Optimal Control of a Two-Wheeled Self-Balancing Robot by Reinforcement Q-learning.
Linyuan GuoSyed Ali Asad RizviZongli LinPublished in: ICCA (2020)
Keyphrases
- optimal control
- reinforcement learning
- mobile robot
- real robot
- function approximation
- control problems
- feedback control
- reinforcement learning algorithms
- dynamic programming
- state space
- inverted pendulum
- model free
- risk sensitive
- robotic systems
- path planning
- infinite horizon
- markov decision processes
- autonomous robots
- continuous state and action spaces
- policy iteration
- multi robot
- motion planning
- class of nonlinear systems
- action selection
- optimal policy
- learning agent
- reinforcement learning methods
- optimal control problems
- humanoid robot
- learning algorithm
- actor critic
- temporal difference
- control law
- lyapunov function
- markov decision problems
- adaptive control
- state action
- robot manipulators
- average cost
- real time
- machine learning
- single agent
- dynamic environments