Data-driven distributed formation control of under-actuated unmanned surface vehicles with collision avoidance via model-based deep reinforcement learning

This paper addresses the distributed formation control with collision avoidance for multiple under-actuated unmanned surface vehicles (USVs) subject to fully unknown models. A fully data-driven distributed control ap-proach is proposed for multiple USVs to achieve a desired formation based on model-based deep reinforcement learning. Specifically, a deep neural network is firstly trained to approximate the dynamic model of each USV by utilizing recorded input and output data. Then, by taking collision avoidance requirements into account, the model predictive formation controllers are proposed for USVs to achieve the safe formation control task based on the learned vehicle dynamics. It is shown that after learning with offline and online data, the proposed fully data-driven distributed controllers are able to achieve a safe formation. Simulations results are given to substantiate the feasibility and efficacy of the proposed model-based deep reinforcement learning method for distributed formation control of under-actuated USVs with fully unknown models.

Automated summary

This paper proposes a fully data-driven distributed control approach based on model-based deep reinforcement learning for multiple under-actuated unmanned surface vehicles (USVs) with fully unknown models, in order to achieve a desired formation and collision avoidance. The dynamic models of each USV are approximated by training a deep neural network with recorded input and output data. Model predictive formation controllers are then proposed to achieve safe formation control while considering collision avoidance. Simulation results demonstrate the feasibility and efficacy of the proposed method.