En poursuivant votre navigation sur ce site, vous acceptez le dépôt de cookies dans votre navigateur. (En savoir plus)

PhD (H/F) on multi-robot localization and navigation for infrastructure monitoring

This offer is available in the following languages:
Français - Anglais

Date Limite Candidature : lundi 23 mai 2022

Assurez-vous que votre profil candidat soit correctement renseigné avant de postuler. Les informations de votre profil complètent celles associées à chaque candidature. Afin d’augmenter votre visibilité sur notre Portail Emploi et ainsi permettre aux recruteurs de consulter votre profil candidat, vous avez la possibilité de déposer votre CV dans notre CVThèque en un clic !

General information

Reference : UMR6004-ISAFAN-002
Workplace : NANTES
Date of publication : Monday, May 2, 2022
Scientific Responsible name : Isabelle FANTONI
Type of Contract : PhD Student contract / Thesis offer
Contract Period : 36 months
Start date of the thesis : 1 October 2022
Proportion of work : Full time
Remuneration : 2 135,00 € gross monthly

Description of the thesis topic

The main objective of the PhD thesis is the multi-robot collaboration to achieve an infrastructure monitoring mission using the shareable map of the environment around the infrastructure, obtained by different sensors embedded on multiple robots. Efficiently deploy a few drones and mobile robots for an infrastructure monitoring mission remains an open problem, especially in outdoor and often constrained environments (semi-urban, with obstacles, ...). The objectives will be to propose multi-sensors localization and navigation strategies with at least a couple of drones and mobile robots to efficiently monitor an infrastructure. The robots will be controlled and located to predefined points of interest using visual or sensor-based servoing. One main objective is the accurate positioning of robots relative to the environment, the observed infrastructure and between them. The thesis aims to answer the following questions: how to choose appropriate sensors (cameras, UWB?...) embedded on robots to localize them in presence or absence of GPS ? How to fuse them in order to optimize the positions of drones and robots to accurately monitor the infrastructure ? Which strategies to adopt when we loose the localization ? How the robots share their positions ?
We will consider two kind of scenarios in which two drones and one mobile robot will be continuously operating during the mission. In the first kind of scenario called “Tight collaboration”, the two drones will assist the vehicle in localization for navigation in the following GPS denied situations (for the vehicle) :
• The ground robot is surrounded by vegetation but some clear spots allow the ground robot to “see” the drones
• The ground robot is in a tunnel (completely surrounded by trees) without clear spots but it can “see” the drones at the beginning and end of tunnel
• The ground robot is too close to high buildings
In the second kind of scenario, called “Loose collaboration” all the robots collaborate to create and update a map of the environment with its own sensors. We will explore different configurations such as:
• each drone has a camera, the vehicle has lidar and cameras
• one drone has a camera, one drone has a lidar, the vehicle has lidar and cameras
In all the above scenario we will suppose that the drones will be able to receive the GPS and communicate with vehicles for localization. We plan to consider strategies using couple geo-localization and relative localization between drones and the robot (e.g. via Ultra-wideband (UWB) measurements) to make vehicle localization more robust. A recent survey [1] has shown that UWB-based localization focused on multi-UAV systems and heterogeneous multi-robot systems has emerged in recent years as high-accuracy solutions for collaborative missions. Combinations of UWB ranging measurement with GPS can also provide higher-precision [2]. The objective of the work will also be to control the robots to compensate potential residual uncertainties.
In order to increase the autonomy of the multi-robot system, we will suppose that two other drones are available on the ground robot charging their batteries when the first couple is on air. When the batteries of the first couple of drones need to be recharged, they will be replaced by the second couple. Coordinating the drone replacement while ensuring the continuity of service needs to define a “high-level” strategy and optimal task allocation. In order to define which task to which robot could be assigned, we will explore different “high-level” strategies, for example task allocation algorithms [5], described as optimization problems in multi-robot systems. Many approaches exist to assign tasks to robots in a multi-robot environment in both a centralized and distributed fashion. Auction-based approaches could offer an attractive solution due to their flexibility and ability to distribute solution computation to different agents, while a global objective function is defined to meet certain system performance requirements [6], [7].

