改进的蚁群算法在灭火机器人多火源路径规划的应用

Application of Improved Ant Colony Algorithm in Multi-Fire Path Planning of Fire Fighting Robot

摘要: 为研究消防灭火机器人在避障环境下寻求到达多火源的最优路径规划问题，针对蚁群算法进行路径规划时易陷入局部最优、收敛速度慢等缺陷，提出一种改进的蚁群算法。首先利用栅格地图建立机器人工作环境模型，将综合权值优先规划策略引入构建好的旅行商算法中，求解出灭火的次序；进而改进转移概率算法，求解出到达各火源的具体路径，增强算法的全局搜索能力及加快收敛速度。在追求路径最短的同时考虑到机器人的转向会消耗时间，提出多指标评价函数来评价路径质量。最后进行仿真，结果表明本算法跳出局部最优能力和收敛速度有很大改进，并证明了改进算法应用在灭火机器人多火源路径规划问题上有很强的可行性和有效性。

Abstract: In order to study the problem of optimal path planning for the fire fighting robot to reach multiple fire sources in obstacle avoidance environment, the based ant colony algorithm is prone to fall into local optimality and slow convergence speed when path planning. First, the grid map is used to establish the robot working environment model, and the integrated weight priority planning strategy is introduced into the constructed traveling salesman algorithm to solve the order of fire extinguishment; then the transition probability algorithm is improved to solve the specific path to each fire source. Meanwhile, the global search ability is improved and the convergence speed is accelerated. While pursuing the shortest path and considering the robot's turning will consume time, a multi-index evaluation function is proposed to evaluate the path quality. Finally, the simulation shows that the algorithm has greatly improved the local optimal ability and the convergence speed, and proved that the improved algorithm has strong feasibility and effectiveness in the problem of multi-fire path planning for fire-fighting robots.