Robot arm reaching through neural inversions and reinforcement learning.
Pedro MartínJosé del R. MillánPublished in: Robotics Auton. Syst. (2000)
Keyphrases
- robot arm
- reinforcement learning
- natural actor critic
- fitted q iteration
- control strategies
- real robot
- motion planning
- inverse kinematics
- function approximation
- markov decision processes
- neural network
- position and orientation
- temporal difference
- nonlinear systems
- reinforcement learning algorithms
- model free
- optimal policy
- state space
- computer vision
- function approximators
- skill learning
- optimal control
- control strategy
- dynamic programming
- control system
- viewpoint
- object recognition
- three dimensional
- learning algorithm
- machine learning