Ifsttar PhD subject

French version
Detailed form :

Title : Movement prediction for accessibility analysis

Main host Laboratory - Referent Advisor   -

Director of the main host Laboratory   -

PhD Speciality Biomécanique

Axis of the performance contract 1 - COP2017 - Efficient transport and safe travel

Main location Bron

Doctoral affiliation UNIVERSITE CLAUDE-BERNARD-LYON 1

PhD school MEGA (MECANIQUE, ENERGETIQUE, GENIE CIVIL, ACOUSTIQUE)

Planned PhD supervisor DUMAS Raphaël  -  Université Gustave Eiffel  -  TS2 - LBMC

Planned PhD co-supervisor PRONOST Nicolas  -  Univeristé Claude Bernard Lyon 1  -  LIRIS

Planned financing Contrat doctoral  - Ifsttar

Abstract

Due to the aging of the population, accessibility for the elderly when traveling is a growing issue. When aging, motor and cognitive abilities decline, which increases the risk of falling and therefore of injury and death. Moreover, injury apprehension can lead to a sedentary lifestyle, whereas physical activity is recommended to maintain good health, both physical and mental. One inconvenience to the physical activity of people with reduced mobility is the accessibility of infrastructures in urban areas. Indeed, if they are not adapted, they can become an obstacle to mobility and therefore promote a sedentary lifestyle. The best solution to attest to the accessibility of a system is in-vivo experimentation. However, this solution has certain limitations, whether in terms of the design of the experimental set-up, which cannot easily simulate all the possible configurations, the experimental protocol, which must take into account the tiredness of the participants, and the recruitment of the participants, which may prove to be challenging. An axillary solution to increase the quantity of results or to act as pre-tests to help design the experimental set-up is the use of simulation. Gait simulation is used in different fields such as biomechanics, computer graphics or robotics. Each field has its own objectives and constraints to respect and therefore uses different control methods. The objective of this thesis is to design a movement simulator which must then be able to simulate pathological walking when passing obstacles. To this purpose, a control method generally used in computer graphics and robotics was used to achieve kinematics and dynamics sufficiently biofaithful for use in the field of biomechanics. The contribution of this work and this approach can then be described in three points. First, it explores the feasibility of this method, highlight potentials problems and guide the choices for future work. Secondly, the design of a control adopting a different control from what is usually done in the biomechanics field allows for a new walking study point of view by taking a direct interest in the roles of each joint and the how to control them depending on the situation. Finally, if it turns out that this method is feasible and that a complete simulator is designed, then it will make it possible to carry out simulations combining the advantages of computer graphics and robotics methods, namely the simulation time and intuitiveness of the control, and the advantages of biomechanics, namely the biofidelity of the results.

Keywords : Gait simulation, obstacle passing, pathological gait, direct dynamics, joint kinematics control, control loop, OpenSim

Main host Laboratory - Referent Advisor	-
Director of the main host Laboratory	-
PhD Speciality	Biomécanique
Axis of the performance contract	1 - COP2017 - Efficient transport and safe travel
Main location	Bron
Doctoral affiliation	UNIVERSITE CLAUDE-BERNARD-LYON 1
PhD school	MEGA (MECANIQUE, ENERGETIQUE, GENIE CIVIL, ACOUSTIQUE)
Planned PhD supervisor	DUMAS Raphaël - Université Gustave Eiffel - TS2 - LBMC
Planned PhD co-supervisor	PRONOST Nicolas - Univeristé Claude Bernard Lyon 1 - LIRIS
Planned financing	Contrat doctoral - Ifsttar
Abstract Due to the aging of the population, accessibility for the elderly when traveling is a growing issue. When aging, motor and cognitive abilities decline, which increases the risk of falling and therefore of injury and death. Moreover, injury apprehension can lead to a sedentary lifestyle, whereas physical activity is recommended to maintain good health, both physical and mental. One inconvenience to the physical activity of people with reduced mobility is the accessibility of infrastructures in urban areas. Indeed, if they are not adapted, they can become an obstacle to mobility and therefore promote a sedentary lifestyle. The best solution to attest to the accessibility of a system is in-vivo experimentation. However, this solution has certain limitations, whether in terms of the design of the experimental set-up, which cannot easily simulate all the possible configurations, the experimental protocol, which must take into account the tiredness of the participants, and the recruitment of the participants, which may prove to be challenging. An axillary solution to increase the quantity of results or to act as pre-tests to help design the experimental set-up is the use of simulation. Gait simulation is used in different fields such as biomechanics, computer graphics or robotics. Each field has its own objectives and constraints to respect and therefore uses different control methods. The objective of this thesis is to design a movement simulator which must then be able to simulate pathological walking when passing obstacles. To this purpose, a control method generally used in computer graphics and robotics was used to achieve kinematics and dynamics sufficiently biofaithful for use in the field of biomechanics. The contribution of this work and this approach can then be described in three points. First, it explores the feasibility of this method, highlight potentials problems and guide the choices for future work. Secondly, the design of a control adopting a different control from what is usually done in the biomechanics field allows for a new walking study point of view by taking a direct interest in the roles of each joint and the how to control them depending on the situation. Finally, if it turns out that this method is feasible and that a complete simulator is designed, then it will make it possible to carry out simulations combining the advantages of computer graphics and robotics methods, namely the simulation time and intuitiveness of the control, and the advantages of biomechanics, namely the biofidelity of the results.
Keywords :	Gait simulation, obstacle passing, pathological gait, direct dynamics, joint kinematics control, control loop, OpenSim

List of topics

Applications closed