一种基于增强型烟花算法的自抗扰控制的机器鱼路径跟踪控制方法

A Path Tracking Method of Robot Fish Based on Enhanced Fireworks Algorithm-Auto Disturbance Rejection Control

机器鱼在未知水域进行路径跟踪时受到内外多种未知干扰因素的影响。自抗扰控制(auto disturbance rejection control,ADRC)具有较强的适应性、鲁棒性,是解决非线性、不确定性、强干扰、强耦合、大时滞等控制问题的一种有效方法。为了获得高质量的跟踪效果,提出基于增强型烟花算法的自抗扰控制(enhanced fireworks algorithm-auto disturbance rejection control,EFWA-ADRC)的路径跟踪控制方法。针对ADRC中参数的优化问题,采用增强型烟花算法(EFWA)在线调整,以提高机器鱼在路径跟踪过程中的抗干扰性。以多关节仿生机器鱼为研究对象,结合机器鱼的数学模型,在Serret-Frenet坐标系下建立路径跟踪误差模型,重点设计前向速度和转向速度控制率,构建基于EFWA-ADRC的路径跟踪控制系统。最后仿真与实验结果显示基于EFWA-ADRC与常规ADRC的控制方法使机器鱼分别在2.8 s、3.3 s左右跟踪给定路径,且跟踪误差分别保持在正负0.09 m、0.1 m内,相比于常规ADRC,所提出的控制方法使跟踪的稳态误差降低10%,证明提出的EFWA-ADRC控制器对受到强干扰的被控系统具有更优的控制效果。

The robot fish is affected by many unknown internal and external interference factors when it performs path tracking in unknown waters.It is an effective method to solve the control problems of nonlinearity,uncertainty,strong interference,strong coupling and large time lag.A path tracking method was proposed based on the enhanced fireworks algorithm-auto disturbance rejection control(EFWA-ADRC)to obtain high-quality tracking effect.ADRC has strong adaptability and robustness.In order to optimize the parameters in ADRC,the enhanced fireworks algorithm(EFWA)was dopted to adjust online,and the anti-interference of the robot fish in the path tracking process was improved.The multi-joint bionic robot fish was used as the research object by combining with the mathematical model of the robot fish,the path tracking error model was established in the Serret-Frenet coordinate system,the forward speed and steering speed control rate were designed as the key points,a path tracking control system based on EFWA-ADRC was constructed.The simulation and experimental results show that the control method based on EFWA-ADRC and conventional ADRC makes the robotic fish track the given path at 2.8 s and 3.3 s respectively,and the tracking error is kept within plus or minus 0.09 m and 0.1 m respectively.This control method reduces the tracking steady-state error by 10%.It was proved that the proposed EFWA-ADRC controller has better control effect on the controlled system,which is subject to strong interference.