Iterative Learning without Reinforcement or Reward for Multijoint Movements: A Revisit of Bernstein's DOF Problem on Dexterity.
Suguru ArimotoMasahiro SekimotoKenji TaharaPublished in: J. Robotics (2010)
Keyphrases
- iterative learning
- reinforcement learning
- end effector
- degrees of freedom
- iterative learning control
- agent receives
- minimally invasive surgery
- error reduction
- incremental learning
- pose estimation
- hand eye coordination
- trajectory tracking
- joint angles
- machine learning
- control theoretic
- inverse kinematics
- robot arm
- robotic arm
- vision system
- significant improvement
- robot manipulators
- reward function
- motion planning
- learning process
- training set
- data sets