Proposed work plan:
- Making a literature review of different strategies for the localization and allocation/navigation strategies using multi-sensors for multiple robots (drones and mobile robot).
- Proposition of different methods for efficiently optimize the localization and navigation of robots during the missions.
- Defining a simulation environment to evaluate a collaborative mission using Gazebo and ROS.
- Study and implementation of different strategies in simulations.
- Experiments in flight arenas (indoor and outdoor).

[1]. W. Shule, C. Martinez Almansa, J. Pena Queralta, Z. Zou, T. Westerlund, “UWB-Based Localization for Multi-UAV Systems and Collaborative Heterogeneous Multi-Robot Systems”, International Conference on Future Networks and Communications (FNC), 2020.
[2]. Y. C. Chen, A. Lai, R. Wu, “UWB-assisted high-precision positioning in a UTM prototype”. IEEE Topical Conference on Wireless Sensors and Sensor Networks (WiSNeT), pp. 42-45, 2020.
[3]. K. Guo, Z. Qiu, C. Miao, A. H. Zaini, C.-L. Chen, W. Meng, L. Xie, “Ultra-Wideband-Based Localization for Quadcopter Navigation”. Unmanned Systems, Vol. 4, No. 1, pp. 23–34, 2016.
[4]. K. Guo, X. Liu, L. Xie, “Ultra-Wideband and Odometry-Based Cooperative Relative Localization with Application to Multi-UAV Formation Control”. IEEE Transactions on Cybernetics, Vol. 50, Issue 6, pp. 250-2603, June 2020.
[5]. A. Khamis et al: “Multi-robot Task Allocation: A Review of the State-of-the-Art”, Cooperative Robots and Sensor Networks, 2015.
[6]. G. Lozenguez, L. Adouane, A. Beynier, A. Mouaddib, P. Martinet: ”Punctual versus continuous auction coordination for multi-robot and multi-task topological navigation”, Autonomous Robots, 40 (4), 599-613, 2016.
[7]. M. Dias et al. : “Market-based multirobot coordination: a survey and analysis”, Proceedings of the IEEE,94(7), 1257–1270, 2006.

Work Context

This thesis project will be done at LS2N in Nantes, with a co-supervision between Isabelle FANTONI at LS2N, Nantes, in ARMEN team, Ezio MALIS and Philippe Martinet at the Centre INRIA, Université Côte d'Azur, in ACENTAURI team in Sophia Antipolis.

The proposed subject is proposed in the context of the SAMURAI project funded by ANR whose partners are INRIA (ACENTAURI team), LS2N (ARMEN team) and the MIS. The ambition of the SAMURAI project is to design new approaches for the long term navigation of a multi-robot system collaborating on a common monitoring tasks, in a urban or peri-urban environment using heterogeneous sensors in order to facilitate their implementation (reduction of preparation time and costs). By monitoring task, we intend to collect accurate data in a specific area and for a specific data processing (which depends on the targeted application, for instance defects detection and following the evolution). The scientific objectives of the project are: (i) to build shareable maps of a complex environment using high-end heterogeneous sensors (lidar, vision, IMU, GPS, ...); (ii) to utilize the map to perform long term infrastructure monitoring using collaborative robots having low-end sensors different from the high-end sensors used to build the shareable map; (iii) to update the map when changes are detected using the data collected by the robots with limited sensor capability during their monitoring task. The developed approaches will be validated experimentally on a scenario concerning the monitoring of a building in an urban or peri-urban environment (e.g. a church or an historical building) and the update of the shareable map using ground and aerial robots.

Additional Information

Required degree and skills
- Master and/or engineer diploma in Robotics.
- Knowledge of programming languages (Python, C++). Knowledge of the ROS2 environment would be appreciated.
- Good English language skill, oral and written.
- Analytical skills, ability to formulate and conduct a scientific project, ability to communicate and valorize work.
- Ability to work in a team on collaborative projects.

Documents to be provided :
- Curriculum Vitae
- Motivation letter

We talk about it on Twitter!