Active disturbance rejection decoupling control for nonlinear MIMO uncertain systems with application to path following of self-driving bust

This paper focuses on the control problem for a general class of multi-input multi-output (MIMO) uncertain systems with unknown coupling dynamics, uncertain control input gain matrix and external disturbances. The active disturbance rejection decoupling control is proposed to force the closed-loop systems to achieve desired performance despite these uncertainties. In particular, the general extended state observer corresponding the relative degree vector of considered system is constructed to timely estimate the total disturbance vector. The tuning laws for our controller to handle the uncertainty of control input gain matrix are quantitatively presented. Also, it is proven that the tracking error between the state vector and its ideal trajectory in the entire transient process can be small enough by tuning ESO's bandwidths. Moreover, the necessary and sufficient condition for closed-loop system's tracking error and estimation error to converge to zero is given. The proposed method is applied to the path following of self-driving bus (SDB), where the lateral error subsystem and the velocity subsystem are severely coupled with each other. Both unknown road-tire friction and steering wheel friction need to be handled. The experiment results on SDB demonstrate that the desired performance can be achieved despite both external disturbance and nonlinear uncertainties.

Automated summary

This paper proposes an active disturbance rejection decoupling control method to address the control problem of multi-input multi-output uncertain systems. The method can handle unknown coupling dynamics, uncertain control input gain matrix, and external disturbances. A general extended state observer is constructed to estimate the total disturbance vector, and tuning laws are presented to handle the uncertainty of the control input gain matrix. The proposed method is applied to the path following of self-driving bus, and experimental results demonstrate its effectiveness in achieving desired performance despite external disturbances and nonlinear uncertainties.