Robust Navigation for Racing Drones based on Imitation Learning and Modularization
Abstract
This paper presents a vision-based modularized drone racing navigation system that uses a customized convolutional neural network (CNN) for the perception module to produce high-level navigation commands and then leverages a state-of-the-art planner and controller to generate low-level control commands, thus exploiting the advantages of both data-based and model-based approaches. Unlike the state-of-the-art method which only takes the current camera image as the CNN input, we further add the latest three drone states as part of the inputs. Our method outperforms the state-of-the-art method in various track layouts and offers two switchable navigation behaviors with a single trained network. The CNN-based perception module is trained to imitate an expert policy that automatically generates ground truth navigation commands based on the pre-computed global trajectories. Owing to the extensive randomization and our modified dataset aggregation (DAgger) policy during data collection, our navigation system, which is purely trained in simulation with synthetic textures, successfully operates in environments with randomly-chosen photorealistic textures without further fine-tuning.
Cite
@article{arxiv.2105.12923,
title = {Robust Navigation for Racing Drones based on Imitation Learning and Modularization},
author = {Tianqi Wang and Dong Eui Chang},
journal= {arXiv preprint arXiv:2105.12923},
year = {2021}
}
Comments
Published at the 2021 International Conference on Robotics and Automation (ICRA 2021)