基于改进蜂群算法的倒立摆控制器优化设计

LQR Controller Optimization for Double Inverted Pendulum Based on Improved Artificial Bee Colony Algorithm

摘要:

环形倒立摆系统以其非线性、不稳定、高阶的、强耦合的特性,已经成为控制理论教学以及模拟机器人行走、卫星飞行姿态等复杂系统控制的重要实验平台;文章中引入了LQR控制算法,实现对倒立摆的稳定控制；而为了避免LQR控制器设计过程中反复调节加权矩阵Q,R的复杂性,同时得到更优异的控制效果,将人工蜂群(ABC)算法应用于LQR控制器设计中;而为了克服传统蜂群算法的缺陷,加快收敛速度和增加种群的多样性,在雇佣蜂阶段和跟随蜂阶段对算法进行了改进;仿真结果表明,改善的ABC算法的能够很好地完成LQR控制器参数寻优,控制效果良好。

Abstract:

As a nonlinear, instability, high-order, strong coupling system, the rotary double inverted pendulum has become an ideal platform. In this paper, LQR control is introduced to achieve its stable control. In order to avoid the complexity of adjusting parameters manually, meanwhile obtain optimal control effect, the artificial bee colony (ABC) algorithm is applied to optimize the LQR controller. Furthermore, in order to accelerate the convergence speed and increase the diversities of population, the ABC algorithm has been improved during employed bees phase and the onlooker bees phase. The simulation results show that the improved ABC algorithm is able to complete parameter optimization for LQR controller of inverted pendulum system.