Abstract
This article proposes inverse reinforcement learning (IRL) algorithms to solve the optimal synchronization problems for multiagent systems (MASs) subject to disturbances. A framework of expert–learner MASs is developed to reconstruct the unknown cost functions of the expert MAS. Within this framework, the learner MAS imitates the behavior trajectories of expert MAS under observation. We first present a model-based IRL algorithm including two iteration loops: an inner loop on optimal control and an outer loop on inverse optimal control (IOC). Moreover, a model-free IRL algorithm is further proposed using the observed input and state data without requiring the dynamics knowledge of the two MASs. Then, generalized fuzzy hyperbolic models (GFHMs) are utilized to implement the model-free algorithm, and we provide the convergence and stabilization proofs of our proposed algorithms. Finally, the effectiveness of the proposed algorithms is verified through a simulation example.
Funding Information
  • National Natural Science Foundation of China (62573238)
  • Basic Research Program of Jiangsu Province (BK20241889)
  • Natural Science Foundation of Nanjing University of Posts and Telecommunications (NY223080)
  • Natural Science Foundation of Nanjing University of Posts and Telecommunications (NY224174)