Neural network-based finite-horizon approximately optimal control of uncertain affine nonlinear continuous-time systems.
Hao XuQiming ZhaoTravis DierksSarangapani JagannathanPublished in: ACC (2014)
Keyphrases
- optimal control
- infinite horizon
- finite horizon
- class of nonlinear systems
- dynamic programming
- average cost
- lyapunov function
- optimal policy
- control strategy
- reinforcement learning
- linear quadratic
- stochastic demand
- control law
- feedback control
- optimal control problems
- production planning
- brownian motion
- neural network
- single item
- inventory models
- continuous stirred tank reactor
- markov decision process
- markov decision processes
- periodic review
- multistage
- non stationary
- decision making