Distributed Finite-Time ADP-Based Optimal Control for Nonlinear Multiagent Systems
Longjie Zhang, Yong Chen
Abstract
In this brief, the finite-time stability of the optimal control is considered and the ADP-based finite-time distributed optimal controller (DFTOC) is proposed for the nonlinear multiagent systems. Firstly, to obtain the candidate optimal value function with finite-time convergence space, the sign function is applied to map the value function into the finite-time convergence space. The finite-time optimal controller is obtained by solving the redesigned Hamilton-Jacobi-Bellman (HJB) equation with the mapped value function, and the finite-time stability is illustrated theoretically. Moreover, to solve the proposed finite-time optimal controller, the finite-time adaptive dynamic programming (FTADP) algorithm is further implemented based on the finite-time optimization method. Finally, compared with optimal control under the sense of asymptotic stability, the result analysis with Van der Pol’s oscillator system shows the superiority of the proposed DFTOC.