Detection, Estimation and Avoidance of Mobile Objects Using Stereo-Vision and Model Predictive Control
Abstract
We propose a complete loop (detection, estimation, avoidance) for the safe navigation of an autonomous vehicle in presence of dynamical obstacles. For detecting moving objects from stereo images and estimating their positions, two algorithms are proposed. The first one is dense and has a high computational load but is designed to fully exploit GPU processing. The second one is lighter and can run on a standard embedded processor After a step of filtering, the estimated mobile objects are exploited in a model predictive control scheme for collision avoidance while tracking a reference trajectory. Experimental results with the complete loop are reported for a micro-air vehicle and a mobile robot in realistic situations, with everything computed on board.
Cite
Text
Roggeman et al. "Detection, Estimation and Avoidance of Mobile Objects Using Stereo-Vision and Model Predictive Control." IEEE/CVF International Conference on Computer Vision Workshops, 2017. doi:10.1109/ICCVW.2017.245Markdown
[Roggeman et al. "Detection, Estimation and Avoidance of Mobile Objects Using Stereo-Vision and Model Predictive Control." IEEE/CVF International Conference on Computer Vision Workshops, 2017.](https://mlanthology.org/iccvw/2017/roggeman2017iccvw-detection/) doi:10.1109/ICCVW.2017.245BibTeX
@inproceedings{roggeman2017iccvw-detection,
title = {{Detection, Estimation and Avoidance of Mobile Objects Using Stereo-Vision and Model Predictive Control}},
author = {Roggeman, Helene and Marzat, Julien and Derome, Maxime},
booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
year = {2017},
pages = {2090-2099},
doi = {10.1109/ICCVW.2017.245},
url = {https://mlanthology.org/iccvw/2017/roggeman2017iccvw-detection/}
}