On Your Own - Pro-Level Autonomous Drone Racing in Uninstrumented Arenas
Michael Bosello, Flavio Pinzarrone, Sara Kiade, Davide Aguiari, Yvo Keuter, Aaesha AlShehhi, Gyordan Caminati, Kei Long Wong, Ka Seng Chou, Junaid Halepota, Fares Alneyadi, Jacopo Panerati, Giovanni Pau
AI summary
Problem
Current autonomous drone systems are typically trained and evaluated only in highly controlled, instrumented environments, limiting their real-world applicability. There is a need to demonstrate robust, vision-based autonomy in unstructured, uninstrumented arenas where traditional navigation fails.
Approach
The authors developed a perception and control stack using stereo-camera vision, drift-corrected visual-inertial odometry, and model predictive control to navigate gates at high speeds without relying on external tracking or fine-tuning with ground truth.
Key results
- Achieved pro-level autonomous racing speeds in both instrumented and uninstrumented tracks
- Matched professional human pilot performance in head-to-head uninstrumented flights
- Released a new dataset of pro-level human-piloted drone flights
- Demonstrated a resilient perception stack requiring minimal ground-truth fine-tuning
Why it matters
This work bridges the gap between controlled lab benchmarks and real-world commercial drone operations by proving vision-based autonomy can handle unstructured environments at professional speeds.
Abstract
Drone technology is proliferating in many industries, including agriculture, logistics, defense, infrastructure, and environmental monitoring. Vision- based autonomy is one of its key enablers, particularly for real-world applications. This is essential for operating in novel, unstructured environments where traditional navigation methods may be unavailable. Autonomous drone racing has become the de facto benchmark for such systems. State-of-the-art research has shown that autonomous systems can surpass human-level performance in racing arenas. However, the direct applicability to commercial and field operations is still limited, as current systems are often trained and evaluated in highly controlled environments. In our contribution, the system’s capabilities are analyzed within a controlled environment—where external tracking is available for ground-truth comparison—but also demonstrated in a challenging, uninstrumented environment—where ground- truth measurements were never available. We show that our approach can match the performance of professional human pilots in both scenarios. We also publicly release the data from flights carried out by a world-class human pilot: github.com/tii-racing/drone-racing-dataset. Video: youtube.com/watch?v=SNw-zXgv vA