Online Replanning in Belief Space for Partially Observable Task and Motion Problems.
Caelan Reed GarrettChris PaxtonTomás Lozano-PérezLeslie Pack KaelblingDieter FoxPublished in: ICRA (2020)
Keyphrases
- belief space
- partially observable
- partial observability
- decision problems
- classical planning
- motion planning
- belief state
- state space
- planning under uncertainty
- reinforcement learning
- markov decision problems
- partially observable markov decision processes
- fully observable
- markov decision processes
- dynamic environments
- space time
- dynamical systems
- image sequences
- multiple agents
- planning domains
- approximate solutions
- initial state
- domain independent
- markov chain
- dynamic programming