基于遗传算法的仿生肌肉多目标优化控制研究

Research on Multi-objective Optimization Control of Artificial Bionic Muscle Based on the Genetic Algorithms

以动物肌肉结构和工作特性为模拟对象的仿生肌肉,其控制对象数量多,控制较为复杂。首先将仿生肌肉控制转化为多目标优化控制问题,将多目标函数合成为单一目标函数,以对该多目标优化问题进行求解;其次就遗传算法在该多目标优化求解问题中的应用进行了讨论;最后对由形状记忆合金构成的仿生肌肉进行仿真实验,结果表明仿生肌肉的多目标优化控制方法是一种有效的优化控制方法。

The control of bionic - structure artificial muscle based on the natural muscle is complex. It uses the multi - objective optimization to control bionic - structure artificial muscle, and applies the genetic algorithm to get the optimal solution. A test shows that the method is effective.