Section: Bilateral Contracts and Grants with Industry

Bilateral Contracts with Industry

BPI-PCR Robo-KII

Participant : Armel Crétual [contact] .

This contract has started in February 2017 and ended in October 2018. In M2S, it involved two permanent members of MimeTic team, Armel Crétual and Franck Multon, and two engineers, Antoine Marin (18 months grant) and Brice Bouvier (10 months grant).

This project was a collaboration between BA Healthcare and M2S lab. It aimed at developing a robotics platform to allow physicians to start gait rehabilitation as soon as possible, even before patients are able to maintain upright posture alone. The usual way to perform such rehab sessions is to make the patient walk on a treadmill benefiting from a harness to prevent patient from falling. The two main limits of this approach are that:

• only straightforward at constant speed gaits are feasible whereas falling risks are much higher when modifying speed or turning

• walking on a treadmill when motor abilities are very affected can be challenging and can generate strong apprehension.

In a previous project, Robo-K, ended in september 2016, BA Healthcare has developed a first prototype of a mobile robot which strongly modified the approach: the harness is mobile and follows the patient displacement. In this way, the patient walks on the ground at his/her desired speed and the physician can include curved trajectories in the rehab process.

The main novelty of Robo-KII project was to implement a biofeeedback system onto the robotics platform to reinforce rehab sessions. Closely working with physicians from two PMR services, CHU Rennes and Kerpape center, we tested several parameters of the feedback to be given to the patients. In particular, in a clinical pre-test, we focused on the temporal aspect, i.e. providing the feedback at each gait cycle or only after one rehab exercise (up to 20 steps) to avoid dual tasks situation as patients in this early stage after stroke usually also suffer from cognitive issues.

Unity - Cinecast

Participants : Marc Christie [contact] , Quentin Galvane.

Cinecast is a research collaboration between Unity and Inria Rennes. The collaboration is focused on automated cinematography and automated editing technologies for creating video casts of 3D game sessions. The project has started in July 2018 for one year, and with a budget of 45kE. The challenge consists in specializing the general editing techniques proposed in our Automated Editing paper (AAAI 2015), reducing the knowledge of the editing algorithm from the full sequence to only 3 seconds. A first demonstration of the results was presented at the Unite 2018 event in Los Angeles.

SolidAnim - Solidtrack

Participants : Marc Christie [contact] , Xi Wang.

In the scope of the ANR project LabCom, the purpose of this research collaboration is to develop SLAM technologies which are robust to changes in the lighting conditions. The collaboration started in October 2018, with a budget of 180kE for a duration of three years. The budget is mostly dedicated to hiring PhD student Xi Wang. The work is a co-supervision with Eric Marchand (from Rainbow team).

Cifre Faurecia - Monitoring de l'efficience gestuelle d'opérateurs sur postes de travail

Participants : Franck Multon [contact] , Georges Dumont, Charles Pontonnier, Olfa Haj Mahmoud.

This Cifre contract has started in September 2018 for three years and is funding the PhD thesis of Olfa Haj Mamhoud. It consists in designing new methods based on depth cameras to monitor the activity of workers in production lines, compute the potential risk of musculoskeletal disorders, and efficiency compared to reference workers. It raises several fundamental questions, such as adapting previous methods to assess the risk of musculoskeletal disorders, as they generally rely on static poses whereas the worker is performing motion. Based on previous works in the team (previous Cifre PhD thesis of Pierre Plantard) we will provide 30Hz motion capture of the worker, that will enable us to evaluate various time-dependent assessment methods.

We will also explore how to estimate joint forces based and torques on such noisy and low-sampling motion data. We will then define a new assessment method based on these forces and torques.

The Cifre contracts funds the PhD salary and 30K€ per year for the supervision and management of the PhD thesis.