2022/10/18 by Andrea Tagliabue, Jonathan P. How, Tagliabue, Andrea +1
Computer Science · Engineering · #Advanced Control Systems Optimization #Algorithm #Artificial intelligence #Computation #Computer science #Control (management) #Control theory (sociology) #Data-driven #Engineering #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine learning #Process (computing) #Reinforcement Learning in Robotics #Robot Manipulation and Learning #Robotics (cs.RO) #Robustness (evolution) #Task (project management) #cs.LG #cs.RO
paper · pdf · doi:10.48550/arxiv.2210.10127
published in arXiv (Cornell University) (Cornell University) · Accepted to IROS 22
arxiv created 2022/10/18 · openalex publication_date 2022/10/18 · arxiv updated 2022/10/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
Imitation learning (IL) can generate computationally efficient sensorimotor policies from demonstrations provided by computationally expensive model-based sensing and control algorithms. However, commonly employed IL methods are often data-inefficient, requiring the collection of a large number of demonstrations and producing policies with limited robustness to uncertainties. In this work, we combine IL with an output feedback robust tube model predictive controller (RTMPC) to co-generate demonstrations and a data augmentation strategy to efficiently learn neural network-based sensorimotor policies. Thanks to the augmented data, we reduce the computation time and the number of demonstrations needed by IL, while providing robustness to sensing and process uncertainty. We tailor our approach to the task of learning a trajectory tracking visuomotor policy for an aerial robot, leveraging a 3D mesh of the environment as part of the data augmentation process. We numerically demonstrate that our method can learn a robust visuomotor policy from a single demonstration--a two-orders of magnitude improvement in demonstration efficiency compared to existing IL methods.