Abstract:
Time-driven learning updates prediction model parameters as new data arrives in this work. The direct heuristic dynamic programming (dHDP) algorithm has been shown to solve several complex learning control problems. As system states change, it updates control policy and critic. Thus, noise should not update the time-driven dHDP. We propose an event-driven dHDP. A Lyapunov function candidate proves the uniformly ultimately boundedness (UUB) of system states and weights in the critic and control policy networks. Thus, the approximate control and cost-to-go function approach Bellman optimality within a finite bound. We also compare the event-driven dHDP algorithm to the time-driven one.
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
