Real-time Motion Planning For Autonomous Car in Multiple Situations Under Simulated Urban Environment
MetadataShow full item record
Publisher:The Ohio State University
Series/Report no.:The Ohio State University. Department of Electrical and Computer Engineering Honors Theses; 2020
Advanced autonomous cars have revolutionary meaning for the automobile industry. While more and more companies have already started to build their own autonomous cars, no one has yet brought a practical autonomous car into the market. One key problem of their cars is lacking a reliable active real-time motion planning system for the urban environment. A real-time motion planning system makes cars can safely and stably drive under the urban environment. The final goal for this project is to design and implement a reliable real-time motion planning system to reduce accident rates in autonomous cars instead of human drivers. The real-time motion planning system includes lane-keeping, obstacle avoidance, moving car avoidance, adaptive cruise control, and accident avoidance function. In the research, EGO vehicles will be built and equipped with an image processing unit, a LIDAR, and two ultrasonic sensors to detect the environment. These environment data make it possible to implement a full control program in the real-time motion planning system. The control program will be implemented and tested in a scaled-down EGO vehicle with a scaled-down urban environment. The project has been divided into three phases: build EGO vehicles, implement the control program of the real-time motion planning system, and improve the control program by testing under the scale-down urban environment. In the first phase, each EGO vehicle will be built by an EGO vehicle chassis kit, a Raspberry Pi, a LIDAR, two ultrasonic sensors, a battery, and a power board. In the second phase, the control program of the real-time motion planning system will be implemented under the lane-keeping program in Raspberry Pi. Python is the programming language that will be used to implement the program. Lane-keeping, obstacle avoidance, moving car avoidance, adaptive cruise control functions will be built in this control program. In the last phase, testing and improvement works will be finished. Reliability tests will be designed and fulfilled. The more data grab from tests, the more stability of the real-time motion planning system can be implemented. Finally, one reliable motion planning system will be built, which will be used in normal scale EGO vehicles to reduce accident rates significantly under the urban environment.
Academic Major: Electrical and Computer Engineering