Synthèse sur la conception, commande et planification de trajectoire d'une interface de locomotion pour la réadaptation de la marche
|Advisor:||Gosselin, Clément; Kövecses, Jozsef|
|Abstract:||This thesis summarizes the design of a locomotion interface for gait rehabilitation. The aim of the mechanism is to alleviate the workload of therapists by automating the repetitive movements involved in the rehabilitation exercises. Moreover, by offering a larger panel of exercises, the locomotion interface should be an asset compared to standard treadmills or rehabilitation walkways. Walking simulation is a challenge in terms of performance, power and safety since the mechanism includes the user in the workspace of the effectors. The balance of the user should be ensured during the swing phase with a reduced human-robot interaction and reliable during the stance phase. First, Chapter 1 describes the walking motion, the stair climbing up and down movement and highlights their main kinematic and dynamic features. Chapter 2 then introduces the architecture of the locomotion interface based on independent belt routings which transmit the movement to two end-effectors that carry the user. Each foot platform has two degrees of freedom (dofs) corresponding to the horizontal and vertical translations in the sagittal plane. Decoupling the dofs simplifies the control of the locomotion interface and increases the efficiency of the torque of the motor sent to the end-effectors compared to systems with co-dependent degrees-of-freedom. Then, the thesis presents the strategies used to supervise the human-robot interaction. The kinematic and dynamic requirements are different during the swing phase and the stance phase of the human gait. Therefore, Chapter 3 introduces the force controllers that lighten the apparent inertia of the mechanism as well as the additional mechanism based on passive cables in order to further alleviate the impedance of the effector. Chapter 4 presents the controller that generates the vertical virtual constraint in order to produce the required reliable floor during the stance phase. The rendering of the virtual environment is improved with the implementation of a static balancing system based on gas springs that alleviates the workload of the motors that handle the weight of the user. Finally, Chapter 5 introduces the cancellation algorithm that generates the infinite environment. Horizontally, the user is brought backward such as on a treadmill. Vertically, the user is moved in the opposite direction of his/her movement such as in a reversed escalator.|
|Document Type:||Thèse de doctorat|
|Open Access Date:||24 April 2018|
|Collection:||Thèses et mémoires|
All documents in CorpusUL are protected by Copyright Act of Canada.