Modular production control using deep reinforcement learning: proximal policy optimization.
Sebastian MayerTobias ClassenChristian EndischPublished in: J. Intell. Manuf. (2021)
Keyphrases
- reinforcement learning
- control policy
- optimal policy
- action selection
- control policies
- control problems
- optimal control
- markov decision process
- optimization algorithm
- industrial production
- policy search
- state space
- robot control
- control system
- partially observable
- optimization problems
- production processes
- function approximation
- machine learning
- actor critic
- production process
- control strategies
- infinite horizon
- markov decision processes
- partially observable environments
- learning algorithm
- policy gradient
- inverse reinforcement learning
- state and action spaces
- robotic control
- continuous state
- reward function
- function approximators
- finite horizon
- production cost
- reinforcement learning algorithms
- temporal difference
- model free
- long run
- control strategy
- production system
- mobile robot
- dynamic programming
- genetic algorithm