Recovering from External Disturbances in Online Manipulation through State-Dependent Revertive Recovery Policies.
Hongmin WuShuangqi LuoHongbin LinShuangda DuanYisheng GuanJuan RojasPublished in: RO-MAN (2018)
Keyphrases
- state dependent
- optimal policy
- steady state
- markov decision processes
- state space
- long run
- queueing networks
- external disturbances
- stationary distribution
- dynamic programming
- average cost
- reinforcement learning
- real time
- radial basis function neural network
- linear combination
- sufficient conditions
- markov chain
- arrival rate
- asymptotically optimal
- single server
- partially observable markov decision processes
- initial state
- service times