基于PSO的非静止流场中机器鱼转向控制参数优化

Optimization of Steering Control Parameters of Robot Fish in Variable Flow Field Based on PSO

机器鱼在航行中容易受到非静止流场的干扰而偏离目标航向。文中采用航向角反馈来解决未搭载流场感知相关传感器的机器鱼航向偏离问题。首先,建立机器鱼的关节动力学模型与水动力学模型,得到关节角运动与关节力矩之间的关系,并且获得机器鱼推进力和转向力矩与机器鱼摆动姿态之间的关系。而后,将中枢模式发生器控制器运用于机器鱼的控制,与受控对象构成闭环控制系统,使机器鱼在控制器的调节作用下保持航向稳定。最后,为获取最佳控制器参数,采用粒子群优化算法,以机器鱼从一个设定的航向角偏离量收敛到零的时间长度作为优化指标,最终实现快速转向。根据建立的机器鱼动力学模型进行了仿真分析,验证了文中设计方法的有效性和合理性。

Robotic fish are susceptible to interference from non-stationary flow fields and thus may deviate from the target course during navigation.In this study,course angle feedback is used to solve the problem of course deviation in a robotic fish without flow field sensors.First,a relationship between joint angular motion and joint torque is obtained by establishing the joint dynamics model of a robotic fish.In addition,a relationship between the propulsion and steering torque and swing posture is obtained.Subsequently,to maintain the stability of the robotic fish,a central pattern generator controller is used to adjust the closed-loop control system.Furthermore,this study takes the length of time during which the robot fish converges from a set course angle deviation to zero as an optimization index,and uses the particle swarm optimization algorithm to obtain the best controller parameters that can achieve rapid steering.The simulation analysis is performed based on the established dynamics model of a robotic fish,and the results verify the effectiveness and rationality of the proposed design method.