Distributed control by Lagrangian steepest descent.
David H. WolpertStefan BieniawskiPublished in: CDC (2004)
Keyphrases
- distributed control
- autonomous agents
- cooperative
- learning rate
- control architecture
- multi agent
- prize collecting
- information technology
- numerical methods
- objective function
- relaxation algorithm
- multi agent reinforcement learning
- lagrangian relaxation
- optimal solution
- agent interactions
- information exchange
- vision system
- real time