Designing Decentralized Controllers for Distributed-Air-Jet MEMS-Based Micromanipulators by Reinforcement Learning.
Laëtitia MatignonGuillaume J. LaurentNadine Le Fort-PiatYves-André ChapuisPublished in: J. Intell. Robotic Syst. (2010)
Keyphrases
- reinforcement learning
- multi agent
- distributed systems
- peer to peer
- cooperative
- decentralized control
- cooperative multi agent systems
- function approximation
- distributed agents
- control system
- distributed environment
- state space
- structured peer to peer
- lightweight
- peer to peer networks
- temporal difference
- real time
- markov decision processes
- neural network
- distributed constraint optimization
- control policies
- distributed data
- load balancing
- model free
- open systems
- learning classifier systems
- optimal policy
- machine learning
- fault tolerant
- learning problems