LEE- Mekatronik Mühendisliği Lisansüstü Programı

Bu topluluk için Kalıcı Uri

http://hdl.handle.net/11527/19385

Gözat

A control-theoretic approach for vision based quality aware autonomous navigation and mapping toward drone landing

(Graduate School, 2023-12-15) Sözer, Onuralp ; Kumbasar, Tufan ; 518172009 ; Mechatronics Engineering

This thesis presents a novel autonomous navigation approach that is capable of increasing map exploration and accuracy while minimizing the distance traveled for autonomous drone landings. For terrain mapping, a probabilistic sparse elevation map is proposed to represent measurement accuracy and enable the increasing of map quality by continuously applying new measurements with Bayes inference. For exploration, the Quality-Aware Best View (QABV) planner is proposed for autonomous navigation with a dual focus: map exploration and quality. Generated paths allow for visiting viewpoints that provide new measurements for exploring the proposed map and increasing its quality. To reduce the distance traveled, we handle the path-cost information in the framework of control theory to dynamically adjust the path cost of visiting a viewpoint. The proposed methods handle the QABV planner as a system to be controlled and regulate the information contribution of the generated paths. As a result, the path cost is increased to reduce the distance traveled or decreased to escape from a low-information area and avoid getting stuck. The usefulness of the proposed mapping and exploration approach is evaluated in detailed simulation studies including a real-world scenario for a packet delivery drone.
A novel gripper design based on series elastic actuator for object recognition and manipulation

(Graduate School, 2023-03-03) Kaya, Ozan ; Ertuğrul, Şeniz ; 518162009 ; Mechatronics Engineering

Because of Industry 4.0 and its following releases, robotic applications are becoming more significant. The goal of using robots is to automate industrial processes and increase production yield. However, there are still study topics that need to be explored for other problems, such as safety and cooperation. Furthermore, sensor technologies are another important subject for automation. In general, sensors like encoders, cameras, lidar, and proximity are chosen for the control algorithm's feedback sensors. Many times, when only one sensor is used, sensor technologies are insufficient to identify or describe incidental obstacles. Due to this, two or more sensors may be required for continuity and safety. Alternatively, it is proposed that a gripper design with external effect sensitivity may be useful in both reducing the number of sensors and inherently sensing the external effects. For this purpose, a novel gripper mechanism design based on SEA is achieved for object recognition and manipulation. For a low-cost solution, one actuator with a ball-screw mechanism as a linear actuator is used for the fingers' positions. As it is based on SEA, the spring is placed between the linear actuator and the fingers. With this method, the finger can be actuated by one motor. However, they can be rotated independently by external effects. To estimate the external force, the length of the spring is computed by using absolute encoders. As a result of these, the proposed gripper mechanism is sensitive to external effects and can be used for estimating force without any force/torque sensor or tactile sensor. For object recognition, the proposed gripper interacts with the objects placed at the workspace. However, this is not enough to recognize an object. Hence, a DNN model is needed to interpret the interaction between the gripper and an object. Therefore, a DNN model is created in order to achieve recognition by using the points on the defined objects' surfaces. For the training part of DNN, a synthetic data set is generated via CloudCompare. As a result of different hyperparameters' effects on the DNN model, the best model is achieved for the recognition of 11 objects. The experiments are conducted in MEAM laboratory with the gripper mounted on the Staubli Rx160 robot arm. It is proposed for object manipulation that the gripper has the ability to compensate for position faults caused by controller error, an inaccurate model, and so on. The proposed gripper can successfully perform common industrial tasks such as peg-in-hole and surfaces following in collaborative applications. To prove this approach, the experiments are conducted with a haptic device and the gripper mounted on the Staubli Rx160 robot arm and used untrained operators. The results are compared according to control strategies. For this purpose, the user operates the tasks in cases of no guidance and a rigid gripper mechanism, guidance and a rigid gripper mechanism, and a series elastic gripper mechanism with guidance.
Addressing parametric uncertainties in autonomous cargo ship heading control

(Graduate School, 2023-07-13) Jambak, Ahmet Irham ; Bayezit, İsmail ; 518211022 ; Mechatronics Engineering

The progress of computer technology has had a profound influence on the maritime sector, particularly in relation to ship autonomy. The incorporation of computer-assisted systems for guidance, navigation, and control has emerged as a promising approach to minimize the potential risks associated with human errors during ship maneuvering. These errors have the potential to lead to disastrous consequences such as collisions with obstacles, vessel damage, and endangering the safety of passengers and cargo. Several notable studies have explored ship control systems using different control approaches and conducting simulation-based and real-time implementation tests. These studies include adaptive path following control for accurate target tracking, stabilization of heading angle using cascade control simulations and fuzzy logic-based decision-making mechanisms, integration of sensor information to reduce accidents on common ship routes, collision scenario analysis with fuzzy control-based decision-making aligned with international maritime regulations, design and analysis of planar autonomous position control using two independent propellers, rudder, speed, and position control experiments on a catamaran prototype using a PID approach enhanced by a Kalman Filter, and modeling and real-time control of unmanned surface vehicles with accurate parameter identification. These studies have provided valuable insights and advancements in ship control systems. Extensive research has also been carried out to address the control of ships with uncertain hydrodynamic coefficients, resulting in the exploration of various approaches. One approach involves using a Multi-level Model Predictive Control (MPC) method to regulate ship speed and calculate propulsion energy while considering uncertainties. Another approach is adaptive control, which continuously adjusts control inputs based on real-time measurements of ship behavior. Fuzzy logic-based control is also utilized, incorporating fuzzy rules that account for uncertainty in hydrodynamic coefficients to adapt the ship's control inputs. Machine learning techniques, like reinforcement learning, are employed to adjust control inputs based on available data. Additionally, robust control techniques are developed to ensure stable and predictable ship behavior even when faced with uncertain hydrodynamic coefficients. These studies make valuable contributions to ship motion control by addressing the challenges associated with uncertain hydrodynamic coefficients and proposing effective control strategies. This thesis aims to address two significant aspects in the field of ship control. First, the development of a comprehensive ship motion model for the ship's heading angle. Second, the effective handling of parametric uncertainty in ship hydrodynamics. The research focuses on enhancing the understanding of ship behavior and designing control strategies that can robustly handle uncertainties, ultimately improving the performance, safety, and efficiency of ship operations. The first objective of this thesis is to construct a control-oriented ship motion model that accurately represents the ship's heading angle dynamics. The model incorporates both linear and nonlinear dynamics to account for the complexities of ship motion under various operating conditions. To derive the linear model, certain assumptions are made to simplify the typical six degrees of freedom (6DOF) equation of motion into a three degrees of freedom (3DOF) equation. From a motion perspective, slow-speed displacement vessels exhibit minimal heave, roll, and pitch motions. Therefore, the variables $Z$, $K$, $M$, and their derivatives, along with the angular velocities $w$, $p$, $q$, can be neglected. From a geometric perspective, conventional ships are symmetrical on the $xz$-plane. As a result, it is assumed that the $y$-coordinate of the ship's center of gravity is zero $y_G = 0$. For the nonlinear model, a well-established 3DOF ship model provided by the Manevering Modelling Group (MMG) is utilized. This model extends the previously derived control-oriented ship equation of motion by incorporating the dynamics of the ship's propeller. By integrating the propeller dynamics into the model, a more comprehensive understanding of the ship's behavior and the interaction between its motion and the propeller can be achieved. The derivations and explanations in this subsection aim to provide valuable insights into the nonlinear ship equation of motion with propeller dynamics, serving as a useful tool for ship motion control design and optimization. By establishing a reliable ship motion model, it becomes possible to design effective control strategies for heading angle control. This aspect of the research involves reviewing and analyzing existing studies on ship autopilot design, including control methods such as adaptive path following controllers and cascade control simulations. In order to validate the model's accuracy, a turning circle test was conducted. This validation process done by comparing the model's predictions with experimental data from previous studies. To ensure a reliable validation process, a 1/80 scale model of the DTC Container ship was chosen for physical validation. The study's results convincingly demonstrate the model's ability to accurately approximate the trajectory with a fixed rudder angle of $\delta = 35$ degrees. Both the linear and nonlinear models developed in this study exhibit a high level of agreement with the actual system, comparable to the results obtained from other studies utilizing CFD and system-based approaches. The second objective of this thesis is to address the challenge of parametric uncertainty in ship hydrodynamics, particularly focusing on the uncertainty associated with hydrodynamic coefficients. Hydrodynamic coefficients play a crucial role in understanding the interactions between the ship and the surrounding water. However, these coefficients often exhibit variations due to various factors, such as environmental conditions, vessel modifications, or manufacturing tolerances. The presence of such uncertainty can significantly impact the performance and reliability of ship control systems. To address this issue, optimization techniques such as Particle Swarm Optimization (PSO) and Genetic Algorithms (GA) have been used to determine the optimal coefficients that best match the ship's motion. However, these methods can be slow, particularly if a large number of samples are needed to represent the uncertainties in the hydrodynamic coefficients. To answer this challenge, this thesis designed a fast optimization algorithm to find a robust optimal controller for ship heading angle control with a certain level of error in hydrodynamic coefficients that is difficult to solve deterministically. The proposed optimization algorithm is referred to as "Bisection Optimization via Blind Search". This methodology involves simulating the model for a total of $n$ samples, utilizing $m$ randomly sampled controller parameters $K$. The resulting cost value $\omega$ is recorded for each simulation. Subsequently, the controller parameter $K$ that produces the lowest $\omega$ value is selected. In the second stage, a bisection-based optimization technique is employed to further refine the chosen controller parameter $K$. In this thesis, the uncertainty levels considered in this study are set at $\pm$20$\%$, representing the permissible deviations of parameters from their nominal values. To mitigate the adverse effects of uncertainty, a simulation scenario is created where the controller is tasked with adjusting the yaw angle $\psi$ from 0 degrees to 40 degrees. To evaluate the optimization performance, four different cost functions are employed: Integral Absolute Error (IAE), Integral Squared Error (ISE), Integral Time Absolute Error (ITAE), and Integral Time Squared Error (ITSE). However, the results indicate that the optimal controller parameters obtained using these different cost functions do not exhibit significant differences. Two types of controllers, namely proportional (P) and proportional-derivative (PD) controllers, are investigated in this study. The inclusion of the derivative term in the PD controller aims to mitigate oscillations and overshoot in the system's response, thereby improving stability and settling time. However, the integral term is excluded from consideration due to the inherent stability of the system and its expected slow response. The results demonstrate that the proposed algorithm outperforms existing methods such as Particle Swarm Optimization (PSO) in terms of both execution speed and task performance. By achieving these 2 objectives, this thesis contributes to the field of ship control by providing a comprehensive understanding of ship motion dynamics and offering effective solutions for addressing parametric uncertainty in ship hydrodynamics. The outcomes of this research can significantly impact various maritime applications, including autonomous navigation, collision avoidance, and energy-efficient ship operations. The developed ship motion model and the proposed optimization approaches can be implemented in real-world scenarios, enabling improved control strategies for ship heading angle control and enhancing the overall performance and safety of ship operations. In conclusion, this thesis combines the development of a control-oriented ship motion model with the effective handling of parametric uncertainty in ship hydrodynamics. The research contributes to the advancement of ship control by providing valuable insights, methodologies, and tools that can be utilized in designing robust control strategies for ship heading angle control under uncertain operating conditions. The findings of this thesis have the potential to enhance the efficiency, safety, and sustainability of ship operations, ultimately benefiting the maritime industry as a whole.
Applications of deep reinforcement learning for advanced driving assistance systems

