Control of an AUV with completely unknown dynamics and multi-asymmetric input constraints via off-policy reinforcement learning.
Mehdi MohammadiMohammad Mehdi ArefiNavid VafamandOkyay KaynakPublished in: Neural Comput. Appl. (2022)
Keyphrases
- reinforcement learning
- control problems
- optimal control
- robot control
- control signals
- control policy
- constraint satisfaction
- constrained optimization
- control strategies
- physical constraints
- feedback loop
- dynamical systems
- control system
- learning process
- neural network
- situational awareness
- cooperative
- highly nonlinear
- robotic control
- adaptive control
- function approximation
- tracking error
- multi agent
- policy search
- autonomous underwater vehicle