# Learning to Fly by Crashing

arXiv: https://arxiv.org/abs/1704.05588 · HTML: https://mlsanity.com/paper/1704.05588/

## Key facts

- Cited by 41 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Apr 2017.

## Built on by (most cited first, 30 of 41)

- [LM-Nav: Robotic Navigation with Large Pre-Trained Models of Language, Vision, and Action](https://mlsanity.com/paper/2207.04429/) (2022)
- [How to Train Your Robot with Deep Reinforcement Learning; Lessons We've Learned](https://mlsanity.com/paper/2102.02915/) (2021)
- [Imitation Is Not Enough: Robustifying Imitation with Reinforcement Learning for Challenging Driving Scenarios](https://mlsanity.com/paper/2212.11419/) (2022)
- [Learning High-Speed Flight in the Wild](https://mlsanity.com/paper/2110.05113/) (2021)
- [Holo-Dex: Teaching Dexterity with Immersive Mixed Reality](https://mlsanity.com/paper/2210.06463/) (2022)
- [Semantics for Robotic Mapping, Perception and Interaction: A Survey](https://mlsanity.com/paper/2101.00443/) (2021)
- [Learning to See before Learning to Act: Visual Pre-training for Manipulation](https://mlsanity.com/paper/2107.00646/) (2021)
- [SEAL: Self-supervised Embodied Active Learning using Exploration and 3D Consistency](https://mlsanity.com/paper/2112.01001/) (2021)
- [How to Certify Machine Learning Based Safety-critical Systems? A Systematic Literature Review](https://mlsanity.com/paper/2107.12045/) (2021)
- [NavRL: Learning Safe Flight in Dynamic Environments](https://mlsanity.com/paper/2409.15634/) (2024)
- [Faults in Deep Reinforcement Learning Programs: A Taxonomy and A Detection Approach](https://mlsanity.com/paper/2101.00135/) (2021)
- [Learned Visual Navigation for Under-Canopy Agricultural Robots](https://mlsanity.com/paper/2107.02792/) (2021)
- [Structure from Silence: Learning Scene Structure from Ambient Sound](https://mlsanity.com/paper/2111.05846/) (2021)
- [SOUS VIDE: Cooking Visual Drone Navigation Policies in a Gaussian Splatting Vacuum](https://mlsanity.com/paper/2412.16346/) (2024)
- [Reinforcement Learning for Collision-free Flight Exploiting Deep Collision Encoding](https://mlsanity.com/paper/2402.03947/) (2024)
- [MAVRL: Learn to Fly in Cluttered Environments with Varying Speed](https://mlsanity.com/paper/2402.08381/) (2024)
- [Deep reinforcement learning for guidewire navigation in coronary artery phantom](https://mlsanity.com/paper/2110.01840/) (2021)
- [New Challenges in Reinforcement Learning: A Survey of Security and Privacy](https://mlsanity.com/paper/2301.00188/) (2022)
- [Self-Supervised Motion Retargeting with Safety Guarantee](https://mlsanity.com/paper/2103.06447/) (2021)
- [RAPID: Robust and Agile Planner Using Inverse Reinforcement Learning for Vision-Based Drone Navigation](https://mlsanity.com/paper/2502.02054/) (2025)
- [Robust Navigation for Racing Drones based on Imitation Learning and Modularization](https://mlsanity.com/paper/2105.12923/) (2021)
- [Mutation Testing of Deep Reinforcement Learning Based on Real Faults](https://mlsanity.com/paper/2301.05651/) (2023)
- [BatMobility: Towards Flying Without Seeing for Autonomous Drones](https://mlsanity.com/paper/2307.11518/) (2023)
- [On-device Self-supervised Learning of Visual Perception Tasks aboard Hardware-limited Nano-quadrotors](https://mlsanity.com/paper/2403.04071/) (2024)
- [Training on the Fly: On-device Self-supervised Learning aboard Nano-drones within 20 mW](https://mlsanity.com/paper/2408.03168/) (2024)
- [SAFER: Safe Collision Avoidance using Focused and Efficient Trajectory Search with Reinforcement Learning](https://mlsanity.com/paper/2209.11789/) (2022)
- [A3D: Adaptive, Accurate, and Autonomous Navigation for Edge-Assisted Drones](https://mlsanity.com/paper/2307.09880/) (2023)
- [Fly360: Omnidirectional Obstacle Avoidance within Drone View](https://mlsanity.com/paper/2603.06573/) (2026)
- [Deep Reinforcement Learning for Adaptive Exploration of Unknown Environments](https://mlsanity.com/paper/2105.01606/) (2021)
- [Channel-Aware Distillation Transformer for Depth Estimation on Nano Drones](https://mlsanity.com/paper/2303.10386/) (2023)

## Abstract (arXiv)

How do you learn to navigate an Unmanned Aerial Vehicle (UAV) and avoid obstacles? One approach is to use a small dataset collected by human experts: however, high capacity learning algorithms tend to overfit when trained with little data. An alternative is to use simulation. But the gap between simulation and real world remains large especially for perception problems. The reason most research avoids using large-scale real data is the fear of crashes! In this paper, we propose to bite the bullet and collect a dataset of crashes itself! We build a drone whose sole purpose is to crash into objects: it samples naive trajectories and crashes into random objects. We crash our drone 11,500 times to create one of the biggest UAV crash dataset. This dataset captures the different ways in which a UAV can crash. We use all this negative flying data in conjunction with positive data sampled from the same trajectories to learn a simple yet powerful policy for UAV navigation. We show that this simple self-supervised model is quite effective in navigating the UAV even in extremely cluttered environments with dynamic obstacles including humans. For supplementary video see: https://youtu.be/u151hJaGKUo