Les commentaires sont clos.

Two-year Research fellow (postdoc)

4 Mai 2023

Catégorie : Post-doctorant


Two-year Research fellow (postdoc): call for application


Localization integrity and Uncertainty propagation in multi-robot systems

1-Context and job description

This two-year position is part of the ANR-JCJC project ToICar (Towards Collaborative Integrity for improved autonomy of a multi-robot system). The context of this project is the navigation of autonomous multi-robot systems. These are robots that can localize themselves in their environment, perceive it, interpret it without any human interaction, interact with other robots by exchanging relevant information and that make safe decisions to achieve their mission. However, so far there is no solution able to operate autonomously in different environments without human intervention. The integrity problem leading to the safety of mobile robots hinders their deployment. The notion of integrity is very important when autonomous navigation presents high safety risks where the probability of accident should be reduced as much as possible as in the case of autonomous vehicles [1]. To do so, navigation requires an accurate and consistent estimation of the pose with reliable inputs from the perception system. However, dealing with this problem from a single robot perspective requires increased redundancy of high cost sensors at each perception and decision level. Hence, collaboration between robots appears to be an interesting approach that can lead to an accurate and high integrity solution thanks to the numerous and redundant information that can be shared [2].

The goal of this two-year research fellowis to advance the state-of-the-art by contributing to the design,development and testing of innovative algorithms in the fields of perception and localization for safe autonomous navigation of multi-robot system.The challenges on which the applicant will focus are the following:

- Multi-sensors data fusion and integrity study of multi-robot system.

- - The link between the integrity of the state estimation and the perception part [3].

- - Formulation of the uncertainty quantification problem in multi-robot systems.

- - Adaptation of the state estimation method and development of fault identification in multi-robot systems (collaboration with a PhD student).

-Implementation and tests on real multi-vehicles datasets.

The research fellow will collaborate with a PhD student working on the same project. The PhD is working on the development of an accurate and consistent state estimation that takes advantage of parameter learning. The state estimation method should be adapted for the collaborative system. Likewise, a decentralized collaborative localization using buildings and Lidar point clouds is developed in another PhD [4]. The research fellow will have access to all resources.

The solutions will be integrated and tested on a case study using three experimental vehicles of the Heudiasyc laboratory. These vehicles are equipped with different sensors: GNSS receivers, wheel speed sensors, Lidars, standard and event-based cameras. High-definition maps are also available.

References (more references on the subject can be given on request)

[1]J. Al Hage, P. Xu, P. Bonnifait, and J. Ibanez-Guzman, “Localization Integrity for Intelligent Vehicles Through Fault Detection and Position Error Characterization,” IEEE Transactions on Intelligent Transportation Systems, 2020.

[2]S. Schön et al., “Integrity and Collaboration in Dynamic Sensor Networks,” Sensors, vol. 18, no. 7, Art. no. 7, Jul. 2018.

[3]H. Chen, Y. Huang, W. Tian, Z. Gao, and L. Xiong, “MonoRUn: Monocular 3D Object Detection by Reconstruction and Uncertainty Propagation,” arXiv, arXiv:2103.12605, Mar. 2021.

[4]M. Escourrou, J. A. Hage, and P. Bonnifait, “Decentralized Collaborative Localization with Map Update using Schmidt-Kalman Filter,” in 2022 25th International Conference on Information Fusion (FUSION), Jul. 2022, pp. 1–8. (another publication with real experiment is under review)


2-Candidates’ profile

Candidates must have a solid background in one or several of the following research fields: state estimation, multi-sensor data fusion, localization, robotic perception and deep learning.

Likewise, they must be able to:

- manage their own academic research and students (e.g. master internship).

- communicate in English and write for publications.

The candidates should have good programming skills in Python or C++. Knowing ROS is a plus.

Scientific curiosity, autonomy, rigor and ability to communicate and collaborate are also expected.

A PhD degree in robotics, automation or computer science (or a related field) is required.



Place: Laboratoire HeudiasycUMR CNRS 7253

Université de Technologie de Compiègne


Web site:

The researcher will be part of SIVALab (joint laboratory between Renault/Heudiasyc/UTC) and will benefit from the means of this joint laboratory.

Employer: CNRS

Duration: 2 years

Salary: ~2500€/month (after taxes) depending on the experience

Application: CV with publications record and cover letter to be sent to:

Joelle Al Hage: