针对蛇形机器人链式CPG(central pattern generator)复杂模型,提出了采用粒子群算法(particle swarm optimization)优化CPG控制模型参数,用来解决CPG参数整定效率低、效果不理想的问题,然后结合Webots机器人仿真软件,实现了PSO算法优化蛇形机器人CPG控制模型参数。最后利用Webots仿真软件,通过实验仿真与遗传算法的结果相对比,证实了PSO算法的优越性,为蛇形机器人的运动步态转换提供好的基础。
The particle swarm optimization algorithm is introduced to optimize the chain CPG model parameters of snake-like robot in order to deal with unsatisfactory, inefficient parameter setting method of CPG network, and then the PSO to optimize parameters of CPG control model is achieved in the Webots simulation software. As result, it is confirmed that the PSO algorithms is more advantage than the genetic algorithm by experiments in the Webots software, which provides a good basis for the snake robot gait conversion.