矿井中多机器人搜救系统路径规划

Path Planning for Multi-robot Rescue System under Coal Mine

矿难发生后,井下通信设施可能已有一定的损害、无法正常使用,因而无法知道被困人员的确切位置并且井下环境复杂危险,可能对营救人员造成伤害发生二次事故。为了快速地搜索到被困人员,结合井下无线传播环境的特点,提出井下多机器人组网搜救系统,其中包括机器人自由空间环境模型的创建、机器人搜索的局部和全局路径的规划。利用MAKLINK图论理论对井下环境进行建模,并且采用Dijkstra算法规划出避障初始路径,最后依据距离改进蚁群算法节点选择得出优化的最终路径,完成搜索路径的规划。MATLAB仿真结果表明,优化的路径总长度明显优于初始路径并且改进的蚁群算法有着较好的收敛速度,可以满足多机器人搜救系统的要求。

Because communication facilities from underground are damaged and may not work normally after mine disaster,accu-rate position about trapped workers cannot be known.The dangerous environment may cause second accident and hurt rescue workers. To find out trapped workers quickly,the multi-robot rescue system was proposed combined with the characteristics of under coal mine wireless environment,including creation of a free space model for robot,global and local path plan for multi-robot rescue system.The MAKLINK graph theory was used to establish the free space model of the mobile robot,and then Dijkstra algorithm was utilized to find a sub-optimal collision-free path.Lastly,the improved algorithm based on distance to optimize the location of sub-optimal path was a-dopted to generate the optimal path of the mobile robot.The result of MATLAB simulation experiment shows that the proposed algorithm has a better performance in convergence speed and matches to the multi-robot rescue system.