Finite-time adaptive optimal consensus control for multi-agent systems subject to time-varying output constraints

Highlights：

• Compared with the existed optimal control algorithms with infinite-time convergence, the proposed control protocol can not only achieve the goal of optimization under worst-case perturbations, but also guarantee the stability of MASs in the finite time, which effectively accelerates the convergence rate.

• In order to reduce the complexity of the network weights updating algorithm, a new positive function is constructed in the RL algorithm proposed in this paper to simplify the design of adaptive learning rules. Meanwhile, the persistent excitation condition that is difficult to verify online is no longer needed.

摘要

•By designing proper barrier functions, the output of every follower is successfully constrained in the desired time-varying set. Compared with most existing constraint control strategies, the dynamic output constraint problem considered in this paper is a more general case. Here, the constraint boundary functions can be asymmetric, or have the same symbol, or even one of the boundary functions is equal to zero. Therefore, the proposed control strategy is more in line with the practical engineering requirements.•Compared with the existed optimal control algorithms with infinite-time convergence, the proposed control protocol can not only achieve the goal of optimization under worst-case perturbations, but also guarantee the stability of MASs in the finite time, which effectively accelerates the convergence rate.•In order to reduce the complexity of the network weights updating algorithm, a new positive function is constructed in the RL algorithm proposed in this paper to simplify the design of adaptive learning rules. Meanwhile, the persistent excitation condition that is difficult to verify online is no longer needed.