Extended-state-observer-based distributed model predictive formation control of under-actuated unmanned surface vehicles with collision avoidance

In this paper, distributed formation tracking control with collision avoidance is addressed for a group of under-actuated unmanned surface vehicles subject to physical constraints and dynamical uncertainties. An extended-state-observer-based distributed model predictive control method is proposed for achieving a safe formation. Specifically, the vehicle dynamics is firstly transformed into an almost spherical form consisting of a position motion subsystem and an angular motion subsystem. Next, an extended state observer is used to estimate unknown model uncertainties and external disturbances in each subsystem. After that, by taking physical constraints and collision avoidance requirements into account, a distributed model predictive position tracking controller and a model predictive angular motion controller are designed based on the recovered model information through the extended state observers. The distributed formation control with collision avoidance problem is formulated as a constrained quadratic programming problem, which can be locally solved in a decentralized manner. Finally, the simulation results of five under-actuated unmanned surface vehicles substantiate the effectiveness of the proposed extended-state-observer-based distributed model predictive control method for multiple under-actuated unmanned surface vehicles.

Automated summary

A distributed formation tracking control method for under-actuated unmanned surface vehicles is proposed in this paper, utilizing an extended-state-observer-based model predictive control approach. The vehicles' dynamics are transformed and uncertainties are estimated to design position and angular motion controllers, solving the problem as a constrained quadratic programming. Simulation results validate the effectiveness of the proposed method for multiple vehicles.