SYM-17: Rehabilitation Robotics


Viet Anh Dung Cai1, Philippe Bidaud2, Viet Thang Nguyen1, Consuelo Granata2, Aurélien Ibanez2, Minh Tam Nguyen1

1Ho Chi Minh City University of Technology and Education, Vietnam;
2Institut des Systèmes Intelligents et de Robotique, Université Pierre et Marie Curie, France

The authors aim to present a robust gait phase detection method which is developed for the control of a multi-contacts lower limb exoskeleton. The device comprises passive mechanical linkages connecting the user limbs to an external rigid mechanical structure. This approach allows not only a more effective control of the system’s transparency but also provides the possibility to use different kinds of sensors to capture the user’s kinematic data, which can be used to detect the gait phases. For this purpose, a Principal Component Analysis (PCA) is applied to each measure. The resulting vector of the principal components is then compared to the reference ones in order to identify the actual gait phase using K-Nearest Neighbors algorithm. A Discrete Time Markov Chain (DTMC) is also used to define the phases shift probability during the gait cycle. This gait detection algorithm was tested experimentally with a percentage of success of more than 95% for repetitive gait cycles. As a result, these principles can now be used to program knee assistance exercises. To this end, a hybrid predictive control, which relies on a robust locomotion phases detection algorithm, will be implemented on the exoskeleton device, providing torque assistance precisely at specific gait phases.

Organised by

Endorsed by


        Supported by





        Supporting Media