摘要
针对传统粒子群算法容易陷入局部最优且进化后期速度较慢等缺点,提出一种基于改进粒子群算法的消防机器人路径规划方法。该方法在粒子群的迭代过程中加入动态的惯性权重,并考虑约束条件,约束条件为路径不能经过障碍物,优化目标为整个路径最短。消防机器人工作空间中的障碍物可以是任意的多边形,模拟其在实际环境中可能会遇到的各种情况,并对障碍物进行编号。惯性权重随迭代次数动态改变,可以帮助算法逃出局部极值,提高搜索的精度。仿真结果验证该算法的正确性和有效性。
A new global path planning approach based on particle swarm optimization (PSO) for a fire robot is presented. The particle swarm optimization is used to get a global optimized path. Considering path planning as an optimization problem with constraints, the constraints are the path can not pass by the obstacles and the optimization target is the distance of the path is shortest .The obstacles in the robot's environment are described as polygons and the vertexes of obstacles are numbered from1 to n. Inertia weight dynamically change with the number of iterations,?simulation results prove the effectiveness and practicability of this approach.
出处
《微计算机信息》
2011年第4期207-209,共3页
Control & Automation
基金
陕西省自然科学基金(2009JM8002)
关键词
粒子群算法
路径规划
消防机器人
惯性权重
Particle swarm optimization
Path planning
Fire robot
Inertia weight