Robust Model Reference Adaptive Consensus with Neural Networks
D. Yue (Southeast University, TU Delft - Team Bart De Schutter)
S Baldi (TU Delft - Team Bart De Schutter, Southeast University)
Jinde Cao (Southeast University)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
This paper addresses distributed and robust leaderless consensus control for a class of uncertain multiagent systems with matched unknown nonlinearities and disturbances. The problem is challenging due to the lack of a leader (reference signal), the large uncertainties in agent dynamics, and the asymmetric communications among the agents. A novel neural network embedded model reference adaptive consensus (NN-MRACon) framework is proposed, which bridges NN and MRACon by means of nonsmooth control. Asymptotic consensus is proved based on robust analysis and input-to-state stability theory. Numerical examples on networks of second-order integrators and two-mass-spring systems are included to validate the effectiveness of NN-MRACon.