# An End-to-end Flight Control Network for High-speed UAV Obstacle Avoidance based on Event-Depth Fusion

arXiv: https://arxiv.org/abs/2603.27181 · HTML: https://mlsanity.com/paper/2603.27181/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 24 references that are in the map.
- First posted on arXiv 28 Mar 2026.

## Builds on (most cited first, 9 of 9)

- [Event-based Vision: A Survey](https://mlsanity.com/paper/1904.08405/) (2019)
- [Video to Events: Recycling Video Datasets for Event Cameras](https://mlsanity.com/paper/1912.03095/) (2019)
- [Learning High-Speed Flight in the Wild](https://mlsanity.com/paper/2110.05113/) (2021)
- [Fully neuromorphic vision and control for autonomous drone flight](https://mlsanity.com/paper/2303.08778/) (2023)
- [FAST-Dynamic-Vision: Detection and Tracking Dynamic Objects with Event and Depth Sensing](https://mlsanity.com/paper/2103.05903/) (2021)
- [Vision Transformers for End-to-End Vision-Based Quadrotor Obstacle Avoidance](https://mlsanity.com/paper/2405.10391/) (2024)
- [Monocular Event-Based Vision for Obstacle Avoidance with a Quadrotor](https://mlsanity.com/paper/2411.03303/) (2024)
- [Deblur e-NeRF: NeRF from Motion-Blurred Events under High-speed or Low-light Conditions](https://mlsanity.com/paper/2409.17988/) (2024)
- [Multi-UAV trajectory planning for 3D visual inspection of complex structures](https://mlsanity.com/paper/2204.10070/) (2022)

## Abstract (arXiv)

Achieving safe, high-speed autonomous flight in complex environments with static, dynamic, or mixed obstacles remains challenging, as a single perception modality is incomplete. Depth cameras are effective for static objects but suffer from motion blur at high speeds. Conversely, event cameras excel at capturing rapid motion but struggle to perceive static scenes. To exploit the complementary strengths of both sensors, we propose an end-to-end flight control network that achieves feature-level fusion of depth images and event data through a bidirectional crossattention module. The end-to-end network is trained via imitation learning, which relies on high-quality supervision. Building on this insight, we design an efficient expert planner using Spherical Principal Search (SPS). This planner reduces computational complexity from O(n2) to O(n) while generating smoother trajectories, achieving over 80% success rate at 17m/s--nearly 20% higher than traditional planners. Simulation experiments show that our method attains a 70-80% success rate at 17 m/s across varied scenes, surpassing single-modality and unidirectional fusion models by 10-20%. These results demonstrate that bidirectional fusion effectively integrates event and depth information, enabling more reliable obstacle avoidance in complex environments with both static and dynamic objects.