基于粒子群算法的直线二级倒立摆LQR控制器优化控制方法

Optimal control method of LQR controller two inverted pendulum based on particle swarm optimization algorithm

针对线性二次型调节器在倒立摆最优控制设计过程中对加权矩阵选择的盲目性,采用粒子群算法,引入惯性权重优化加权矩阵,获得状态反馈控制率,设计最优控制器。以直线二级倒立摆控制系统为研究对象,对比分析传统线性二次型调节器算法与基于粒子群优化线性二次型调节器算法的稳摆控制和抗干扰能力。结果表明:优化后的最优控制器能使系统的响应时间更快、超调量更小;直线二级倒立摆控制系统在4 s内达到稳定状态,在实时控制中表现出了较强的抗干扰能力。该算法对直线二级倒立摆的控制效果更理想。

This paper is an attempt to address the blindness affecting linear quadratic regulators used to select weighted matrices and during the optimal control design of an inverted pendulum. The research involves introducing inertial weights using particle swarm optimization and thereby optimizing the weighted matrices and, obtaining state feedback control rate, designing the optimal controller; and using the linear double inverted pendulum control system as the research object, comparing and analyzing the stability control and anti-jamming capability between the traditional linear quadratic regulator algorithm and the linear quadratic regulator algorithm based on particle swarm optimization. The results show that the optimal controller designed after optimization gives the system a faster response time and less overshoot, and allows the linear double inverted pendulum control system to achieve a stable state within 4 s, demonstrating a stronger anti-jamming capability in real-time control . The algorithm proves more ideal for controlling the linear double inverted pendulum.