Improved Bidirectional A* and Time-Elastic Band Algorithm Based Mobile Robot Path Planning

Abstract

This paper proposes an improved A* algorithm combined with the time-elastic band (TEB) algorithm for path planning. Using the angle information of related nodes, the 8 neighborhood search in the the traditional A* algorithm is simplified to 5 neighborhood search, which improves the computing efficiency and reduces the node expansion. Using the distance information of related nodes, a dynamic weight factor is added to the heuristic function to balance the speed and precision of path planning. The bidirectional search strategy is used instead of the traditional unidirectional search to shorten the search time. By considering the robot kinematics and other constraints, the TEB algorithm optimizes the global path to achieve smooth and safe local obstacle avoidance. Numerical simulation and physical experiments show that the proposed algorithm effectively achieves global path planning and local obstacle avoidance in complex environments. Compared with the traditional algorithm, the number of extended nodes decreases by 85.80 % and 71.85 %, and the search time decreases by <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{8 4. 8 9 \%}$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{6 6. 1 0 \%}$</tex> under two algorithms respectively.