基于连续动态运动基元的移动机器人路径规划

针对二维动态场景下的移动机器人路径规划问题,提出了一种新颖的路径规划方法——连续动态运动基元(continuous dynamic movement primitives, CDMPs).该方法将传统的单一动态运动基元推广到连续动态运动基元,通过对演示运动轨迹的学习,获得各运动基元的权重序列,利用相位变量的更新,实现对未知动态目标的追踪.该方法克服了移动机器人对环境模型的依赖,解决了动态场景下追踪运动目标和躲避动态障碍物的路径规划问题.最后通过一系列仿真实验,验证了算法的可行性.仿真实验结果表明,对于动态场景下移动机器人路径规划问题, CDMPs算法比传统的DMPs方法在连续性能和规划效率上具有更好的表现.

Path Planning for Mobile Robots Based on Continuous Dynamic Movement Primitives

Aiming at the problem of mobile robot path planning in two-dimensional dynamic scenes, we propose a novel path-planning method, called continuous dynamic movement primitives(CDMPs). The method is an extension of the traditional dynamic movement primitives. By learning the motion trajectory of the demonstration, we obtain the weight sequence of each movement primitive. We can track unknown dynamic targets by updating the phase variables. The method overcomes the dependence of mobile robots on the environment model and solves the path-planning problem of tracking moving targets and avoiding dynamic obstacles in dynamic scenes. Finally, through a series of simulation experiments, the feasibility of the algorithm is verified. The simulation results show that the CDMPs algorithm has better continuous performance and planning efficiency than the traditional DMPs method for mobile robot path planning in dynamic scenes.