Finite-horizon controllability and reachability for deterministic and stochastic linear control systems with convex constraints.
Daniel DueriBehçet AçikmeseMorgan BaldwinRichard Scott ErwinPublished in: ACC (2014)
Keyphrases
- finite horizon
- control policies
- convex constraints
- optimal policy
- control system
- infinite horizon
- markov decision processes
- linear constraints
- single product
- state space
- inventory control
- markov decision process
- convex optimization
- multistage
- reinforcement learning
- global consistency
- ordering cost
- quadratic program
- dynamic programming
- long run
- quadratic programming
- average cost
- non stationary
- convex programming