Research on Local Obstacle Avoidance of Power Inspection Robot Based on Improved TEB-DWA Algorithm

Aiming at the local path planning problem of power plant environment, this paper proposes a hierarchical trajectory planning framework based on improved TEB-DWA algorithm. Firstly, the explicit smoothness function is introduced into the TEB algorithm to solve the trajectory jitter problem caused by the implicit parameter dependence of the traditional method by directly quantifying the curvature and jerk changes. Secondly, the dynamic weight model of multi-factor nonlinear coupling is constructed to realize the dynamic adjustment of weight. Finally, in order to vehiclery out scenario analysis, a fuzzy evaluation system is constructed, and the rationality of the improvement is verified by setting six rules. Aiming at the obstacle avoidance requirements of dynamic obstacles, the DWA fusion speed obstacle (VO) theory is improved, and the threat evaluation function is constructed. The two work together through a dynamic triggering mechanism: the improved TEB is responsible for generating a time-optimal and smooth trajectory, and the improved DWA is responsible for avoiding short-period dynamic obstacles. Matlab and ROS simulation experiments show that the electric inspection vehicle can avoid obstacles and reach the target point in a smooth path in both static and dynamic obstacle environments, which verifies the feasibility of the improved algorithm in this paper.