Path Planning Algorithm for Dynamic Obstacles Based on Support Vector Machines Towards Autonomous Navigation

Autonomous car navigation algorithm is proposed to achieve a safe and smooth path to the target destination by integrating dynamic obstacles avoidance and path tracking algorithms. Multi obstacle avoidance algorithm integrates both support vector machines and Gaussian process regression. When obstacles are detected, Gaussian regression predicts the position of the obstacles for the forthcoming time steps, while the SVM classifier finds the optimal separating hyperplane, which guarantees maximum margin to obstacles. The generated path to target goal is smooth, and safe while travels all the way through numerous static and dynamic obstacles. Path tracking control algorithm is designed to ensure minimum tracking error, this algorithm is based on a nonlinear model predictive controller considering vehicle dynamics. The simulation results reveal the effectiveness of proposed algorithms in different complicated scenarios.