Faites connaître cette offre !
Reference : UMI2958-NADWER-024
Workplace : METZ
Date of publication : Monday, September 07, 2020
Scientific Responsible name : CEDRIC PRADALIER
Type of Contract : PhD Student contract / Thesis offer
Contract Period : 36 months
Start date of the thesis : 1 December 2020
Proportion of work : Full time
Remuneration : 2 135,00 € gross monthly
Description of the thesis topic
This thesis subject is to study various exploration strategies in an outdoor natural environment by an unmanned ground vehicle (UGV). Autonomous robotics in this kind of environment is not much studied, by comparison to autonomous driving, focusing mainly on urban environments.
In this type of environment, the challenges are undeniably different. Firstly, they are related to navigation (how to drive from A to B without an available road or detailed map). Secondly, they are related to localization (intermittent GPS signal, issues in the perception: sun glare for the camera, semi-transparent trees for the laser, aspect variation of the environment due to seasonal changes, weather conditions…).
Besides, outdoor natural environment suffer from the lack of labeled data for machine learning techniques. The objective of this thesis is to contribute to the research in this area, by developing policies for exploration and coverage to allow an UGV to be adaptative, able to integrate in its exploration and coverage policy a geometric and semantic understanding of its surroundings. Indeed, we assume here that those policies are different in an outdoor natural environment and in a more structured environment such as an urban or indoor one. For example, entering a room and covering it entirely is not done in the same way than going around a sparse grove of trees, nor even a dense bush.
This thesis will be done in the DREAM laboratory of the UMI 2958 GT-CNRS
Constraints and risks
Field work (outdoor experiments, with any weather condition)
We talk about it on Twitter!