基于自适应粒子群遗传算法的柔性关节机器人动力学参数辨识

提出一种基于自适应粒子群遗传算法的柔性关节机器人动力学参数辨识方法。该算法采用动态自适应调整策略,提高了粒子群算法收敛速度;同时引入新型遗传算法混合交叉变异机制,避免了粒子群陷入局部最优。将自适应粒子群遗传算法与标准粒子群算法、遗传算法、人工蜂群算法进行了比较,仿真实验结果表明该算法在迭代60次左右完成参数辨识,各参数的辨识相对误差均降低到了1%以内。最后利用旋转柔性关节实验平台进行了实验验证,实验结果证明了该算法具有更好的收敛速度和寻优精度。

Dynamic Parameter Identification of Flexible Joint Robot Based on Adaptive Particle Swarm Optimization-genetic Algorithm

A method of dynamic parameter identification of flexible joint robot based on adaptive particle swarm genetic algorithm was proposed. The algorithm adopted dynamic adaptive adjustment strategy to improve the convergence speed of particle swarm algorithm. At the same time, a new hybrid genetic algorithm was introduced to avoid particle swarm optimization. Adaptive particle swarm genetic algorithm was compared with the standard particle swarm algorithm, genetic algorithm and artificial swarm algorithm, and the simulation results showed that the algorithm performs parameter identification after about 60 iterations, and the relative error of each parameter was reduced to less than 1%. Finally, the experimental verification was carried out by using the rotating flexible joint experimental platform, and the experimental results showed that the algorithm has better convergence speed and optimization precision.