(Graduate School, 2023-07-23) Yavaş, Muharrem Uğur ; Kumbasar, Tufan ; 518162005 ; Mechatronics Engineering

Nowadays, advanced driving support systems are becoming more prevalent every day. For instance, although adaptive cruise control has been present in some mass-produced vehicles since 1980, it is now available in almost every new vehicle model and is becoming usable, especially in congested traffic situations, with the help of developing technology. On the other hand, the autonomous lane centering function developed for highway environments reduces the driving load on drivers. One of the main reasons for the advancement and prevalence of technology is the progress in environmental perception sensors. Decision-making algorithms can obtain high-accuracy positions of lanes and other vehicles' speed and positions on the road by blending data from intelligent camera and radar sensors. Thanks to advancements in artificial intelligence research, the main topic of this thesis is to evaluate the conditions of surrounding vehicles to achieve cruise follow speed, the amount of gas or brake applied, and finally, the lane changing decision by deep reinforcement learning. Deep reinforcement learning is the integration of reinforcement learning theory into new generation artificial neural networks that emerged with the deep learning revolution. In the proposed methods, both the adaptive cruise control and autonomous lane-changing functions designed with deep reinforcement learning have taken more optimal decisions than classical algorithms and the similarity between the decisions taken and those taken by human drivers has been revealed. Adaptive cruise control systems typically calculate the amount of acceleration required to maintain a safe following distance by using information about the distance to the closest vehicle. However, this method is not compatible with human driving behavior, as it involves scanning the entire traffic and taking into account the dynamic elements surrounding the vehicle being driven. In one of our proposed solutions, we designed the adaptive cruise control function using a model-based deep reinforcement learning method. In model-based reinforcement learning, the decision-making policy uses its own internal model during training to minimize interaction with the system. Therefore, one artificial neural network creates the decision-making policy, while a second network creates the internal model. By using the proposed meta-learning approach to train the two neural networks in a closed-loop fashion, we selected two leader vehicle data inputs for the algorithm instead of a single one. In our simulation environment, the model-based artificial intelligence algorithm performed better than the classical intelligent driver model. Additionally, we proposed a hybrid method that switches to the classical driver model if the internal model and real-world observations do not match for a certain period of time, with a fallback mechanism added to the system's internal model. xxiii In the second proposed study on adaptive cruise control, we suggested a discrete driver model inspired by human drivers' use of gas and brake pedals to manipulate them directly. In the analysis performed using data collected from real life, it was observed that drivers were driving at a stable state with certain gas and brake pedals and coped with dynamic conditions by applying delta brake or pedal. Different gas and brake delta levels were determined through statistical inference based on this dataset. In this case, as the inputs of the artificial intelligence algorithm, the position and speeds of all vehicles in a multi-lane highway in front of the vehicle were determined. When considering the superiority of the algorithms that work with a single leader vehicle compared to two leader vehicles on a single lane, the information of the vehicles on the adjacent lanes will help in case of changes in the leading vehicle of the ego vehicle. The deep Q-learning algorithm, which provides the best results in discrete outputs, was used as the decision-making algorithm. In the evaluations performed on both simulation and real test data, the proposed algorithm obtained the highest score. Especially, slowing down the vehicle in line with its own friction by giving a 0 output without pressing both gas and brake pedals, which can be evaluated as tactical decision-making, was frequently preferred by the designed algorithm. The other advanced driver assistance system studied in the thesis work is the autonomous lane-changing function. In the first original study, autonomous lane-changing was designed using deep reinforcement learning method, and the normally long training process was accelerated 5 times with the proposed safety reward feedback. In the autonomous lane-changing problem, the critical task is to process the position and speed information from all vehicles in front and behind in traffic and make safe maneuvers that will cause speed increase at the right time. Especially in complex traffic scenarios created in simulated environments, classical algorithms are adversely affected by sensor uncertainties and noises, and they cannot show optimal performance in the dynamic driving of multiple vehicles. With the uncertainty calculation in the designed deep reinforcement learning algorithm, the confidence level of the decisions made is observed, and progress is made in the important research area of explainable artificial intelligence. It seems that although deep reinforcement learning techniques have achieved significant successes, they still face integration issues in real-world applications. One of the main problems is the lengthy training process, which can take millions of steps, and the fact that policies are optimized through trial and error, making training in real systems impossible. One promising area of research is sim2real transfer, which involves transferring policies trained in simulation directly to real-world applications. In the second original study on autonomous lane changing, a new approach was introduced to measure the transferability between two simulators with different resolutions. The transferability was evaluated using a human-like usage score generated from the traffic situations when lane-changing decisions were made. In the training process, an adjusted reward function was used, and the proposed method outperformed reference methods in terms of both efficiency and safety, achieving the highest human-like lane-changing score.
Applications of multi-agent systems in transportation

(Graduate School, 2023-03-21) Tunç, İlhan ; Söylemez, Mehmet Turan ; 518152012 ; Mechatronics Engineering

Traffic density is a growing drawback of the crowding of cities in contemporary societies. As a consequence of financial and technological innovations, the living standards of people are improving yet this increases the number of cars and traffic density accordingly. Thus, the density of traffic is reducing the quality of life for individuals in metropolitan areas in particular. Traffic is an important factor affecting human life quality in crowded cities. The increasing population and increasing individual vehicle ownership lead to an increase in traffic density. This causes an increase in loss of time and pollution. Traffic density in big cities is an important factor that reduces the quality of human life. Due to the growing population in metropolitan areas and the inadequate infrastructure to accommodate this density, traffic problems are on the rise. As a result, passengers waste more time in traffic, and the amount of emissions, and hence air pollution, also increases. The issue of traffic congestion is a significant concern for numerous metropolitan areas across the globe, as it causes delays, increases commuting time, and contributes to air pollution. Controlling the flow of traffic is problematic in terms of many complexities and uncertainties. Despite this situation, this problem needs to be solved as it reduces productivity and living standards. Modern traffic control methods offer a more effective solution, unlike traditional methods. As traffic congestion continues to increase rapidly in the world, the need to research and apply more effective methods of traffic control than the traditional method is increasing. Solving traffic congestion is one of the most important and complex problems, as it causes chaos in metropolitans, especially during heavy traffic hours. Traditional methods that continue to be used have proven to be inadequate, and as a result, the developing technology has affected all areas as well as the solutions to the traffic control problem. With the emergence of Intelligent Transportation Systems (ITS), utilizing artificial intelligence and communication technologies, a more effective and efficient solution to traffic congestion is possible. Transportation techniques are improving day by day with the pace of growing technology. Intelligent Transportation Systems (ITS) provide advanced services such as high-tech traffic controllers and various transportation modes, reducing the burden on drivers and thus enabling them to meet the need for complex decision-making while on the road. Intelligent transportation solutions have enabled an unprecedented level of data collection within the industry, leading to significant advancements in transportation system management and operation. With the increasing demand and rate of data collection, ITS has also been advancing day by day and increasing the speed of progress of smart transportation systems. ITS can be described as systems consisting of technologies such as electronics, data processing and wireless networks that provide security and efficiency in the transportation network. ITS provides communication and information exchange between each transport unit. These units can be centres that provide information to pedestrians, vehicles, infrastructure, transportation and other peripherals such as traffic lights and other communication and control units. The application of MAS (Multi-Agent Systems) techniques, as a new development in information technology, can help to increase interest in traffic and promote energy-efficient transportation. ITS-based multi-agent technology is an important approach to solving complex traffic problems. The complexity of the elements of the traffic makes them convenient for multi-agent structures. ITS-based multi-agent technology provides us with safer controllers and makes us feel more comfortable in our daily lives. It increases the quality of our lives by decreasing the amount of time spent in traffic and by lowering the amount of emission gases released by our vehicles. The structurally dispersed nature of components in heterogeneous environments causes application difficulties, such as interoperability between agents forming a demand for a unified software platform as an underlying infrastructure. Therefore, it is preferable to use centralized solutions for relatively simple problems such as the one considered in this paper. For both transport decision-makers and drivers, ITS have a great potential for efficient and intelligent traffic management, threat identification, driving comfort and safety. ITS can also provide a flexible approach for the effective management of complex networked transportation systems letting traffic management decision-makers to control signal changes, regulate route flows, and broadcast real-time traffic information. In addition to providing route scheduling, weather forecasting, and emergency services for drivers, ITS (Intelligent Transportation Systems) can also help to reduce driving loads and improve safety. The implementation of ITS (Intelligent Transportation Systems) can generate positive outcomes across a range of areas, spanning from environmental and national security issues to emergency management and transportation. ITS applications can reduce time spent on the road. Short travel times provide economic savings for both individual and commercial vehicles and usually mean less environmental pollution. Intelligent Intersection Management (IIM) technology has started to develop in traffic intersections as part of Traffic Light Control (TLC) systems. Intersections are some of the busiest parts of roads. Therefore, the control of traffic lights plays an important role in decreasing the density. In this thesis, particular attention is given to the control of intersections in order to find solutions to decrease traffic density leading to an increased quality of life in big cities. Intelligent traffic control methods, the use of which is increasing with the development of new methods, are used especially in traffic intersections with high traffic density in order to provide efficient solutions. Control of a single intersection with traffic lights is considered first in the thesis. Various methods, including Fuzzy Logic Control (FLC), Proportional Integral (PI) control and State Space Model Control techniques, have been proposed and compared for a better traffic light controller architecture so as to increase the traffic flow and reduce the overall waiting time of the cars and the emissions released by them. It is demonstrated that the proposed architectures give better results compared to the traditional fixed-time traffic light control method. Different types of traffic intersections are considered in the study. Initially, a simple single-lane traffic intersection with no left or right turn is taken into consideration. Later on, intersections on which three-lane (or four-lane) roads meet with vehicles turning left and right are considered. Finally, a realistic case study, in which the Altunizade Region of Istanbul, is examined to demonstrate the efficiency of some of the proposed methods. The results of simulations indicate that the FLC, in which the positions of the vehicles are used as the state variables, gives superior results in comparison to the other classical methods. In order to increase the efficiency of the FLC further, a built-in learning algorithm is proposed to be used in addition to the FLC. A deep Q-learning method is employed for this purpose as a part of the agent-based traffic light controller. Hence, the resulting intelligent traffic light controller is named DQ-FLSI. In this method, a state matrix which divides the arms of the traffic intersection into cells is used. The traffic light durations are determined using fuzzy logic, and traffic light actions are determined by the help of deep Q-learning. A stability analysis is also carried out for this newly proposed method. Another important traffic problem is route planning. This is particularly important in large cities with complex traffic networks. In order to address this problem, an agent-based traffic route planning method has also been proposed as part of this thesis with the motivation of vehicles choosing the fastest route. In this method, route planning is made by deciding at traffic intersection points. Vehicle agents make decisions when they reach traffic intersections. In this way, dynamic route planning becomes possible for the vehicles. Another solution for the traffic intersection problem is multi-agent reservation-based traffic intersection control. With this method, all vehicles (called agents) can pass the intersection without the need for a traffic light thanks to a traffic intersection agent. A platoon method, which can work in harmony with reservation-based traffic intersection management, is proposed as an improvement in this part of the study. The proposed method aims to reduce the slowdowns that occur when approaching the traffic intersection by properly lining up the vehicles approaching the traffic intersection. It is shown by simulations that the proposed platoon method reduces energy consumption and gas emissions while increasing the average speed of the vehicles, especially as the density of the traffic increases. Work environments for all studied traffic problems are designed and simulated using the SUMO program. Simulation of Urban MObility (SUMO) is an open-source simulation package that works on networks imported from maps, provides various workspaces at micro levels, also allows pedestrian simulation, and has a sufficient set of tools that makes it more reachable.
Automatic landing with model predictive control

