Monocular Event-Based Vision for Obstacle Avoidance with a Quadrotor

TL;DR

This paper introduces a novel monocular event-camera-based obstacle avoidance system for quadrotors, leveraging depth prediction and transfer learning to enable autonomous navigation in complex environments.

Contribution

It presents the first static-obstacle avoidance method using only a monocular event camera on quadrotors, combining simulated pre-training with real-world fine-tuning.

Findings

Higher speeds improve depth estimation and obstacle avoidance.

Indoor scenes pose more challenges than outdoor scenes.

Event-based depth prediction is more accurate at higher speeds.

Abstract

We present the first static-obstacle avoidance method for quadrotors using just an onboard, monocular event camera. Quadrotors are capable of fast and agile flight in cluttered environments when piloted manually, but vision-based autonomous flight in unknown environments is difficult in part due to the sensor limitations of traditional onboard cameras. Event cameras, however, promise nearly zero motion blur and high dynamic range, but produce a very large volume of events under significant ego-motion and further lack a continuous-time sensor model in simulation, making direct sim-to-real transfer not possible. By leveraging depth prediction as a pretext task in our learning framework, we can pre-train a reactive obstacle avoidance events-to-control policy with approximated, simulated events and then fine-tune the perception component with limited events-and-depth real-world data to achieve obstacle avoidance in indoor and outdoor settings. We demonstrate this across two quadrotor-event camera platforms in multiple settings and find, contrary to traditional vision-based works, that low speeds (1m/s) make the task harder and more prone to collisions, while high speeds (5m/s) result in better event-based depth estimation and avoidance. We also find that success rates in outdoor scenes can be significantly higher than in certain indoor scenes.