Abstract:

Abstract This paper presents the enhanced operation of the path planner integrated with a predictive controller for a self-driving vehicle to accomplish trajectory planning and avoid obstacles. The path planner used the Batch Informed Trees (BIT*) planning algorithm approach and the tracking controller is designed based on the model predictive control (MPC). BIT* algorithm is used to find the best path between the start and the goal nodes. Then the MPC tracks the route and controls the vehicle's movement to its destination. Path planning control is vital point in avoiding autonomous car the obstacles during serious traffic scenarios. The MPC controls the main parameters of the vehicle: velocity, acceleration, and orientation. The traditional BIT* operation is enhanced by subjecting the generated trajectory to a basis spline (B-Spline) planner. This enhancement solves the hard angle and manoeuvre presented in the path, improves the trajectory points connections, and then swiftly obtains a collision-free trajectory. In addition, this paper tackles the issues related to avoiding local obstacles and the follow up of dynamic goal points in a complex and dynamic world. The model predictive controller is used to track the enhanced trajectory plan generated by the BIT*planner approach by using the kinematic model of the vehicle. A modal description of the approach for building the graph-search for these cases and displaying simulated and real-world examining data shows this method's practical application. In the simulation, the controller selects the best trajectories as references. Also, it enhances the performance of trajectory planning and ensures that the casual obstacle can be avoided in real-time and the robot can arrive at the final point smoothly. The results of the simulation show a reasonable accomplishment in navigation performance, the planned path is softer, and the efficiency of the search is higher in composite environments and different scenarios. Also, the test shows that the autonomous car can pursue the reference path accurately, even with sharp corners.

Keywords: