ROS Resources: Documentation | Support | Discussion Forum | Service Status | Q&A answers.ros.org

Post-doc position in the domain of lifelong, 3D mapping of populated indoor environments

A 1.5 year post-doctoral position is immediately available in the domain of lifelong, 3D mapping of populated indoor environments.

– Context –

The position is available in the context of a bilateral collaboration program between IMT Atlantique, Brest (https://www.imt-atlantique.fr/en) and ENSIBS Lorient, France. The objective of the project is to develop multi-level map representations of habitable environments, capturing 3D metric and semantic properties of the static as well as the dynamic scene (cf. life-long mapping domain). The project is in line with earlier work related to activity recognition, object, scene and social-interactions mapping (see [1-4] for indicative examples), applied to personal assistance provided by intelligent robots and smart environments.

– Technical skills –
Experience in the following technologies/tools: Robot Operating System (ROS), Gazebosim

For further information related to the project and for submitting your candidature, you may contact:

Panagiotis PAPADAKIS, Ass. prof.


e-mail: panagiotis.papadakis@imt-atlantique.fr

Christophe LOHR, Ass. prof.
e-mail: christophe.lohr@imt-atlantique.fr

– References –

[1]. M. Devanne, P. Papadakis, S. M. Nguyen, Recognition of Activities of Daily Living via Hierarchical Long-Short Term Memory Networks, IEEE International Conference on Systems, Man and Cybernetics (SMC), 2019
[2]. P. Papadakis and P. Rives, Binding Human Spatial Interactions with Mapping for Enhanced Mobility in Dynamic Environments, Autonomous Robots, 41, 5, 2017, pp 1047-1059
[3]. P. Papadakis, A Use-Case Study on Multi-View Hypothesis Fusion for 3D Object Classification, Multi-View Relationships in 3D Data workshop, IEEE International Conference on Computer Vision (ICCV), 2017
[4]. P. Papadakis, A. Spalanzani, C. Laugier, Social Mapping of Human-Populated Environments by Implicit Function Learning, IEEE International Conference on Intelligent Robots and Systems (IROS), Tokyo, Japan, 2013