(Graduate School, 2022) Ulukır, Talha ; Üstoğlu, İlker ; 518191029 ; Mechatronic Engineering Programme

In flight control systems, the landing maneuver is one of the most critical time periods for the aircraft, and it is of great importance to both respond to the disruptors and ensure durability in this time period when the disruptor activity is high. Within the scope of this thesis, 4 different controller type automatic landing systems were designed for a twin-engine passenger aircraft, this landing system provides fully automatic landing in both longitudinal and lateral planes. Within the scope of the thesis, different control architectures in the literature for the automatic landing autopilot were examined. Within the scope of the thesis, the change of system inputs and system outputs as a result of linearization under different conditions has been examined. The consistency of the nonlinear model of the aircraft with the linear model was compared, and this comparison was made by examining the behavior of the system variables in response to the binary commands given to the control surfaces. Within the scope of the thesis, what the sub-phases of the automatic landing autopilot are and according to which criteria and conditions these sub-phases are separated from each other are examined. The classical control architectures in the flight control system (stability-enhancing system and control-enhancing system) are discussed, and for what purposes and with what standards these architectures are designed. In fixed-wing aircraft control systems, the longitudinal and lateral states of the system are separable. In the scope of the thesis, the automatic landing architectures in the literature for these separated states are examined. The controllers designed for the descent system in the thesis are: PID, Linear Quadratic Integral, Model Predictive Control and Algebraic Model Predictive Control architectures. One of these four different control architectures (PID) is in a single-input-single-output control structure, while the other three control architectures (LQI, MPC, AMPC) are in a multi-input-multi-output control structure.
Battery management system design with embedded electrochemical impedance spectroscopy

(Graduate School, 2023-04-27) Babacan, Medet Kerem ; Erol, Osman Kaan ; 518201019 ; Mechatronics Engineering

Humanity is developing day by day in engineering and technical fields. Engineers and scientists all over the world are trying to take humanity one step further. As a result of these studies, the technology provides comfort in our daily lives. One of the best examples of this are transportation, mobility and automotive. Each year, the studies progressing cumulatively in this field have pioneered the presentation of more developed cars than the previous years, and to populate different transportation methods and concepts in our lives. Each development has been formed as a result of some needs. One of the most important motivations in the field of automotive is user requests, the limited resources and the environmental sensitivity. The transition to electric vehicles, one of the biggest revolutions in transportation, has undoubtedly accelerated due to the depletion of fuel resources and the environmental sensitivity. Fuels used to operate internal combustion vehicles are obtained from petrol. While the formation of a petrol reserve takes for thousands of years, the current reserves are running out. Considering the demand that will rise in transportation due to the population and industrialization increase over time, it is predicted that the petrol reserves will be drained in the coming years. In addition, as a result of reactions inside the internal combustion engine, harmful gases are produced and released into world atmosphere. These gases, which are released from millions of vehicles, accumulate in the atmosphere and disrupt the balance of nature. Therefore, humanity has now become unable to lean on to petrol fuels and have been in search of new fuel sources. In this context, the most innovative transportation methods seem to be hydrogen and electrical based. While the comparison of these two methods with each other is the subject of a separate study, this study will focus on the electrical transportation method and the batteries to be used in these means of transportation. Transportation has been mainly provided by internal combustion vehicles until today. This naturally allowed many engineers working on this field to accumulate a lot of knowledge cumulatively. Over the years, engineers have solved the problems in the designs one by one and have reached the present knowledge level by pushing the limits of the existing technology. There have been many developments in internal combustion engines and vehicles in areas such as safety, efficiency, practicality and comfort. However, electric vehicles that have just started to become widespread have opened a new page. Compared to internal combustion propulsion systems, studies on electric propulsion systems are still in their infancy. Engineers and scientists are conducting a lot of work to fill the gap in this field. One of the most basic components of electric vehicles are batteries. When we compare one to one with internal combustion vehicle, the battery group corresponds to the fuel tank of the vehicle. The first factors for the user, such as the range of the electric vehicle, the charge time and the performance at different temperatures, are completely related to the battery. When these factors are examined, they are all disadvantaged xx compared to internal combustion vehicles. This offers a negative effect for the market share of electric vehicles. While the criterias mentioned are the factors that the user will experience directly, there is also a factor that the user cannot experience, but in fact, which is even more important than all of them, which is safety. As can be seen from time to time, electric vehicles may caught fire while charging or in a traffic accident. It is not possible to extinguish it when a battery flames. Therefore, this is a great danger for both the vehicle's user and for those around. These situations show that there are many more things to develop in terms of both safety and user experience in the batteries of electric vehicles. Today, Li-ion type cells are widely used in the batteries of electric vehicles. These cells are preferred because they are one of the cell types that give the highest energy per kilogram. As a result of chemical reactions in these cells, electrical energy is generated, which provides power to traction. Battery cells have safe operating ranges. In particular, the voltage and temperature values of the cells should be within some ranges. Otherwise, chemical reactions in the cells come to an uncontrollable point and undesirable fires, explosions or structural deformations may occur. In addition, since these cells are non linear systems, it is not easy to predict the changes in their internal structures as a result of their use. For this reason, it is a research area in itself to predict the energy remaining in the battery of an electric vehicle and therefore the range it can go to. In order to overcome such difficulties, there is a control unit that manages the battery and this module is called the battery management system. In this study, a battery management system will be developed to ensure the safety of the electric vehicle battery and has a new method to estimate the chemical structure of the batteries. The developed battery management system will measure voltage, current and temperature values in the battery and check whether the battery is at safe ranges. It will take the necessary actions to prevent these values get dangerous. In addition, the battery will drive auxiliary elements in battery pack such as contactors. It will send the measurements and calculations taken over the battery over the communication channels and work in harmony with the other components in the vehicle. The designed battery management system will be scalable and the big battery packets will be able to managed with different number of battery management system modules. There are many methods to analyze the chemical structures of batteries. The most important of these is electrochemical impedance spectroscopy. In this method, an alternative current is sent to a battery cell in the laboratory environment and the voltage change in the cell terminals is monitored. This voltage change analyzed in the frequency domain and an idea is obtained about the chemical internal structure of the cell. In spite of obtaining valuable information as a result of this method, the devices that do this analysis are expensive, heavy and stationary devices that can be used only in the laboratory environment. Within the scope of this study, it is aimed to integrate such an analysis method into designed battery management system. Thus, the tests performed in the laboratory will be able to performed on the vehicle also, and the accuracy will increase for battery management system calculations. Within the scope of this study, the hardware of a battery management system will be developed. After circuit schematic and printing circuit board design completed, circuit board will be produced and prototyping work will be done. Low level drivers, battery management system algorithms and electrochemical impedance analysis algorithms will be developed on this board. The developed product will be tested on a battery pack and measurements will be taken. Taken measurements will be used to express cell structure as an equivalent circuit model. Application areas that product can be used and future studies will be discussed.
Classification of ten different motor imagery eeg signals by using deep neural networks

(Graduate School, 2023-08-19) Korhan, Nuri ; Dokur, Zümray ; 518152011 ; Mechatronics Engineering

