The Inria Université Côte d’Azur center counts 36 research teams as well as 7 support departments. The center's staff (about 500 people including 320 Inria employees) is made up of scientists of different nationalities (250 foreigners of 50 nationalities), engineers, technicians and administrative staff. 1/3 of the staff are civil servants, the others are contractual agents. The majority of the center’s research teams are located in Sophia Antipolis and Nice in the Alpes-Maritimes. Four teams are based in Montpellier and two teams are hosted in Bologna in Italy and Athens. The Center is a founding member of Université Côte d'Azur and partner of the I-site MUSE supported by the University of Montpellier.

Research teams:

ACENTAURI is a robotic team that studies and develop intelligent, autonomous and mobile robots that can help humans in their day-to-day lives at home, at work or during their displacements. The team focuses on perception, decision and control problems for multi-robot collaboration by proposing an original hybrid model-driven / data driven approach to artificial intelligence and by proposing efficient algorithms. The team focuses on robotic applications in smart territories, smart cities and smart factories. In these applications several collaborating robots will help humans by using multi-sensor information eventually coming from infrastructure. The team demonstrates the effectiveness of the proposed approaches on real robotic systems like cars AGVs and UAVs together with industrial partners. Innovation and the transfer of the research work towards industrial partners are a concern of ACENTAURI.

Université de Picardie Jules Verne - MIS (Modélisation, Information \& Systèmes) conducts researches in computer science, control and robotics. The Perception on robotics, which is MIS lab team, has a research focus on digital and heritage. Huge heritage buildings, such as the cathedral of Amiens in France are targeted by this team toward their full digitization for archiving, architects and historians of art analyses, renovation planning, and more. To do so, the team considers  cameras, mainly non-conventional (omnidirectional, multimodal), or laser scanners with robotics, on the ground or in the air. One of this team specialties is to design and use the non-conventional cameras as input of the robot motion control.

This subject is proposed in the context of the SAMURAI project funded by ANR. The ambition of the 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 a 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.

Within this context, the main objective of this PhD will be to design and implement an approach for detecting significant changes in already mapped environment and update the existing map. The environment will be initially mapped with high-end sensors (Lidars, cameras, GPS RTK, ...) and stored in a new Robotics BIM (R-BIM)  that will be developed in the SAMURAI project. Then, the environment will be monitored by multiple collaborative robots that uses the map for localization and navigation. During the monitoring task we will look for significant changes between the real environment and our representation.

Several approaches have been proposed for 3D change detection. The main scientific problem still to be solved is to separate significant changes from measurement errors and periodic variations caused by natural processes that we would like to include in the map. A promising research direction that we would like to investigate is to use Artificial Neural Networks with data coming from different sensor modalities embedded on a robot. Since we already have the complete 3D model of the environment we could simulate changes in the environment to train the neural network (avoiding the acquisition of large scale data). Another problem to be addressed is to find the best strategy to decide for an update. Indeed, updating too soon (i.e. for small changes) will add a drift to the map while updating too late may be a problem to register the new data with the old one. We plan to address the spatio-temporal detection problem by extending the classical spatial models with the time dimension to represent the periodicities of the observed events.

We will investigate two different scenarios. In the first scenario, the shareable map is updated using data acquired at high frequency (e.g. every week) by low-end sensors. The main scientific problem to be solved is how to maintain the coherence of the high-end data and low-end data. Indeed, we cannot update the high-end data in a straightforward manner from low-end data. We plan to address this problem by updating only the low-end data but propagating the uncertainty on the high-end data. Therefore, qualitative information abouts the changes will be propagated if high-end data is used to generate low-end data for a different sensor. In this context, we would like to investigate how to solve our problem using probabilistic data fusion techniques.

In the second scenario we update the shareable map using data acquired at low frequency (e.g. every month) by the high-end sensors. Our objective is not to use the GPS RTK in order to study a more flexible approach for map update. The main scientific problem to be solved is the high accuracy multi-sensory alignment of all data with the existing map in the initial R-BIM. Another important problem to be addressed is to separate (local) constant and dynamic parts of the infrastructure through the trajectory replay(s).

The work will be decomposed with incremental steps as follows:

• Bibliography on multi-sensor based reconstruction and model update
• Choice and setup of a simulation environment
• Design of the algorithm
• Simulation and tuning of the algorithm
• Comparison with the state of the art techniques
• Experimental results on real data
• Writing of reports and conference papers
• Improvement on the algorithm
• Experimental results on real data acquired on SAMURAI robots and environment
• Writing Phd Thesis and journal papers

The candidate is expected to have a Master in Robotics, Computer Vision or Signal Processing, as well as solid skills in software development (LINUX, ROS, Git, MATLAB, C/C ++, Python, OpenCV). He/she must also be highly motivated for multidisciplinary studies and all aspects of research ranging from fundamental to experimental work. A good level of written/spoken english is also important

• Subsidized meals
• Partial reimbursement of public transport costs
• Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
• Possibility of teleworking (after 6 months of employment) and flexible organization of working hours
• Professional equipment available (videoconferencing, loan of computer equipment, etc.)
• Social, cultural and sports events and activities
• Access to vocational training
• Social security coverage

Gross Salary per month: 1982€brut per month (year 1 & 2) and 2085€ brut/month (year 3)