AIB-MDP: Continuous Probabilistic Motion Planning for Automated Vehicles by Leveraging Action Independent Belief Spaces.
Maximilian NaumannChristoph StillerPublished in: IROS (2022)
Keyphrases
- motion planning
- belief space
- action space
- configuration space
- degrees of freedom
- path planning
- mobile robot
- markov decision processes
- trajectory planning
- robot arm
- initial state
- state space
- humanoid robot
- probability theory
- robotic tasks
- robotic arm
- reinforcement learning
- probabilistic planning
- multi robot
- autonomous mobile robot
- obstacle avoidance
- manipulation tasks
- markov decision process
- bayesian networks
- inverse kinematics
- nearest neighbor searching
- planning under uncertainty
- real time
- mechanical systems
- climbing robot
- belief revision
- computer vision
- belief functions
- heuristic search
- optimal policy
- control system
- discounted reward
- image sequences