Brain-Computer Interface (BCI) is a research area that aims at establishing a sustainable communication infrastructure between the brain and machines. The primary purpose of BCI is to restore functionality to paralyzed individuals, but it can also be used for gaming applications. Various modalities such as Electroencephalogram (EEG) and Functional Magnetic Resonance Imaging (fMRI) can be employed in this field. This thesis focuses on EEG-based BCI and specifically explores the classification of ten different motor imagery (MI) tasks using deep neural networks. Motor imagery is a BCI method that aims to detect imagined movements through potential changes on the scalp, which are measured by electrodes during the imagined motor movement. Increasing the number of recognizable tasks in BCI systems, specifically in the field of mechatronics, holds considerable importance. The limited scope of a four-command system significantly inhibits the versatility of these applications, particularly as they become more complex. To illustrate, imagine the operational demands of a drone, which requires absolute control over direction, altitude, speed, and elaborate maneuvers to navigate obstacles in three-dimensional space. The limitations of a four-command system decrease the number of controllable actions, thus undermining the efficacy and the scope of BCI applications. A substantial increase in the number of recognizable tasks in a BCI system signifies not only the expansion of its capabilities, but also a progression in advancing its applicability and versatility. In the first chapter, the problems of BCI are introduced, and the relevant literature is reviewed. In the second chapter, the concepts related to MI, the specific BCI area of interest, are explained. The third chapter examines methods to increase the number of commands in the MI paradigm, discussing previous approaches and the proposed methods. In the fourth chapter, deep learning tools commonly used in the field and employed in this research are introduced and discussed. The fifth and final chapter discusses the obtained results, their implications, and potential future research directions. The findings contribute to the advancement of BCI and demonstrate the feasibility of classifying ten different motor imagery EEG signals using deep neural networks, alongside augmentation, and divergence-based feature extraction. In summarizing the research conducted in this study, emphasis must be placed on the success rates achieved through the application of the developed methods. The techniques of artificial EEG signal generation, data augmentation, and regularization have been utilized, resulting in enhancements in the performance and efficiency of the BCI tasks. The methods employed have demonstrated promising results in various test scenarios. The success rates exceed those observed in traditional approaches documented in the literature. These rates are expanded upon in their respective sections and numerically illustrated in tables within the fifth chapter. Looking at the classification of both simple and combined MI-EEG signals across various studies, mean accuracy rates of around 51.6% and 54.2% were reported using different techniques for feature extraction and classification on three simple and one combined MI-EEG signals across a varied number of subjects. When increasing the number of classes used, as in four simple and four combined MI-EEG signals, a trend of increased mean accuracy was observed. Studies reported accuracy rates of 55% (four simple and four combined classes, dataset 3) and a substantial 70% (four simple and three combined classes) using different methods. The methods developed in this study demonstrate a significant improvement. For dataset 1, the proposed approach achieved an 85% mean accuracy with only DivFE on four simple and six combined classes across three subjects. Dataset 2 shows a 78.6% accuracy across nine subjects. Lastly, for the dataset 3 (four simple and four combined), the model achieved a 77.8% accuracy across seven subjects. These success rates not only validate the effectiveness of the proposed methods but also highlight the potential for future enhancements in BCI applications.
Comparative analysis of torque vectoring control strategies in electric vehicles

(Graduate School, 2024-06-11) Sezgin, Emre ; Yumuk, Erhan ; 518211013 ; Mechatronics Engineering

Vehicle dynamics control systems have an important role in accident prevention by decreasing the difference between the desired and actual vehicle response. Torque Vectoring Control is one of these systems which is developed to enhance the steering and handling performances of vehicles. This thesis focuses on the comparison of the control strategies for the potential of improving the steering and handling performances of electric vehicles through torque vectoring control systems. In this thesis, a non-linear multi-body dynamics model is used to represent the real electric vehicle. This vehicle model adopts three independently controllable electric motors; one of them is on the front axle for traction and the other two are on the rear axle. With the help of these two motors on the rear axle, the yaw moment can be controlled with the torque difference between the two electric motors. Furthermore, the control system needs a target value to follow the desired behavior of the system. To achieve this aim, the lateral dynamics reference generation has been included. When working with the vehicle lateral dynamics in the study, the yaw rate response is considered as a representative of steering and handling performance. Then, the system steering and handling capabilities using different maneuvers are evaluated with closed-loop test cases performed in a simulation environment. Various controllers can be used to assess the steering and handling performance on the vehicle. In the first part of this thesis study, the effects of PID controller parameters on system performance for a single maneuver are examined, and then these parameters are adjusted to minimize the integral square error performance metric. The durability of this optimal controller under maneuver changes is also investigated. Simulation results show that under different maneuvers, particularly at high speeds and steering angles, the system performance significantly deteriorates. To overcome this problem, possible maneuvers the vehicle might encounter are determined, and then optimal PID controller parameters are found for each maneuver to minimize the integral square error performance metric. Using the optimal PID parameters found for possible maneuvers, an adaptive PID controller design is proposed with the help of the cubic spline method. The performance of the adaptive PID controller adapted to speed and steering angle is tested under various maneuver changes, yielding satisfactory results. In the second part of the thesis study, the effects of a fuzzy PID controller on system performance are examined due to its success in controlling nonlinear systems. For this purpose, fuzzy PID parameters, i.e. input-output scaling factors are found using the rule table and membership functions from the literature to minimize the integral square error, similar to the optimal PID controller. The fuzzy PID controller is compared to the optimal PID controller using a similar maneuver. Although the fuzzy PID controller does not produce superior results for a single maneuver compared to the optimal PID controller, it shows quite satisfactory results when faced with maneuver changes. The reason for the fuzzy PID controller not producing superior results for similar maneuvers is that the rule table and membership functions from the literature are not suitable for the system. For this purpose, the output membership functions are adjusted (shifted outward and inward), and then a fuzzy PID design is designed for a similar maneuver. When the output membership functions are shifted outward, the fuzzy PID controller produces superior results compared to the optimal PID controller.
Comparison of MPC based motion control algorithms on mujoco hybrid platform

(Thesis (M.Sc.) -- Istanbul Technical University, Graduate School, 2025, 2025-06-17) Kızıldemir Kılıç, Ayşe ; Temeltaş, Hakan ; 518181005 ; Mechatronics Engineering

