关键词: 单元分解, 融合改进RRT算法, 清扫机器人, MCD局部覆盖算法, 全覆盖路径规划

Abstract: Aiming to address the challenges associated with high algorithm operation cost, poor obstacle avoidance performance, and low area coverage in full-coverage path planning for cleaning robots in large-scale unstructured environments in steel rolling workshops, this paper proposes a full-coverage path planning algorithm. The algorithm utilizes the cell decomposition method for partitioned coverage and fuses it with the RRT area transfer technique and the jump-point search method. Initially, the movement cell decomposition algorithm is employed to achieve coverage of the free regions. Subsequently, to solve the obstacle avoidance issue during inter-region path planning, the RRT algorithm with the fused jump-point search strategy is introduced. This strategy involves searching the target region by biasing the node extension’s directivity towards the target region, followed by correction of the path through the removal of redundant points using both the greedy algorithm and the triple B-spline method for smoothing. The feasibility and effectiveness of the algorithm in different environments are verified through simulation and experiments, and the planned paths are shorter and the path repetition rate is greatly reduced compared with other methods, which improves the efficiency of the robot’s obstacle avoidance and achieves the full-area path coverage at the same time.

Key words: cell decomposition, fusion of improved RRT algorithm, cleaning robot, movement cell decomposition (MCD) local coverage algorithm, complete coverage path planning