种群随机化遗传算法在仿生机械手姿态的应用研究

Research on the Application of Population Randomization Genetic Algorithm in the Posture of Bionic Manipulator

针对人机交互仿生机械手模仿人类行为的识别误差和能量消耗等问题,提出一种基于种群随机化的多目标遗传优化算法。该算法同时考虑人机交互机械手姿态识别优化中的终端误差、轨迹误差和能耗的影响,采用基于并行选择的种群随机化方法,将每一代中自适应性最优的个体遗传给下一代种群,实现自适应性和定位功能。为验证所提算法的有效性,通过六自由度机械手进行交互实验。结果表明:所提出的方法不仅可以降低能耗,而且在保证追踪机械手臂跟踪效果的同时,可将终端误差降低到3 mm以下;与其他算法相比,该算法具有能量消耗少、轨迹误差小的优势,可有效提高机械臂姿态适应的准确性。

Aiming at the problems of recognition error and energy consumption of human-computer interaction bionic manipula⁃tor,a multi-objective genetic optimization algorithm based on population randomization was proposed.In this algorithm,the influ⁃ences of terminal error,tracking error and energy consumption in the optimization of human-computer interaction manipulator attitude recognition were considered at the same time;the population randomization method based on parallel selection was adopted,and the adaptive optimal individuals in each generation were inherited to the next generation population to realize the adaptive and positioning functions.In order to verify the effectiveness of this algorithm,an interactive experiment was carried out on a 6 DOF manipulator.The results show that by using the proposed method,not only the energy consumption can be reduced,but also the terminal error can be reduced to less than 3 mm while ensuring the tracking effect of the tracking manipulator;compared with other algorithms,the proposed algorithm has the advantages of less energy consumption and small tracking error,and the accuracy of manipulator attitude adaptation can be effectively improved.