Planning for Human-Robot Interaction Using Time-State Aggregated POMDPs.
Frank BrozIllah R. NourbakhshReid G. SimmonsPublished in: AAAI (2008)
Keyphrases
- human robot interaction
- belief state
- human robot
- partially observable
- state space
- stochastic domains
- partially observable markov decision processes
- gesture recognition
- predictive state representations
- robot programming
- humanoid robot
- service robots
- markov decision processes
- reinforcement learning
- planning problems
- human centered
- heuristic search
- decision making