基于模型的有限预测逆跟踪控制

Finite prediction inversion following control based on model

针对水下无人潜航器(unmanned underwater vehicle,UUV)非最小相位的特性,提出利用有限预测逆方法实现其精确的跟踪控制.利用系统逆方法计算当前时刻t_c的有界的逆输入,需要已知所有未来时刻的理想输出信息.为了有效利用未来时刻的理想输出数据,并提高计算效率,提出将时间窗内理想输出的预测信息用于逆算法中,来求取控制输入值,从而实现对UUV的路径跟踪控制.有限预测逆方法是利用从当前时刻到未来T_p时刻内的理想路径来获得逆输入,可以实现在线控制.通过给定的精确逆输入与利用有限预测逆方法所求得的输入之间的误差,对参数T_p和m进行优化.通过实验验证了该方法的有效性,并对比说明了跟踪误差与优化参数T_p之间的呈指数递减的关系.

Considering the non-minimum phase characteristics for unmanned underwater vehicle(UUV),a finite prediction inversion algorithm was proposed to implement the exact following control.The bounded inversion input at current instant t_c was calculated using system inversion method,and the desired output information at future instants must be known.In order to utilize the desired output data effectively and improve the calculating efficiency,the predictive information of desired output in the time window was introduced into the algorithm to obtain the control inputs,and the path following control would be implemented for UUV.The finite predictive inversion method was to introduce the desired path from current instant to future instant T_p to obtain inputs and to complete the real-time computation.Through the error of exact inversion inputs and inputs solved using finite predictive inversion method,the parameters T_p and m were optimized.Simulations are implemented to verify the viability of the algorithm proposed.The contrast simulations with different T_p clarify following error is exponentially decayed with the parameters Tp.