This paper presents a novel system for autonomous, vision-based drone racing combining learned data abstraction, nonlinear filtering, and time-optimal trajectory planning. The system has successfully been deployed at the first autonomous drone racing world championship: the . Contrary to traditional drone racing systems, which only detect the next gate, our approach makes use of any visible gate and takes advantage of multiple, simultaneous gate detections to compensate for drift in the state estimate and build a global map of the gates. The global map and drift-compensated state estimate allow the drone to navigate through the race course even when the gates are not immediately visible and further enable to plan a near time-optimal path through the race course in real time based on approximate drone dynamics. The proposed system has been demonstrated to successfully guide the drone through tight race courses reaching speeds up to and ranked second at the .
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8827337 | PMC |
| http://dx.doi.org/10.1007/s10514-021-10011-y | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Promoting Public Health Through Drone Sports Within Diverse Communities of Middle- and High School Students.
Public Health Rev
December 2024
RdyTechGo, Minneapolis, MN, United States.
Optimality principles in spacecraft neural guidance and control.
Sci Robot
June 2024
Micro Air Vehicle Lab, Faculty of Aerospace Engineering, Delft University of Technology, 2629 HS Delft, Netherlands.
This Review discusses the main results obtained in training end-to-end neural architectures for guidance and control of interplanetary transfers, planetary landings, and close-proximity operations, highlighting the successful learning of optimality principles by the underlying neural models. Spacecraft and drones aimed at exploring our solar system are designed to operate in conditions where the smart use of onboard resources is vital to the success or failure of the mission. Sensorimotor actions are thus often derived from high-level, quantifiable, optimality principles assigned to each task, using consolidated tools in optimal control theory.
View Article and Find Full Text PDFCoordination of gaze and action during high-speed steering and obstacle avoidance.
PLoS One
March 2024
Cognitive Science Department, Rensselaer Polytechnic Institute, Troy, New York, United States of America.
When humans navigate through complex environments, they coordinate gaze and steering to sample the visual information needed to guide movement. Gaze and steering behavior have been extensively studied in the context of automobile driving along a winding road, leading to accounts of movement along well-defined paths over flat, obstacle-free surfaces. However, humans are also capable of visually guiding self-motion in environments that are cluttered with obstacles and lack an explicit path.
View Article and Find Full Text PDFReaching the limit in autonomous racing: Optimal control versus reinforcement learning.
Sci Robot
September 2023
University of Zurich, Zurich, Switzerland.
Article Synopsis
- The paper examines how to design better control systems for agile mobile robots, specifically in the context of autonomous drone racing.
- Researchers found that a neural network controller using reinforcement learning (RL) outperformed traditional optimal control (OC) methods due to RL's ability to optimize more relevant objectives.
- The study highlights the limitations of OC's planning-control decomposition, which constrains behavior, while RL adapts better to uncertainties, achieving impressive drone performance with rapid training.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!