Multi-fidelity reinforcement learning for time-optimal quadrotor re-planning
Name
ryou-et-al-2025-multi-fidelity-reinforcement-learning-for-time-optimal-quadrotor-re-planning.pdf
Description
Published version
Size
3.71 MB
Format
Adobe PDF
Checksum (MD5)
9766dc8c7705d4692466bda2fe900973
Author(s)
Ryou, Gilhyun
Wang, Geoffrey
Karaman, Sertac
Date Issued
August 22, 2025
Journal
The International Journal of Robotics Research
Publisher
SAGE Publications
Citation
Ryou G, Wang G, Karaman S. Multi-fidelity reinforcement learning for time-optimal quadrotor re-planning. The International Journal of Robotics Research. 2025;0(0).
Version
Final published version
Abstract
High-speed online trajectory planning for UAVs poses a significant challenge due to the need for precise modeling of complex dynamics while also being constrained by computational limitations. This paper presents a multi-fidelity reinforcement learning method (MFRL) that aims to effectively create a realistic dynamics model and simultaneously train a planning policy that can be readily deployed in real-time applications. The proposed method involves the co-training of a planning policy and a reward estimator; the latter predicts the performance of the policy’s output and is trained efficiently through multi-fidelity Bayesian optimization. This optimization approach models the correlation between different fidelity levels, thereby constructing a high-fidelity model based on a low-fidelity foundation, which enables the accurate development of the reward model with limited high-fidelity experiments. The framework is further extended to include real-world flight experiments in reinforcement learning training, allowing the reward model to precisely reflect real-world constraints and broadening the policy’s applicability to real-world scenarios. We present rigorous evaluations by training and testing the planning policy in both simulated and real-world environments. The resulting trained policy not only generates faster and more reliable trajectories compared to the baseline snap minimization method, but it also achieves trajectory updates in 2 ms on average, while the baseline method takes several minutes.
MIT Department
Massachusetts Institute of Technology. Laboratory for Information and Decision Systems
Terms of Use
Creative Commons Attribution-Noncommercial
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1177/02783649251364393
