基于粒子群并行优化的煤矿井下机器人路径规划

Coal Mine Underground Robot Path Planning Based on Parallel Particle Swarm Optimization[HS)]

摘要:

路径规划是煤矿井下搜救探测机器人自主导航的关键步骤,矿井是三维的非机构化的环境,机器人行走过程应该具有高度智能的路径规划,传统的自适应能力与处理非线性的问题能力较差,路径规划误差较大,提出基于粒子群并行优化的煤矿井下机器人路径规划方法,充分考虑井下的环境高低变化,采用栅格法对环境建模,将粒子群独立分布在不同容器中分别进行路径建模,不同容器中粒子分别进行优化操作;因为速度和最优子群被分别保留,在机器人路径规划实验阶段,路径规划的时间较传统方法降低20%,避障成功率高达95%,最优路径的出现概率能保持在99%,这种方法具有很强的指导性与实用价值。

Abstract:

Path planning is mine rescue detection of autonomous navigation key steps, mine is three dimensional non institutional environment, the robot walk process should be highly intelligent path planning, the traditional adaptive ability and processing nonlinear problem ability is bad, the error is bigger, path planning of coal mine underground robot based on parallel particle swarm optimization is proposed. Fully consider down hole environment height change, the grid method is adopted to environment modeling, particle swarm independent distribution in different containers are path modeling, different containers particle separately optimization operation, because of the speed and the most optimal group were retained in robot path planning, the experimental stage, path planning time reduced by 20% than the traditional methods, obstacle avoidance success rate reaches as high as 95%, the optimal path appear probability can maintain at 99%, this method has a strong guidance and practical value.