An Improved Bidirectional RRT* Algorithm for UAV Path Planning Based on Energy Consumption Constraints

Aiming at the problems of ignoring turning cost and slow convergence in traditional path planning, a potential function sampling heuristic optimal path planning (AGB-RRT*) algorithm based on Gaussian distribution is proposed. In this paper, a path energy consumption estimation model combining distance and Angle is proposed. Secondly, on the traditional RRT, this paper adopts Gaussian sampling to achieve the heuristic sampling target, introduces the gravitational potential function of the artificial potential field method, and determines the growth direction of the random number according to the size of the configured gravitational factor, thus speeding up the convergence speed. Finally, integrated with the dynamic step size optimization strategy, it guides the growth of new nodes and the update of parent nodes, removes high-cost nodes and sampling points, and uses segmentation cutting and path interpolation to make the UAV plan a shorter path to shoot the target object. Simulation results show that the proposed algorithm can significantly reduce flight energy consumption and turn times, and accelerate initial path acquisition and convergence speed, which is suitable for UAV path planning.