In this project, a portable cost-efficient active ankle orthosis is designed and controlled through adaptive intelligent algorithms.

The system serves as the first module of the Pediatric Exoskeleton Robot for Lower-limb (PERL) which aims at assisting and facilitating the rehabilitation of children with cerebral palsy.

Modern passive ankle-foot exoskeletons do not exhibit appropriate biomechanics during walking, and are unable to adjust their mechanics for different phases of gait and for different environments. In this project, we introduce FLORA - FLexible Orthoses for Rehabilitation of Ankle, which is a semi-active ankle-foot orthoses with customizable stiffness.

The study of motor adaptation is key to a better understanding of mechanisms underlying motor learning. In collaboration with the Motion Analysis Lab at Harvard University, the ExoRoboWalker - a 6-degree-of-freedom lower-limb overground exoskeleton - was used to test the human motor adaptation during gait. The robot is programmed to generate random perturbation with different angular orientations at the ankle.

The main objective of this project is to develop a system identification algorithm to characterize the joint contribution of sensory modalities to human postural control through experimentation, data analysis, model development and simulations.

We use an experimental apparatus which delivers proprioception and visual perturbations using pedals and a virtual reality (VR) system, while recording responses in terms of kinematic and dynamic data as well as muscle activations (EMGs).