Reinforcement Learning for Robotic Assembly Using Non-Diagonal Stiffness Matrix.
Masahide OikawaTsukasa KusakabeKyo KutsuzawaSho SakainoToshiaki TsujiPublished in: IEEE Robotics Autom. Lett. (2021)
Keyphrases
- reinforcement learning
- covariance matrix
- robotic tasks
- symmetric positive definite
- real time
- real robot
- multi agent
- mobile robot
- markov decision processes
- function approximation
- robotic control
- positive definite
- singular value decomposition
- state space
- low rank
- robotic arm
- position control
- liquid crystal displays
- autonomous robots
- reinforcement learning algorithms
- temporal difference
- robotic systems
- control strategies
- robot control
- model free
- reinforcement learning methods
- optimal control
- manipulation tasks
- force control
- transfer learning
- learning process
- learning algorithm
- machine learning