动态和未知环境中基于改进粒子群优化的路径规划  

Path planning based on improved particle swarm optimization in dynamic and uncertain environment

作　　者：刘彬[1] 陈特放[2] 张仁津[1] 

机构地区：[1]贵州师范大学数学与计算机科学学院,贵阳550001 [2]中南大学信息科学与工程学院,长沙410083

出　　处：《计算机应用研究》2014年第6期1673-1676,共4页Application Research of Computers

基　　金：贵州省科学技术基金资助项目(黔科合J字LKS[2013]29号)

摘　　要：为了提高动态和未知环境中路径规划的能力,提出了一种改进粒子群优化和评估路径优劣程度的适应度函数。先把起始点作为智能体的当前点,如果当前点到目标点的直线路径会发生碰撞,再由改进粒子群优化根据适应度函数循环搜索不会发生碰撞的下一个移动点。适应度函数包括智能体到粒子之间的等效距离和粒子到目标点的等效距离两个部分。当有新的移动障碍物出现时,从智能体当前点起的已搜索到的路径分段判断是否与新的移动障碍物发生碰撞,从发生碰撞的路径开始由改进粒子群优化重新规划路径。仿真测试和比较结果表明,提出的方法在动态和未知环境中有较强的路径预测能力。To improve the ability of path planning in dynamic and uncertain environment,this paper proposed an improved particle swarm optimization and a fitness function evaluating the degree of the pros and cons.It regarded the start point as the current point.If collisions would occur at the straight-line path between the current point and the target point,the improved particle swarm optimization searched the next moving point circularly.The fitness function consisted of two parts,the first part was the equivalent distance between the agent and the particle and the second part was the equivalent distance between the particles and the target position.When new mobile obstacles appeared,whether obstacle would happen it was needed to be judged sectionally from the current position of the agent.The path was needed to be replan from the collision path by improved particle swarm optimization.Emulational tests and comparative results show that the presented method has strong predictive ability.

关 键 词：粒子群优化 移动障碍物 路径规划 避障 动态环境 未知环境 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]