Finite horizon optimal control of non-linear discrete-time switched systems using adaptive dynamic programming with ε-error bound.
Chunbin QinHuaguang ZhangYanhong LuoBinrui WangPublished in: Int. J. Syst. Sci. (2014)
Keyphrases
- optimal control
- infinite horizon
- dynamic programming
- finite horizon
- error bounds
- optimal control problems
- optimal policy
- markov decision processes
- linear quadratic
- feedback control
- production planning
- average cost
- control strategy
- single product
- stochastic demand
- multistage
- state space
- single item
- reinforcement learning
- markov decision process
- lost sales
- fixed cost
- brownian motion
- long run
- worst case
- theoretical analysis
- probabilistic model
- inventory control
- markov decision problems
- inventory models
- markov chain