Dört ayaklı robotlar, son yıllarda özellikle zorlu ve düzensiz arazilerdeki yüksek hareket kabiliyetleri sayesinde robotik alanında giderek artan bir ilgiyle karşılaşmakta ve öncü araştırma konularından biri haline gelmektedir. Bu tür sistemler, yapıları gereği yalnızca düz yüzeylerde değil, aynı zamanda kaygan, eğimli, engebeli veya çukur gibi karmaşık arazi koşullarında da dengeyi koruyarak ilerleyebilme kapasitesine sahiptir. Geleneksel tekerlekli ya da paletli hareket sistemleriyle karşılaştırıldığında, dört ayaklı robotlar çok daha yüksek çeviklik, esneklik ve adaptasyon yetenekleri sunmakta; bu da onları arama-kurtarma operasyonları, keşif görevleri, askeri uygulamalar, tarımsal otomasyon sistemleri ve afet sonrası incelemeler gibi çevresel belirsizliğin yüksek olduğu görevler için ideal hale getirmektedir. Ayrıca bu sistemler, biyolojik canlıların yürüyüş ve denge stratejilerinden esinlenerek tasarlandıkları için, karmaşık çoklu temas senaryolarını yüksek verimlilikle yönetebilme potansiyeline sahiptir. Ancak bu sistemlerin gerçek dünya uygulamalarında güvenilir ve verimli biçimde çalışabilmesi, yalnızca gelişmiş mekanik yapı, dayanıklı gövde tasarımı ve hassas sensörlerle değil; aynı zamanda çevresel değişimlere hızlı adapte olabilen, hesaplama açısından etkin ve yüksek kararlılığa sahip kontrol algoritmalarıyla mümkündür. Dört ayaklı robotların yürüyüş planlaması, dengesizlik durumunda toparlanması veya farklı zemin türlerine anlık tepki verebilmesi gibi işlevleri, klasik kontrol yöntemlerinin sınırlarını zorlamakta; bu nedenle ileri düzey hareket kontrol stratejilerine ihtiyaç duyulmaktadır. Bu bağlamda Model Öngörülü Kontrol (Model Predictive Control - MPC), sistemin mevcut durumunu ve dinamik modelini kullanarak gelecekteki olası davranışları belirli bir zaman ufku içinde tahmin edebilmesi ve kontrol girişlerini bu doğrultuda optimize etmesi sayesinde, dört ayaklı robot kontrolünde dikkat çeken bir yöntem olarak öne çıkmaktadır. MPC, her kontrol adımında bir optimizasyon problemi çözüp, sistemin gelecek durumları üzerinde etkili olacak kontrol girişlerini belirli kısıtlar altında seçer ve yalnızca ilk kontrol adımını uygular. Bu işlem her adımda tekrarlanarak sistemin gerçek zamanlı olarak çevre koşullarına duyarlı, planlı ve güvenli şekilde hareket etmesi sağlanır. MPC'nin bu ileri görüşlü yapısı, özellikle çok eklemli, temasa duyarlı, doğrusal olmayan ve yüksek derecede dinamik davranış sergileyen dört ayaklı robot sistemlerinde; denge koruma, adım geçişi, yön değiştirme ve düşmeden toparlanma gibi görevlerin güvenilir biçimde gerçekleştirilmesine olanak tanır. Bu tez çalışmasında, MPC temelli üç farklı kontrol algoritması İteratif Lineer Kuadratik Gauss (iLQG), Gradyan İnişi (Gradient Descent) ve Öngörülü Örnekleme (Predictive Sampling) karşılaştırmalı olarak incelenmiştir. Söz konusu algoritmalar, Google DeepMind tarafından geliştirilen açık kaynaklı MJPC (Model Predictive Control for MuJoCo) platformu üzerinden, MuJoCo fizik motoru ile entegre bir şekilde Unitree A1 dört ayaklı robot modeli üzerinde uygulanmıştır. MuJoCo (Multi-Joint dynamics with Contact), sürekli temas içeren çok eklemli sistemlerin yüksek hassasiyetle fiziksel simülasyonunu mümkün kılan bir simülasyon altyapısıdır. MJPC ise bu fizik motorunu kullanarak gerçek zamanlı MPC planlayıcılarının uygulanmasına olanak tanır. Tez kapsamında oluşturulan simülasyon ortamı Python diliyle kodlanmış ve Jupyter Notebook platformunda yapılandırılmıştır. Python'un sunduğu esneklik, güçlü veri işleme kütüphaneleri (NumPy, Pandas) ve görselleştirme olanakları (Matplotlib) sayesinde algoritmalar arası karşılaştırmalar sistematik ve tekrarlanabilir biçimde gerçekleştirilmiştir. Simülasyonlarda kullanılan robot modeli olan Unitree A1, 12 serbestlik derecesine (3'er DOF'lu 4 bacak), yüksek torklu fırçasız motorlara ve IMU gibi temel sensör donanımlarına sahiptir. Robotun MJCF (MuJoCo XML) formatında fiziksel özellikleri tanımlanmış ve MJPC çerçevesine entegre edilmiştir. Her algoritma için aynı koşullarda yürütülen simülasyonlarda, farklı yürüyüş senaryoları (yürüme, dörtnala koşma), eğimli yüzeylerde hareket, bozulmuş zeminde ilerleme ve düşme sonrası toparlanma gibi görevler gerçekleştirilmiştir. Her kontrol döngüsünde 2 ms kontrol zaman adımı ve 30 adımlık öngörü ufku kullanılmış; böylece toplam 60 ms ileriye dönük planlama gerçekleştirilmiştir. Simülasyonlar sırasında sistem durumu (pozisyon, hız, eklem torkları, zemin tepkisi vb.) sürekli izlenmiş ve maliyet fonksiyonları her zaman adımında hesaplanarak performans ölçütleri elde edilmiştir. İteratif Lineer Kuadratik Gauss (iLQG) algoritması, sistemin doğrusal ve kuadratik yaklaşımlarına dayanan bir ikinci mertebe optimizasyon yöntemidir. Her iterasyonda sistem dinamiği ve maliyet fonksiyonu lokal olarak doğrusal ve kuadratik hale getirilir; böylece hem ileri besleme (feedforward) hem de geri besleme (feedback) kontrol sinyalleri elde edilir. Bu çift yönlü yaklaşım sayesinde iLQG, yüksek izleme doğruluğu, hızlı yakınsama ve sistemsel kararlılık sağlar. Simülasyon sonuçlarında, iLQG algoritmasının hem yürüme hem de dörtnala senaryolarında en düşük maliyet fonksiyon değerlerine ulaştığı, sistemin düşmeden önce toparlanma kabiliyetinin yüksek olduğu ve temas değişimlerine karşı istikrarlı tepkiler verdiği görülmüştür. Bununla birlikte, ikinci türev bilgisine olan bağımlılığı ve yüksek hesaplama yükü, özellikle donanım sınırlamalarının olduğu uygulamalarda dezavantaj oluşturabilir. Gradyan İnişi yöntemi ise birinci mertebe optimizasyon tekniğidir ve yalnızca maliyet fonksiyonunun gradyan bilgisine ihtiyaç duyar. Bu nedenle iLQG'ye kıyasla daha düşük hesaplama yüküne sahiptir ve uygulaması daha kolaydır. Ancak algoritmanın başarısı, adım büyüklüğü gibi hiperparametrelerin hassas ayarlanmasına bağlıdır. Simülasyonlarda, Gradyan İnişi yönteminin ortalama düzeyde bir performans sergilediği, bazı senaryolarda dengeyi sürdüremediği ve özellikle ani dış etkiler karşısında hassas olduğu gözlemlenmiştir. Geri besleme mekanizması bulunmadığından sistemsel bozulmalar telafi edilememekte, bu da kararlılığı azaltmaktadır. Buna rağmen, sınırlı kaynaklı sistemlerde veya prototip kontrol algoritmalarının geliştirilmesi aşamasında faydalı bir yöntemdir. Öngörülü Örnekleme (Predictive Sampling) algoritması ise, belirli bir dağılımdan rastgele kontrol dizileri örnekleyerek, bu diziler için simülasyonlar yürütür ve en düşük maliyet üreten kontrol dizisini uygular. Bu yöntemin avantajı, türev bilgisine ihtiyaç duymaması ve paralel işlemeye uygun olmasıdır. Ancak örnek sayısının artırılması performansı olumlu etkilerken, aynı zamanda hesaplama maliyetini de yükseltir. Tez kapsamında yapılan deneylerde, Predictive Sampling algoritmasının özellikle basit yürüme senaryolarında kararlı hareketler sergilediği ancak karmaşık ve hızlı adaptasyon gerektiren durumlarda yeterli tepki veremediği gözlemlenmiştir. Ayrıca, geri besleme mekanizmasının olmaması nedeniyle, beklenmedik bozucular karşısında denge kaybı yaşanmıştır. Tüm kontrol algoritmaları, performans karşılaştırmasının adil ve tutarlı bir biçimde yapılabilmesi amacıyla aynı maliyet fonksiyonu tanımı altında test edilmiştir. Bu maliyet fonksiyonu, robotun istenen (referans) yörüngeden sapmalarını ve bu yörüngeyi takip ederken harcadığı kontrol eforunu birlikte değerlendiren bir yapıdadır. Özellikle konum ve hız gibi durum değişkenlerindeki sapmalar ile eklem torklarının büyüklüğü, belirli ağırlık katsayılarıyla dengelemiş ve çok amaçlı bir optimizasyon çerçevesi oluşturulmuştur. Böylece hem sistemin izleme doğruluğu hem de enerji verimliliği göz önünde bulundurulmuş, algoritmaların yalnızca performansı değil aynı zamanda uygulama uygunluğu da dikkate alınmıştır. Simülasyonların sonunda elde edilen çıktılar, zaman serisi verileri üzerinden detaylı şekilde analiz edilmiştir. Bu analizler sırasında robotun üç boyutlu uzaydaki gövde pozisyonu, bacak uçlarının (end-effector) lineer ve açısal hızları, yer ile temas anındaki kuvvet dağılımları ve her bir eklemde üretilen tork değerleri dikkatle incelenmiştir. Bu veriler, Python kullanılarak geliştirilen görselleştirme araçları aracılığıyla grafiksel olarak sunulmuş; algoritmaların performansları, hem sayısal hem de görsel analizlerle desteklenmiştir. Özellikle adım geçiş süreleri, denge merkezinin (Center of Mass - CoM) salınım büyüklüğü, ayak temas süresi ve yönelme (orientation) açılarındaki değişimler, algoritmalar arası farkları ortaya koymak açısından kritik metrikler olarak kullanılmıştır. Ayrıca, gait geçişlerinin performansı, simülasyonlar sırasında robotun adım düzenini koruyabilme yetisi ve geçiş sırasındaki salınımların kontrol edilebilirliği açısından değerlendirilmiştir. Denge koruma analizi kapsamında ise robotun dışsal bozuculara veya eğimli yüzeylere verdiği tepkiler incelenmiş, her bir algoritmanın bozulmaya ne hızda yanıt verdiği, dengenin ne kadar sürede yeniden sağlandığı gibi metrikler karşılaştırılmıştır. Yönelme analizinde ise robotun sapma yaptığı eksen etrafındaki toparlanma davranışı, açısal hız regülasyonu ve gövde stabilizasyonu gibi yönler ele alınmıştır. Bu kapsamlı değerlendirmelere ek olarak, MJPC platformunun sınırlı olan analiz ve senaryo çeşitlendirme kapasitesi, Python diliyle geliştirilen özel modüller sayesinde genişletilmiştir. Kullanıcının manuel olarak tanımlaması gereken senaryolar, otomatikleştirilmiş yapılandırma dosyaları ile betik destekli hale getirilmiş; farklı yürüyüş modları ve zemin profilleri için seri simülasyonlar çalıştırılabilir duruma getirilmiştir. Ayrıca her simülasyon sonunda elde edilen veriler düzenli biçimde kaydedilmekte ve video formatında dışa aktarılmaktadır. Bu süreçlerin otomatize edilmesi sayesinde yalnızca algoritma doğruluğu değil, aynı zamanda tekrarlanabilirlik, simülasyon süresi ve analiz verimliliği gibi pratik uygulama ölçütleri de değerlendirilebilir olmuştur. Sonuç olarak, bu tez çalışması, farklı MPC algoritmalarının dört ayaklı robot hareket kontrolünde sağladığı başarıyı sistematik biçimde karşılaştırmış; algoritmaların hem teorik yapıları hem de uygulama çıktıları yönünden güçlü ve zayıf yönlerini ortaya koymuştur. iLQG algoritması genel performans bakımından en başarılı yöntem olarak öne çıkarken, Gradyan İnişi ve Örnekleme tabanlı algoritmalar belirli görevler ve kaynak kısıtları altında uygulanabilir alternatifler sunmuştur. Çalışmada geliştirilen simülasyon ve değerlendirme altyapısı, gelecekte pekiştirmeli öğrenme (reinforcement learning) tabanlı yöntemlerin de dahil edileceği genişletilmiş araştırmalar için sağlam bir temel teşkil etmektedir. Bu sayede hem akademik literatüre hem de gerçek dünya uygulamalarına katkı sağlanması hedeflenmektedir.
Cooperative control of multi-agent system under time delay

(Graduate School, 2023-09-07) Akkaya, Şirin ; Ergenç, Ali Fuat ; 518142009 ; Mechatronic Engineering

In this Ph.D. dissertation, multi-agent systems are studied in detail using two of the most common examples in practice, which are vehicle platooning systems and formation control of unmanned aerial vehicles. For a better understanding of the study, some basic information such as graph theory, matrix theory, and time-delayed systems are given. Then, the "Cluster Treatment of Characteristic Roots" paradigm, which forms the backbone of the study, is explained, and the existing methods in the literature have been explained. In this study, a new Bezout Resultant matrix-based CTCR method has been proposed, and the steps of the algorithm are explained via simulation examples in detail. The main advantage of the proposed method is that it provides computational convenience for the time-delayed systems in which the degree of characteristic equation is relatively large and not decomposed into factors in obtaining the stability posture of the system in terms of time delay. First, the distributed controller algorithm is selected as the state feedback controller. The closed-loop system matrix is constructed for the cases with and without time delay. The controller coefficients that make the system stable are obtained by using the Routh table and Lyapunov-based methods for the case where the time delay is neglected. However, in the presence of delay, the system is converted into retarded time delay system, and the stability posture is obtained with CTCR methods for single and multiple time delays. Morover, the formation geometry between vehicles is considered as constant policy and constant headway policy. For constant policy, the characteristic equation of the system for delayless and single time delay case, is decomposed into factors, which makes the stability analysis easier. But, this case is not possible for the characteristic equation involved multiple time delay, which direct us to utilize Bezout Resultant matrix-based CTCR method. For constant time headway policy, it is seen that, the characteristic equation cannot decomposed into factors for any cases. So, the sufficient condition is derived for determining the stability of multi-agent system for delay-free case with converted the system matrix to block companion form and block Schwarz form. Then, a PID controller based distributed controller protocol is proposed. The cooperative control problem of multi-agent system with distributed PID controller is converted into an asymptotic stability problem through matrix and state transformations in the absense of time delay. Finally, a Lyapunov function is created and the controller parameters are choosen with the help of linear matrix inequality. In the presence of time delay, the closed-loop system is converted into a neutral-type time delay system. And, the stability posture of the multi-agent system is obtained with the help of Kronecker multiplication and elementary transformation based CTCR method. Finally, all the theoretical studies and simulation results are evaluated with a real-time experimental study. An industrial controller-based real-time simulation for the platoon system with five connected vehicle including a virtual leader is proposed. The constant time headway policy is selected to modeled the desired inter-vehicle distance and the vehicle dynamic states-based distributed control strategy is used to converge to their desired velocities and inter-vehicle distances. Then the multi-agent platooning control problem is converted into LTI system stability analysis problem. The delay-based stability analysis is studied by means of Bezout Resultant matrix-based CTCR method. Numerical simulations are provided to verify the validity of the proposed method. The real-time experiments are carried out on industrial computers to show the applicability of the proposed method in real time systems. The study concluded by evaluating the results and recommendations.
Çamaşır makinelerinde kullanılan plastik kasnakların seri üretim otomasyon sistemi tasarım doğrulama analizi

