Q-learning solution for optimal consensus control of discrete-time multiagent systems using reinforcement learning.
Chaoxu MuQian ZhaoZhongke GaoChangyin SunPublished in: J. Frankl. Inst. (2019)
Keyphrases
- control strategy
- multiagent systems
- optimal control
- reinforcement learning
- multi agent
- control system
- multiagent learning
- control problems
- function approximation
- optimal solution
- control policy
- dynamic programming
- rl algorithms
- learning agents
- reinforcement learning algorithms
- state space
- autonomous agents
- markov games
- multiagent reinforcement learning
- optimal policy
- multi agent reinforcement learning
- markov decision processes
- machine learning
- stochastic approximation
- cooperative
- multiagent planning
- coalition formation
- decentralized control
- action sets
- single agent
- model free
- finite state
- markov chain
- multi agent systems
- learning algorithm
- temporal difference
- action selection
- sufficient conditions
- adjustable autonomy