2021/03/02 by Michael O’Connell, Michael O'Connell, O'Connell, Michael +7 · 1 citation
Computer Science · Engineering · Physics and Astronomy · #Adaptive Dynamic Programming Control #Aerospace and Aviation Technology #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Model Reduction and Neural Networks #Robotics (cs.RO) #Systems and Control (eess.SY) #cs.LG #cs.RO #cs.SY #eess.SY #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2103.01932
7 pages, 7 figures; this article is an early draft and presents preliminary results; the full method and improved results were published in Science Robotics on May 4th, 2022: doi.org/10.1126/scirobotics.abm6597; arXiv: doi.org/10.48550/arXiv.2205.06908
openalex publication_date 2021/03/02 · arxiv created 2022/05/25 · arxiv updated 2022/05/26 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Realtime model learning proves challenging for complex dynamical systems, such as drones flying in variable wind conditions. Machine learning technique such as deep neural networks have high representation power but is often too slow to update onboard. On the other hand, adaptive control relies on simple linear parameter models can update as fast as the feedback control loop. We propose an online composite adaptation method that treats outputs from a deep neural network as a set of basis functions capable of representing different wind conditions. To help with training, meta-learning techniques are used to optimize the network output useful for adaptation. We validate our approach by flying a drone in an open air wind tunnel under varying wind conditions and along challenging trajectories. We compare the result with other adaptive controller with different basis function sets and show improvement over tracking and prediction errors.