(Lisansüstü Eğitim Enstitüsü, 2024-02-02) Yıldız, Hilal ; Yeşiloğlu, Murat ; 518181018 ; Mekatronik Mühendisliği

Çamaşır makinelerinde tamburu döndürmek için tahrik elemanı olarak kullanılan kasnaklar, makinenin en önemli parçalarıdır. Kayış yardımıyla motorun miline, göbek bölgesinden ise tamburdaki flanşın miline bağlıdır. Böylece, motorun dönme hareketi kayış yardımıyla önce flanş miline, ardından da tambura iletilir ve bu, tamburun dönmesini sağlayarak çamaşırların yıkanmasını ve sıkılmasını mümkün kılar. Kasnaklar, çamaşır makinesinin verimli ve etkili bir şekilde çalışmasında ve yıkama performansında kritik bir role sahiptir ve genellikle dayanıklılık ve uzun ömür sağlamak için alüminyum ya da yüksek cam elyaflı plastik ham maddeden üretilirler. Arçelik Çamaşır Makinesi bünyesinde yüksek cam elyaflı ham maddeden üretilen alüminyum göbekli plastik kasnaklar kullanılmaktadır. Dayanıklı ve uzun ömürlü olan plastik kasnaklar, makinenin kullanım ömrünü artırır ve enerji maliyetinde önemli faydalar sunar. Çamaşır makinesi üreticileri, kasnakların yapısında, kullanılan malzeme, kol sayısı, kol tasarımı ve çap gibi özelliklerde farklılıklar gösterir. Örneğin, Arçelik önden yüklemeli çamaşır makinelerinde, yüksek cam elyaflı ham maddeden üretilen göbek bölgesi alüminyum olan 5 kollu plastik kasnak kullanılır. Makinenin sıkma hızına bağlı olarak, yüksek devirli makinelerde küçük çapta, düşük devirli makinelerde ise büyük çapta plastik kasnak kullanılır. Plastik kasnaklarda çap ölçüsünün yanı sıra, kullanıldığı üründeki flanşın miline bağlı olarak göbek bölgesindeki alüminyum parça 3 farklı yapıya sahiptir. Küçük ve büyük çaplı plastik kasnaklarda göbek yapısı ortaktır, kasnak çapına bağlı olarak değişmez. Plastik kasnaklardaki herhangi bir kalite problemi direk çamaşır makinesinin yıkama performansını etkiler. Bundan dolayı plastik kasnakların belirlenen tüm kalite kriterlerini sağlaması gerekir. Plastik kasnaklarda çamaşır makinesinin çalışmasını etkileyecek en önemli kalite kriteri plastik kasnakların yalpa ve salgı ölçüleridir. Bu ölçüler motor miline bağlı olan kayışın plastik kasnağa tutunmasını sağlar. Yalpa ve salgı değerlerinin üretimde kullanılan tüm plastik kasnaklarda teknik resim isterlerindeki toleranslar içerisinde olması gerekir.
Çamaşır makinelerinde yapay sinir ağları ile yıkama performansı ve enerji tüketiminin modellenmesi

(Lisansüstü Eğitim Enstitüsü, 2022-06-23) Aktaş, Yakup ; Altınkaynak, Atakan ; Kalafat Acer, Merve ; 518181036 ; Mekatronik Mühendisliği

Dünyada gelişen teknoloji ve mühendislik yetkinlikleriyle beraber endüstride bir rekabet ortamı oluşmuştur. İlgili sektör üreticileri fark yaratan ürünler otaya koymak için gelişen teknolojik adımları yakalamak ve ürünlerine değer katacak gelişmeleri takip etmek durumundadır. Özellikle tüketicinin doğrudan etkileşim halinde olduğu beyaz eşya ürünlerinde fark yaratan teknolojiler ön plana çıkmaktadır. Ancak ürünlere eklenen birçok yeni özellik beraberinde maliyetleri de doğurmaktadır. Aynı zamanda üreticiler için kaynak ve zaman yönetimi anlamında da ekstra yük getirmektedir. Bu nedenle ürünlerin ar-ge, tasarım ve üretim süreçleri ne kadar iyileştirilebilirse sektörde rekabetçi ve yenilikçi ürünler ortaya koyabilmek o kadar mümkün hale gelecektir. Üretilecek olan ürünlerin ar-ge ve tasarım aşamalarındaki test süreçlerinin iyileştirilmesi maliyet, kaynak ve zaman açısından üreticiler için olumlu katkı sağlamaktadır. Çamaşır makinaları günümüzde yaygın olarak kullanılan dayanıklı tüketim aletleridir. Su ve elektrik enerjisi ile çalıştıkları için, test süreçlerinde her bir çevrimdeki bu tüketimler ek maliyetlere ve aynı zamanda dünya kaynaklarının da tüketilmesine yol açmaktadır. Bununla birlikte zaman açısından da yeni ürün proje süreleri uzamakta ve teknolojik gelişimi yavaşlatmaktadır. Yani test süreçlerinin kısalması, hem sürdürülebilirliğe katkı yapacak, hem maliyetleridüşürecek hem de zamanın verimli kullanılmasına yol açacaktır. Tez kapsamında çamaşır makinalarının ar-ge ve tasarım süreçlerinde gerçekleştirilen test metotlarının kurulacak model yapısında incelenmesi ile çevresel sürdürülebilirliğe katkı sağlanması, üretici maliyetlerinin düşürülmesi ve zaman tasarrufu elde edilmesi amaçlanmaktadır. Çamaşır makinalarının sahip olduğu özelliklerin yanında, standartlarca belirlenmiş çeşitli sınırları da sağlıyor olması gerekmektedir. Bunlardan biri yıkama performansıdır. Çamaşır makinalarının temel özelliği olan yıkama işlemi, standartlarda belirlenmiş yöntemlerile ölçülebilmektedir. Üretilen çamaşır makinalarının da belirlenen limit değerin altına düşmeyecek etkinlikte yıkama performansına sahip olması gerekmektedir. Üreticiler ise bu sınır koşulu sağlayıp sağlamadığını test etmek için standart yıkama performansı testlerini laboratuvar oertamında gerçekleştirmektedir. Ancak farklı sınır koşullarından dolayı yıkama performansını sağlayabilmek adına birçok parametrenin optimize edilmesi gerekmektedir. Birden fazla parametrenin etki ettiği yıkama performansı hedef değerini yakalayabilmek adına yapılan bu deneme testleri ise su ve enerji tüketimlerinden dolayı beraberinde ekstra bir yük getirmektedir. Bu sebeple kurulacak model yapısı ile bu test sonuçlarının tahmin edilebilmesi hedeflenmektedir. Diğer bir yandan, standart olarak sağlanması gereken yıkama performansının belirli enerji tüketimi sınırları içerisinde gerçekleşiyor olması gerekmektedir. Üreticiler, üretilen çamaşır makinasının enerji tüketiminin, standartlarda belirlenen enerji sınıf aralıklarından hangisine denk geldiğini deklare etmek durumundadır. Doğal olarak daha düşük tüketime sahip enerji sınıfındaki ürünler son kullanıcı tarafından daha çok tercih edileceğinden yıkama performansı değerine olabilecek en düşük enerji tüketimi ile ulaşmak ana hedeftir. Bu nedenle yapılan performans testleri yerine yıkama performansını tahmin edecek model ihtiyacının yanında, optimum tasarımın yapılabilmesi için enerji tüketiminin de tahmin edilmesi gerekmektedir. Kurulacak enerji tüketimi modeli ile de enerji tüketimi değerinin test yapmadan tahmin edilebilmesi amaçlanmaktadır. Tez kapsamında kurulacak yıkama performansı ve enerji tüketimi tahmin modellerini elde edebilmek için öncelikle deneysel veriye ihtiyaç vardır. Bu amaçla laboratuvar ortamında deney istasyonları hazırlanmış ve standart yıkama performansı test sonuçları tüm analog ve dijital verileriyle birlikte toplanmıştır. Tahmin edilmek istenen yıkama performansı ve enerji tüketimi değerlerinin yanında model yapılarını girdi sağlayabilecek parametrelerin de değişimleri kaydedilmiştir. Toplanan verilerin analizi yapılarak yıkama performansı ve enerji tüketimi tahmin modelleri için ayrı ayrı girdi parametreleri seçilmiş ve çeşitli model yapıları oluşturulmuştur. Oluşturulan yapılardan en iyi performans gösteren modeller seçilmiştir. Elde edilen modeller sayseinde yıkama performansı ve enerji tüketimi için seçilen girdi parametresi değerleri verildiğinde yüksek doğrulukta sonuçlar alınmaktadır. Tezin ilk bölümünde literatürde çamaşır makinalarında gerçekleştirilen yıkama prosesine etki eden temel parametrelerden bahsedilmiştir. Ayrıca tezin ilk bölümünde çamaşır makinlarında geliştirilmiş makine öğrenmesi, yapay sinir ağı ve bulanık mantık algoritma çalışmalarından örnekler sunulmuştur. Yapılan çalışmalarda tahmin edilmesi kritik parametrelere yer verilmiş ve farklı yöntemler kıyaslanmıştır. Tezin ikinci bölümünde yıkama performansı ve enerji tüketimi modellerine veri girişi sağlamak amacıyla kurulan deney sisteminden, kullanılan ekipmanlardan ve ölçüm yöntemlerinden bahsedilmiştir. Bu bölümde ek olarak toplanan deneysel veri kümesi incelenmiştir. Verilerin makina özellikleri açısından yıkama performansı ve enerji tüketimine göre dağılımları gösterilmiştir. Tezin üçüncü bölümüden yıkama performansı modeli için girdi parametreleri seçilmiştir. Girdi parametrelerinin çıktı değerine etkileri detaylıca açıklanmıştır. Parametrelerin istatistiksel özellikleri elde edilmiş, girdi-çıktı parametreleri arasındaki lineer korelasyon ilişkileri çıkarılmıştır. Tezin bu bölümünde lineer yöntemlerin problemi çözümlemeye yetmeyeceği ve makine öğrenmesi yöntemlerinin denenmesi gerektiği yapılan lineer regresyon analizleri ile vurgulanmıştır. Bu amaçla aynı bölümde modelleme için kullanılacak yapay sinir ağları ile Levenberg-Marquardt geri yayılım algoritması açıklanmıştır. Kurulacak modelin algoritma parametreleri detayları ile verildikten sonra farklı katman ve nöron sayılarındaki yapay sinir ağı sonuçları elde edilmiş ve en iyi performansı veren modeller belrtilmiştir. Yapay sinir ağı modeli Matlab programı kullanılarak Levenberg-Marquardt geri yayılım öğrenme algoritmasının model parametre detayları değiştirilerek oluşturulmuştur. Tezin dördüncü bölümünde de yıkama performansı yapay sinir ağı modeline benzer şekilde enerji tüketimi modeli için de girdi parametreleri belirlenip lineer korelasyon ilişkileri belirtilmiştir. Lineer regresyon analizi sonuçları paylaşılmış ve enerji tüketimi modeli için de yapay sinir ağı modeli kurulmuştur. Yıkama performansı yapay sinir ağı modeli ile aynı ağ yapısı özelliklerinde modeller karşılaştırılmış ve en yüksek performansı veren model seçilmiştir Tezin beşinci bölümünde elde edilen model yapıları, ortak model arayüzü oluşturmak adına Simulink ortamına aktarılmış ve tasarım süreçlerinde kullanıma hazır hale getirilmiştir. İlgili girdi parametrelerinin değerleri verildiğinde elde edilen en iyi modellerin tahmini sonucu yıkama performansı ve enerji tüketimi değerleri elde edilebilmektedir. Tezin beşinci ve son bölümünde ise yapılan tez çalışmasının sonucuna ve gelecek çalışmalar için önerilere yer verilmiştir.
Deep reinforcement learning approach in control of Stewart platform- simulation and control

