Development of Optimization-based Path Planning Method with Virtual Environment
Seongyeon Kim, Jongho Shin, Sung-Hoon Back · Journal of Institute of Control Robotics and Systems · 2023
During autonomous driving, unmanned vehicles encounter various environmental changes. Because such environmental changes degrade the overall autonomous driving performance, an optimal path that considers changes in the surrounding environment is required. In this paper, we propose an optimization-based path planning technique that comprehensively reflects changing traversable areas, vehicle dynamic models, static obstacles, and state/input constraints. To this end, the optimization path planning problem is formulated by considering all the aforementioned constraints and target point following. In addition, the pseudo code for obtaining a numerical optimal solution is derived using the first-order gradient descent method. On the other hand, development of autonomous driving system is time-consuming and expensive. To handle the difficulty, this study constructs a virtual environment autonomous driving system that integrates a robot operating system (ROS) and CarMaker. To verify the performance of the proposed path planning method, autonomous driving simulation is conducted in the developed virtual environment and the results are analyzed.