Motion Planning as Online Learning: A Multi-Armed Bandit Approach to Kinodynamic Sampling-Based Planning.
Marco FaroniDmitry BerensonPublished in: CoRR (2023)
Keyphrases
- motion planning
- online learning
- multi armed bandit
- degrees of freedom
- trajectory planning
- regret bounds
- mobile robot
- humanoid robot
- robotic tasks
- path planning
- multi armed bandits
- multi robot
- e learning
- belief space
- collision free
- robotic arm
- multi modal
- reinforcement learning
- configuration space
- active learning
- mechanical systems
- pose estimation
- viewpoint