Composable Energy Policies for Reactive Motion Generation and Reinforcement Learning.
Julen UrainAnqi LiPuze LiuCarlo D'EramoJan PetersPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- image sequences
- motion segmentation
- motion analysis
- state space
- motion estimation
- markov decision process
- dynamic programming
- reward function
- space time
- motion model
- partially observable markov decision processes
- human motion
- reinforcement learning agents
- moving objects
- control policy
- control policies
- motion detection
- fitted q iteration
- motion field
- energy minimization
- optical flow
- spatial and temporal
- camera motion
- function approximation
- motion capture
- markov decision processes
- image segmentation
- learning process
- markov decision problems
- mobile robot
- feature points
- action selection
- infinite horizon
- finite state
- agent architecture
- motion parameters