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

摘要： 针对钢筋轧制车间中大型非结构化环境下清扫机器人在全覆盖路径规划时所面临的算法运行成本高、避障性能差和区域覆盖率低等问题，提出了一种利用单元分解法分区覆盖和融合跳点搜索法的RRT区域转移的全覆盖路径规划算法。利用MCD（movement cell decomposition）算法实现自由区域覆盖，为了解决区域间路径规划时的避障问题，引入融合跳点搜索策略的RRT算法，通过增加节点扩展的导向性，使其更偏向目标区域进行搜索，并利用贪婪算法裁剪冗余点修正路径以及三次B样条曲线法平滑处理。通过仿真与实验验证了算法在不同环境下的可行性和有效性，相比于其他方法所规划的路径更短且大大降低了路径重复率，提高了机器人避障效率的同时实现了全区域路径覆盖。

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