基于动觉智能图式的仿人智能控制理论设计小车单摆系统的控制器,包括:建立拉格朗日动力学模型,规约简化控制问题,划分控制阶段,设计子控制器并组成总控制器.基于Simulink构建仿真模型,将各控制阶段用Matlab程序实现后,与仿真模型相结合进行摆起和稳定控制的仿真研究.基于二进制和浮点数两种编码方式的遗传算法对小车单摆控制器参数进行优化,寻求最优控制参数.
The controller of inverted pendulum cart system (IPCS) is designed based on sensory-motor intelligent schema (SMIS) of human simulated intelligent control (HSIS) theory. The contriving process includes establishing Lagrange dynam- ics model, simplifying control issue, partitioning control phase, and designing sub-controllers which are gathered to compose chief-controller. The simulation model is built based on Simulink and each control phase is implemented in the form of Matlab program, then both are integrated to study the swing-up and stabilization characteristics of control. For seeking the optimization solution, parameters of IPCS controller are evolved by genetic algorithm (GA) with two different encoded modes, i.e. binary and float.