(Graduate School, 2023-06-08) Yadavari,Hadi ; İkizoğlu, Serhat ; Aghaei Tavakol, Vahid ; 518162002 ; Mechatronics Engineering

As named, this work approaches the Stewart platform's controlling task with reinforcement learning methods, presenting a new simulation environment. The Stewart platform, having a broad range of applications that span from flight and driving simulators to structural test platforms, is a fully parallel robot. Exact control of the Stewart platform is challenging and essential in its applications to deliver the desired performance. The fundamental aim of artificial intelligence is to address complex problems by utilizing sensory information with a high number of dimensions. Reinforcement learning (RL) is a specific area of Machine Learning (ML) that incorporates an agent interacting with its surrounding environment according to some policies to maximize the sum of the future rewards as an objective function. The agent's learning process is based on a reward-penalty scheme according to the quality of the selected action from the policy space. In this manner, RL tries to solve many problems and tasks. The primary focus of this work revolves around acquiring the ability to control a sophisticated model of the Stewart platform through the utilization of cutting-edge deep reinforcement algorithms (DRL) and model-based reinforcement learning algorithms. The question is that why do we need a simulation environment? To learn an optimal policy, reinforcement learning necessitates a multitude of interactions with the environment. Experiences with real robots are expensive, time consuming, hard to replicate, and even dangerous. To safely implement the RL algorithms in real-time applications, a reliable simulation environment that considers all the nonlinearities and uncertainties of the agent environment is inevitable. Therefore, an agent could be trained in the simulation through sufficient trials without concerns about the actual hardware issues. After having accurate parameters of the controller learned by the simulation, they can be transferred to a physical real-time system. With the objective of improving the reliability of learning performance and creating a comprehensive test bed that replicates the system's behavior, we introduce a precisely designed simulation environment. For our simulation environment, we opted for the Gazebo simulator, which is an open-source platform utilizing either Open Dynamic Engine (ODE) or Bullet physics. Integrating Gazebo with ROS can pave the way for efficient complex robotic applications due to the ability to simulate different environments involving multi-agent robots. Although some Computer-Aided Design (CAD-based) simulations of the Stewart platform exist, we choose ROS and Gazebo to benefit from the latest reinforcement learning algorithms with high yield and performance, compatible with the last developed RL frameworks. However, despite many robotic simulations in ROS, it lacks parallel applications and closed linkage structures like the Stewart platform. Consequently, our initial step involves creating a parametric representation of the Stewart platform's kinematics within the Gazebo and Robot Operating System (ROS) frameworks. This representation is then seamlessly integrated with a Python class to facilitate the generation of structures.
Design and control of ros based omnidirectional vehicle

(Graduate School, 2022) Nalbant, İbrahim Dinçer ; Temeltaş, Hakan ; 710624 ; Mechatronics Engineering Programme

Nowadays, with the fact that the hardware and software of mobile robots are cheaper and easier to find, studies on them have increased, and as a result of research and development activities carried out by academia and large companies, mobile robots have been widely used in the areas such as exploration, humanoids, drones, automation, transportation, space missions, patrolling, search and rescue and service robots. In addition, a wide variety of driving systems have been designed to increase the driving characteristics of mobile robots, as well as sensors and software architectures to make them autonomous. In this thesis, a mecanum wheeled, ROS-based autonomous, omnidirectional vehicle prototype with high load carrying capacity, called ITU omnidirectional vehicle, has been designed and produced to be used in future research and development activities at the ITU Robotics Laboratory. ITU omnidirectional vehicle is capable of moving to all directions on the ground plane without changing its heading angle, with the help of its specially designed mecanum wheels. Mecanum wheels are driven individually by four Maxon DC motors. For the autonomous navigation, there are two Hokuyo LIDAR sensors and a Xsens IMU sensor mounted in the vehicle. Vehicle inverse and forward kinematic analyses were carried out according to the mecanum wheel kinematics and chassis dimensions, then these equations were applied to the control software of the vehicle. DC motor selection of the vehicle was made according to the payload and acceleration values needed. A special drivetrain was designed between motors and wheels to reduce the lateral force, vibration and balancing effects of the mecanum wheels to the motors. Electrical design of the vehicle was made by using DC regulators and supplying appropriate power for the hardware. For the software of the vehicle, open source Robot Operating System(ROS), was integrated to the vehicle, so that various control, localization, path planning and mapping algorithms could be developed and tested. Both simulation and experimental results of different path following scenarios for the vehicle is presented in the last chapter of this thesis. Results of these case studies satisfied the requirements of the thesis.
Design and evaluation of energy management systems for connected hybrid and electric vehicles

(Graduate School, 2022-07-04) Özdemir, Abdulehad ; Koç, İlker Murat ; 518132001 ; Mechatronics Engineering

Transportation is one of the most significant sources of emissions across various industries. With the effect of Paris Climate Agreement and the Green Deal, environmental concerns and technological progress push the development and market penetration of electric vehicles and hybrid electric vehicles. The number of electric and hybrid vehicles which can be considered as a stepping stone for electric vehicles are increasing day by day. On the other hand, transportation systems are becoming more efficient and safe by the improvement of the communication systems both on the vehicles and the infrastructure. There are significant improvements in connected and autonomous vehicles which has been started with the development of advanced driver assistance systems. The automotive industry, which plays a key role in the development of many accompanying technological ecosystems, is expected to be enhance more changes in the next 10 years than in the past 100 years. It is estimated that this transformation will dominance especially with the technologies progress in connected and autonomous vehicles. The main purpose of the study is to develop smart energy management strategies for connected, hybrid and electric vehicles and evaluate the benefits of developed smart energy management strategies. At the same time, the effects of the transition to electric vehicles in terms of energy consumption and environment is evaluated. For the optimization studies the Well-to-Wheels emission values are calculated and used in order to ensure apple-to-apple comparison. During the thesis study, three articles have been prepared and the preapared articles have been the substructure of the thesis. As of the date of submission of the thesis, one of the articles has been published and the requested revisions have been made for the other two articles and resubmitted. Prepared articles entitled as " Dynamic Programing Based Green Speed Advisory System Design for Mixed Platooning Vehicles", " Driving Cycle Based Energy Management Strategy Development for Range Extended Electric Vehicles " and " Comparative Study on Well-to-Wheels Emissions between Fully Electric and Conventional Automobiles in Istanbul". The article about the comparative study on Well-to-Wheels emissions has been published in the eighty-seventh issue of the "Transportation Research Part D" journal. Turkey's energy mix is analyzed and the emission factor of electricity production of Turkey is estimated in order to make appropriate comparisons during optimization studies. The Well-to-Wheels equivalent carbon dioxide emissions of the electricity is calculated. By considering energy sources, the Well-to-Wheels emission of Turkey is calculated as 520 g carbon dioxide equivalent per kWh. By using the carbon intensity of electricity, it is possible to compare the same variable for electric energy and fossil fuels for hybrid and electric vehicles. Vehicle models are created to use for model-based optimization studies. In order to develop an energy management system for serial hybrid vehicles, all critical subsystems are tested and a vehicle model which is validated by the test data is created. The model is developed by mathematical modelling of vehicle dynamics and testing the the electric motor, motor driver, battery cells and internal combustion engine. The developed models are validated by vehicle level testing on chassis dynamometer. A driving cycle based energy management strategy is developed for range extended electric vehicles to increase system efficiency and equivalent vehicle range. The results showed that; the optimized strategy can save CO2 emission by 6.21%, 1.77% and 0.58% for heavy, moderate and light traffic respectively. The usage of range extender in an efficient way by taking the traffic data into account extends the vehicle range, especially in heavy traffic conditions. For the hybrid vehicles which consumes both electric energy and fossil fuels, It will is important to compare the same value fort he objective function such as equivalent carbon dioxide emission. This study is a good example from this point of view. The developed energy management system will enable connected hybrid vehicles to be in more efficient way by using the route and traffic density information. In addition, vehicle emission maps are developed as a vehicle feature. The vehicles are tested on the chassis dynamometer and emission maps which are based on speed and wheel force are created. It is offered that vehicle emission maps can be used for optimization studies, especially in traffic with different types of vehicles. Considering that there are many ongoing studies on reducing tranportation based emissions, the use of the standardized emission maps are important for system level efficient use of connected vehicles. From this point of view, a multi-layer dynamic programing based optimizer is designed to minimize platooning Well-to-Wheels emissions of platooning vehicles where the platoon consists of an electric, a gasoline and a diesel vehicle. Vehicle emission maps and longitudinal dynamics are used for vehicle modelling. Tank-to-Wheels emission maps of internal combustion engine vehicles are produced by testing the vehicles on a chassis dynamometer. The optimization process has exploration and exploitation layers. The cost function is total Well-to-Wheels emission, design variable is speed trace, constraints are speed limits, traffic light states and vehicle accelerations limits. The test results show that the developed optimizer helps to achieve a 19.8% reduction in total Well-to-Wheels emissions for the defined use case. Thus, there is a significant emission saving potential in using speed advisory system for platooning vehicles through signalized intersections. On the other hand, driving cycles are used to examine the energy consumption and emission emissions of vehicles. In order to analyze the environmental effects of electric vehicles on a real driving cycle, a driving cycle has been developed for Istanbul by statistically analyzing the data collected on the determined routes. By using the developed driving cycle, the vehicle test are conducted. Acoording to the results electric vehicles emit 73.9 g carbon dioxide equivalent per kilometer on the same route, while gasoline vehicles emit 183.4 g equivalent carbon dioxide emissions. Therefore, the transition to electric vehicles should be strengthened by more widespread use of renewable energy in order to effectively reduce emissions associated with electric vehicles in general. At the same time, the results of this study can be a guide for policy makers. In summary, within the scope of the thesis electric carbon intensity of Turkey is calculated by considering Turkey's energy mix and Well-to-Wheels greenhouse gas emissions are analyzed both for conventional and electric vehicles are measured. A dynamic programing based optimizer is developed to decrease total Well-to-Wheels emissions of the mixed conventional and electric platooning vehicles through signalized intersections. Vehicle emission maps are generated both for electric and conventional vehicles for model-based optimization. A driving cycle based energy management strategy is developed for range extended electric vehicles to increase system efficiency and equivalent vehicle range. The vehicle model is developed by critical subsystem testing. An up to date driving cycle for Istanbul is developed (so called Istanbul Driving Cycle) by using collected traffic data across various sections of the city. An internal combustion engine vehicle and an electric vehicle are tested on a chassis dynamometer under the same conditions to determine specific energy consumption and specific emissions.
Developing mobile robot obstacle avoidance methods with model-based and learning-based methods

