Policy Iteration Adaptive Dynamic Programming for Optimal Control of Multi-Player Stackelberg-Nash Games

This paper investigates multi-player Stackelberg-Nash (SN) game problems of nonlinear continuous-time systems via policy iteration adaptive dynamic programming (ADP). To represent different hierarchical roles, the appropriate cost functions of the leader and each follower are designed. By introducin...

Full description

Saved in:
Bibliographic Details
Published in:Chinese Control Conference pp. 2393 - 2397
Main Authors: Lin, Mingduo, Zhao, Bo, Liu, Derong, Zhang, Yongwei
Format: Conference Proceeding
Language:English
Published: Technical Committee on Control Theory, Chinese Association of Automation 25.07.2022
Subjects:
Adaptive Dynamic Programming
Adaptive systems
Approximation algorithms
Artificial neural networks
Cost function
Games
Heuristic algorithms
Multi-Player systems
Neural Networks
Optimal control
Stackelberg Games
ISSN:1934-1768
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper investigates multi-player Stackelberg-Nash (SN) game problems of nonlinear continuous-time systems via policy iteration adaptive dynamic programming (ADP). To represent different hierarchical roles, the appropriate cost functions of the leader and each follower are designed. By introducing the ADP technique, the policy iteration algorithm is developed to obtain approximate solutions of the coupled HJ equation of each player. Then, the multi-player SN equilibrium is derived to guarantee the stability of the closed-loop system. Furthermore, the developed method is realized by employing the critic neural networks through the gradient-based weight updating algorithm. Finally, simulation example validates the effectiveness of the present method.
ISSN:1934-1768
DOI:10.23919/CCC55666.2022.9901882