sujet2020-exoskeleton

Dynamic simulation of human motions with exoskeleton assistance

Type: Internship (Master level)
Advising team: Pauline Maurice and Serena Ivaldi

General Information

Advisors Pauline Maurice Serena Ivaldi
Address LORIA/INRIA, Nancy, France LORIA/INRIA, Nancy, France
Phone 03 54 95 85 32
Email pauline.maurice@loria.fr serena.ivaldi@inria.fr
Office C122 C104

Background

Exoskeletons are currently receiving a lot of attention, because of their potential to assist people during strenuous tasks [1]. By providing assistive torques and/or structural support, exoskeletons can reduce the users’ physical load, and thereby decrease the risk of developing long term occupational injuries [2]. An exoskeleton is, however, a wearable device that is tightly coupled to the human body: its design and control are therefore of utmost importance for the comfort and efficacy. A difficulty is that even when exoskeletons are designed to assist one specific task or movement, they also need not to perturb peripheral motions. The design of exoskeletons therefore remains an iterative process, including users test and subsequent adjustments.

A possibility to shorten the design process is to leverage the power of digital human simulation to evaluate the device in a digital world, thus suppressing the need for a costly physical mock-up. Specifically, dynamic simulation, which relies on a physics engine, is a promising tool since it enables to estimate both the contact forces (human-exoskeleton) and internal forces (proxy for human effort) during a motion [3]. Simulations have recently been proposed to compare the benefit of different exoskeletal assistances [4,5], but so far, they rely on a simplified version of either the exoskeleton or the human. The goal of this internship is to develop a framework for simulating 3D dynamic motions of a digital human model coupled to a digital mock-up of an exoskeleton, in order to evaluate the effects of the exoskeleton on the human body.

Objectives and Workplan

In the Larsen team of LORIA-INRIA Nancy, we develop a simulation framework which includes a digital human model that can be animated using a quadratic programming control technique (QP controller) [6]. Building on this framework, the goal of the internship is to develop a methodology to integrate the simulation of actuated exoskeletons coupled to the digital human model, and to evaluate their effects.

The main steps of the work are:

  • Develop a module which enables to import exoskeleton designs and their control law in the simulation framework. At this stage, high-level descriptions of exoskeletons will be considered (e.g., number, type and location of the degrees of freedom, actuation capabilities, segments dimensions).

  • Implement the simulation of the digital human model coupled to the exoskeleton, for a variety of movements. A database of human motions during industry-like tasks is available and will be used as a reference for the motions to reproduce [7]. Additional data may be collected using the motion capture tools available in the lab.

  • Propose and implement evaluation metrics to estimate the physical effects of the exoskeleton of the human.

  • Analyze the possibilities and limitations of the simulation tool, with respect to the state of the art. Criteria can include the type of exoskeletons and the type of motions that can be successfully simulated, the information that can be measured, etc.

Requirements

Technical skills:

  • Robotics, control theory.
  • Programming: C++, Python.
  • Experience with exoskeletons and/or motion capture is a plus.

Language:

  • English or French

Advising and Organization

The internship is for a duration of 6 months, in the interval between January and December 2020.
The intern will be located in LORIA/INRIA research center in Nancy.
The intern will receive a salary of ~550euros/month plus benefits.

The internship will be co-supervised by:

Application

Applicants should send their CV, motivation letter describing their specific interest for the topic, and their Master’s grades to Pauline Maurice (pauline.maurice@loria.fr).

References

[1] M. P. de Looze et al., “Exoskeletons for industrial application and their potential effects on physical work load,” Ergonomics, vol. 59, no. 5, pp. 671–681, 2016.

[2] A. Parent-Thirion et al., Fifth European Working Conditions Survey, Eurofound, Pub. Office of the European Union, 2012.

[3] P. Maurice et al., “Human-oriented design of collaborative robots”, International Journal of Industrial Ergonomics, Elsevier, vol. 57, pp. 88–102, 2017.

[4] C. L. Dembia et al., “Simulating ideal assistive devices to reduce the metabolic cost of walking with heavy loads”, PloS one, vol. 12, no 7, 2017.

[5] M. Harant et al., “Cost function evaluation for optimizing design and actuation of an active
exoskeleton to ergonomically assist lifting motions”, Proceedings of the IEEE/RAS International Conference on Humanoid Robots (Humanoids), 2019.

[6] W. Gomez et al., “Humanoid Whole-Body Movement Optimization from Retargeted Human Motions”, Proceedings of the IEEE/RAS International Conference on Humanoid Robots (Humanoids), 2019.

[7] P. Maurice et al., “Human movement and ergonomics: An industry-oriented dataset for collaborative robotics”, The International Journal of Robotics Research, vol. 38, no. 14, pp. 1529-1537, 2019.

 

Comments are closed.