Neural network-based modeling of robot action effects in conceptual state space of real world.
Witold JacakStephan DreiseitlRobert MuszynskiPublished in: IROS (1996)
Keyphrases
- state space
- real world
- mobile robot
- neural network
- reinforcement learning
- wide range
- synthetic data
- case study
- real robot
- data sets
- computer vision
- heuristic search
- robot control
- indirect effects
- action space
- robot navigation
- real world environments
- obstacle avoidance
- initial state
- robot manipulators
- simulated robot
- state transition
- autonomous robots
- state variables
- robotic systems
- conceptual model
- markov decision processes
- domain independent
- particle filter
- vision system
- control system