Model-Free Cooperative Control for Multi-Agent Systems Using the Approximate Dynamic Programming Approach.
Yanhua QuAnna WangJinglu LiuPublished in: IEEE Access (2018)
Keyphrases
- model free
- approximate dynamic programming
- multi agent systems
- policy iteration
- reinforcement learning
- cooperative
- control policy
- multi agent
- impedance control
- game theory
- reinforcement learning algorithms
- markov decision processes
- dynamic programming
- linear program
- function approximation
- optimal control
- average reward
- control system
- machine learning
- state space
- temporal difference
- fixed point
- action selection
- sufficient conditions
- step size
- monte carlo
- belief state
- control strategy
- linear programming
- support vector
- learning algorithm