From semantics to execution: Integrating action planning with reinforcement learning for robotic tool use.
Manfred EppePhuong D. H. NguyenStefan WermterPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- action selection
- plan execution
- plan generation
- durative actions
- partially observable domains
- graphical representation
- reward shaping
- state space
- mission planning
- action space
- ai planning
- learning algorithm
- control flow
- formal semantics
- function approximation
- real time
- multi agent
- mobile robot
- derived predicates
- temporal difference
- condition action rules
- macro actions
- robotic tasks
- concurrent execution
- heuristic search
- active rules
- robotic systems
- planning process
- action sequences
- action theories
- optimal policy
- action language
- markov decision processes
- logic programs
- conceptual model
- complex domains
- planning problems
- reinforcement learning algorithms