(Graduate School, 2023-07-19) Özdemir, Aykut ; Bogosyan, Seta O. ; 518162010 ; Mechatronics Engineering

Mobile robot navigation is a crucial area of research and development in robotics that focuses on enabling robots to move autonomously in their environments. Mobile robots are increasingly being used in a wide range of applications, including manufacturing, healthcare, transportation, and search and rescue missions. These robots have the potential to improve efficiency, reduce costs, and enhance safety in a variety of industries. However, for mobile robots to be effective, they must be able to navigate their surroundings with accuracy and reliability. Navigation involves the robot's ability to perceive its environment, plan a path, and execute that path while avoiding obstacles and other hazards. The development of mobile robot navigation systems has been a major area of focus in robotics research for several decades, and it continues to evolve rapidly. Advances in technologies such as sensors, computing, and machine learning have enabled mobile robots to navigate more complex environments and perform increasingly sophisticated tasks. As such, mobile robot navigation is a critical area of study for researchers and engineers who seek to develop intelligent and autonomous systems that can operate in real-world environments. Path planning and obstacle avoidance are two important topics in robotics that are closely related. Path planning refers to the process of determining a safe and efficient path for a robot to travel from its current location to a desired destination. This process takes into account the robot's movement capabilities, the environment it is operating in, and any obstacles that may be present. Obstacle avoidance, on the other hand, involves the robot's ability to detect and avoid obstacles as it navigates its environment. This is an essential component of path planning, as the robot must be able to react to changes in its environment and modify its path accordingly in order to avoid collisions and ensure safety. Both path planning and obstacle avoidance are critical for the development of autonomous robots that can navigate complex environments and perform tasks without human intervention. These topics are the focus of ongoing research in the field of robotics, and advances in technologies such as sensors, mapping algorithms, and machine learning are enabling robots to navigate increasingly complex environments with greater efficiency and safety. This study proposes three novel contributions in the field of robotics. The first is a novel model-based obstacle avoidance method that plans local trajectories by passing through gaps between obstacles. The second is a learning-based sampling method that improves the efficiency of trajectory planning for path planning algorithms. Finally, we proposed a non-holonomic local planner that uses a CNN-based sampling technique. These contributions aim to improve the navigation and path planning capabilities of robots, allowing them to operate more efficiently and safely in complex environments. Overall, this thesis demonstrates the potential of using advanced techniques and technologies, such as machine learning and local planning, to enhance the performance and capabilities of mobile robots.
Development and control of an active torsional vibration damper for vehicle powertrains

(Graduate School, 2021-05-25) Yüceşan, Alişan ; Mugan, Ata ; 518152002 ; Mechatronics Engineering

The emission regulations on internal combustion engines (ICEs) have become more stringent and the importance of fuel efficiency has enhanced due to environmental pollution concerns. As a result, studies on optimization of ICEs put forward the consideration of downsizing, downspeeding and turbo supercharging concepts in designing modern ICEs and powertrains. Despite their numerous advantages, they result in boosted engine torsional vibrations which demands innovative vibration isolation solutions. Such a design solution should be uncomplicated and simple from an automotive manufacturer's point of view, meanwhile, be an extreme performer besides being a cost-effective solution. Passive and active dampers have been utilized to suppress torsional vibrations in the literature. At this point, the passive dampers appear more preferable at the first sight due to their cost advantage while active systems have the disadvantages of having higher costs due to the presence of actuators, sensors and peripherals, advanced complexity and potential of lower efficiency. But conventional passive torsional vibration damper systems like dual-mass flywheel (DMF) reached their limits and they are no longer able to sufficiently isolate the torsional vibrations of the state-of-the-art ICEs. Also, as seen in the literature survey section, there is no kind of active dampers for torsional vibrations which is equipped alone or proposed to be used in this context for vehicle powertrains and there is a huge demand for a new design solution to beat the performance constraints of passive isolation systems.
Development of quality prediction model and control mechanism for clinching process

(Graduate School, 2024-07-04) Kazancı, Emin Abdullah ; Kocaarslan, İlhan ; 518211011 ; Mechatronics Engineering

Many mass production lines in the industry use the joining technique known as clinching. Reasons for the high demand for the clinching process are the unnecessity of additional binding agent, process speed, waterproofness, eco-friendliness, and ease of implemantation. In the clinching process, metal sheets are formed under mechanical force that is applied by punch and die tools. The tools are designed and produced specifically according to the thickness and material properties of metal sheets. Despite the fact that there are electromechanical or hydro-pneumatic powered, conventional hydraulic powered clinching stations are the most preferred as sources of mechanical force because of their investment cost, process speed, versatility and size advantages. However, hydraulic powered systems bring along some drawbacks such as a lack of precision on the quality of clinched joints, eccentricity between punch and die, power consumption and control difficulty because of the single pump that feeds multi-cylinder systems. Although there are three major quality indicators of clinched joints, the bottom thickness of the joint is the most used and critical one because it is both the simplest measurement in an production environment and the most related to quality. Nevertheless, inspection of all produced clinched joints is not feasible based on the measurements of a single operator. Therefore, a quality prediction model is developed in this study. The study is conducted with force and displacement data that is collected from 16 different clinching cylinders at a 1200 Hz sampling rate. Linear, ridge, lasso, decision tree, random forest, extreme gradient boosting, support vector machine and k-nearest neighbors machine learning models are experimented with and validated systematically. The random forest regressor is found to be the best validation scored model. Additionally, a smart decision mechanism (SDM) is developed and implemented based on force and displacement sensor data to overcome major malfunctions that cause a remarkable amount of scrap and production line stoppage. Moreover, a part-to-part feedback control mechanism is developed and implemented to control clinching quality in the optimum range. The bottom thickness of a clinched joint for 0.4 and 0.5 mm stainless metal sheet joining must be between 0.3 mm and 0.4 mm in order to be evaluated as optimum, while the range of 0.25–0.5 mm is accepted as a proper joint. The control mechanism uses force and displacement sensor data to observe system behavior, and utilizes the prediction model and periodic manual measurements to build reference thresholds. In conclusion, an application that stores sensor data, runs control algorithms and makes visualization, is developed for two clinching stations that consist of 16 hydraulic cylinders. In future, the study can be maintained to predict quality more precisely and maintenance dates with regard to the expanding data set and the advanced machine learning algorithms.
Diagnosis of brain cancer and contour normal tissue for radiation therapy based on deep learning methods

(Graduate School, 2024-07-18) Halili, Navid ; Doğan, Mustafa ; 518162006 ; Mechatronics Engineering

Brain tumors are one of the deadliest types of cancer ever identified. Rapid and accurate diagnosis of brain tumors, followed by surgical intervention or appropriate treatment, increases the probability of survival. Accurate identification of brain tumors in MRI scans allows precise location of surgical intervention or chemotherapy. Accurate segmentation of brain tumors in MRI scans is challenging due to their varied shapes and requires knowledge and accurate image interpretation. This thesis starts with analyzing machine learning and traditional methods and focuses on the study of edge detection using the Sobel and Canny edge detector algorithm. After that, we use morphology-based techniques to segment the images and evaluate the results. We use K-means techniques for Clustering. Despite various advances, these methods still show limitations in complex situations such as tumor detection and segmentation. In the next step, we analyze the process of dividing photos into parts using transformations. Specifically, we discuss the Wavelet and Contourlet transforms. By using these transformations, we get more detailed information about the analysis of the images. These transformations have many applications, and we can identify the borders of the image and combine them. Finally, we can use this transformation to process and generate deep learning masks using a supervised model. In the following, we analyze new techniques using supervised and deep learning approaches in two specific areas: image classification and image segmentation. As we introduce these methods, we introduce the obstacles facing deep learning and discuss potential strategies to solve and enhance them. Using deep neural networks and the Resnet 50 model, we classify brain images into tumor and non-tumor categories and achieve a satisfactory score of 97% in the F1 criterion. In addition, we introduce and analyze the Unet deep network in deep learning and upgrade it to a RESUNET network for segmentation. The results of this segmentation show that the proposed approach, with different criteria, such as the DICE metric with a score of 0.9434, performs exceptionally during training compared to conventional topologies and shows a faster convergence rate. In the last part, we presented the unsupervised learning system and developed the adversarial generative network to generate brain MRI images. The adversarial generative network is an intelligent network for generating the desired data, and the results show the effectiveness of the adversarial generative network in generating new data. It is of exceptional quality.

Gözat

Başlık ile LEE- Mekatronik Mühendisliği Lisansüstü Programı'a göz atma

Sayfa başına sonuç

Sıralama Seçenekleri