LEE- Elektrik Mühendisliği Lisansüstü Programı

Bu topluluk için Kalıcı Uri

http://hdl.handle.net/11527/19259

Gözat

Konu "accident rate" ile LEE- Elektrik Mühendisliği Lisansüstü Programı'a göz atma
  • Öge
    Accident data analysis for formal scenario generation and traffic simulation
    (Graduate School, 2023) Kutlu, İlke ; Akıncı, Tahir Çetin ; Akbaş, Mustafa İlhan ; 835232 ; Electrical Engineering Programme
    Autonomous vehicles have become one of the most important topics in the automotive industry today. Vehicles, which have started to be equipped with many electronic control units and sensors in order to increase the comfort and safety of drivers, are preparing for the transition to fully autonomous driving after breakthroughs in processor power and artificial intelligence in recent years. Advanced driver assistance systems, such as cruise control system, collision avoidance system, autonomous emergency braking system or lane tracking system, that support the driver or support systems which are activated only when it is necessary, are also present in vehicles on the road today. Although such systems are used to increase safety and driver comfort, either the driver is still in control of the vehicle or driver supervision is required during these systems and the driver can override these systems at any time. The ultimate goal of recent research and development studies on autonomous driving and autonomous vehicles is to bring self-driving vehicles to a level that they can be in traffic without any need for human attention or intervention, without causing any safety problems. Autonomous driving and advanced driver assistance systems attract the attention of the end consumers because they both are innovative technologies and because of the advantages they will provide for driving comfort. Due to this interest of the consumer, autonomous driving and advanced driver assistance systems have become an area for manufacturers and companies working in the field of research and development in the automotive sector, where they can show their technology. For this reason, the development of autonomous driving and advanced driver assistance systems and making them safer is one of the most important research and development topics for both large vehicle manufacturing companies and supplier companies that supply systems and components to vehicle manufacturers. The result of many different scenarios experienced in traffic in real life can end with major or minor accidents. One of the most important issues in autonomous driving development is to be able to test how the driverless car will behave before and during a possible accident. It is close to impossible to define these traffic accidents with only logical and mathematical equations to create test cases. Moreover, there are variables like speed, departure time, braking time etc. in real-life scenarios even for the similar cases. Accident situations are one of the most critical issues in the use of machine learning for autonomous driving, as they are the scenarios that can endanger human safety the most. A lot of different situations should be tested to validate the artificial intelligence in this regard. However, the traffic accident data for every kind of incident are very difficult to find and collect, as they are rare and real-life data. For example, one of the important data sources on traffic accidents is traffic accident reports.