Abstract:
This paper introduces a model-reference reinforcement learning algorithm for intelligent tracking control of uncertain autonomous surface vehicles with collision avoidance. The proposed control algorithm combines reinforcement learning with conventional control to improve control accuracy and intelligence. A nominal system is used to design a baseline tracking controller using conventional control. Uncertain autonomous surface vehicles in an obstacle-free environment should behave according to the nominal system. Reinforcement learning allows the tracking controller to compensate for model uncertainties and avoid collisions in obstacles. Our learning-based control provides stability guarantees and better sample efficiency than deep reinforcement learning methods. The new algorithm is tested using autonomous surface vehicles.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
