Active end-effector pose selection for tactile object recognition through Monte Carlo tree search.
Mabel M. ZhangNikolay AtanasovKostas DaniilidisPublished in: IROS (2017)
Keyphrases
- monte carlo tree search
- end effector
- object recognition
- force feedback
- degrees of freedom
- d objects
- monte carlo
- inverse kinematics
- pose estimation
- robot arm
- position and orientation
- robot manipulators
- joint angles
- vision system
- visual servoing
- evaluation function
- computer vision
- configuration space
- temporal difference learning
- human pose
- game tree
- reinforcement learning methods
- visual feedback
- motion planning
- learning algorithm