Real-Robot Deep Reinforcement Learning: Improving Trajectory Tracking of Flexible-Joint Manipulator with Reference Correction.
Dmytro PavlichenkoSven BehnkePublished in: CoRR (2022)
Keyphrases
- real robot
- trajectory tracking
- control law
- closed loop
- reinforcement learning
- visual servoing
- mobile robot
- control system
- bi directional
- dynamic model
- iterative learning
- control algorithm
- control theory
- control scheme
- sliding mode
- physical constraints
- nonlinear systems
- optimal control
- control strategy
- adaptive control
- iterative learning control
- control method
- simulated robot
- variable structure
- motion planning
- autonomous robots
- robotic systems
- linear model
- tracking error
- computer vision
- state space
- real time
- real world
- machine learning
- learning process
- dynamic programming
- robot control
- vision system
- path planning
- stability analysis
- function approximation
- multi robot
- control parameters