Print Email Facebook Twitter Model+Learning-based Optimal Control Title Model+Learning-based Optimal Control: An Inverted Pendulum Study Author Baldi, S. (TU Delft Team Bart De Schutter; Southeast University) Rosa, Muhammad Ridho (Student TU Delft; Telkom University, Bandung) Wang, Yuzhang (Southeast University; Student TU Delft) Date 2020 Abstract This work extends and compares some recent model+learning-based methodologies for optimal control with input saturation. We focus on two methodologies: a model-based actor-critic (MBAC) strategy, and a nonlinear policy iteration strategy. To evaluate the performance of the algorithms, these strategies are applied to the swinging up an inverted pendulum. Numerical simulations show that the neural network approximation in the MBAC strategy can be poor, and the algorithm may converge far from the optimum. In the MBAC approach neither stabilization nor monotonic convergence can be guaranteed, and it is observed that the best value function is not always corresponding to the last one. On the other side the nonlinear policy iteration approach guarantees that every new control policy is stabilizing and generally leads to a monotonically decreasing cost. To reference this document use: http://resolver.tudelft.nl/uuid:552ddb83-fa0d-4d9f-943f-b50c4b27b027 DOI https://doi.org/10.1109/ICCA51439.2020.9264402 Publisher IEEE, Piscataway, NJ, USA ISBN 978-1-7281-9093-8 Source Proceedings of the IEEE 16th International Conference on Control and Automation, ICCA 2020 Event 16th IEEE International Conference on Control and Automation, ICCA 2020, 2020-10-09 → 2020-10-11, Virtual, Sapporo, Hokkaido, Japan Bibliographical note Accepted Author Manuscript Part of collection Institutional Repository Document type conference paper Rights © 2020 S. Baldi, Muhammad Ridho Rosa, Yuzhang Wang Files PDF root_swing3_short.pdf 957.98 KB Close viewer /islandora/object/uuid:552ddb83-fa0d-4d9f-943f-b50c4b27b027/datastream/OBJ/view