基于遗传算法的欠驱动机器人模糊控制器设计

Fuzzy Controller Design for Underactuated Robots Based on GA Method

主要对二自由度平面欠驱动机器人的操作空间控制问题进行了研究,提出了一种适用于此类欠驱动机器人的模糊控制新方法。将欠驱动机器人末段位置控制问题进行分解,设计了一种新型的分层模糊控制器,模糊规则简单容易制定,并且大大减少了模糊规则数目。在控制规则制定方面引入遗传算法进行全局优化,根据定义的适应度函数从系统模型中自动生成全局最优的控制规则。最后进行了数值仿真,验证了方法的有效性。

The endpoint position control of two-dof underactuated planar robot was focused and a new fuzzy control method was proposed for this kind of underactuated robots, The endpoint position control of the underactuated system could be regarded as the resultant motion of folding or unfolding of the passive joint and the rotation of the active joint, A new kind of hierarchical fuzzy controller which could reduce a lot of fuzzy rules was proposed, Genetic algorithm was introduced to achieve the global optimization. Fuzzy rules were automatically generated by the way of the Genetic Algorithm from the system model according to the defined fitness function. At last, numerical simulation demonstrates the effectiveness of the proposed method.