In this project, our main goal is to develop a robotic platform for upper limb rehabilitation of post-stroke patients by incorporating a robotic manipulator and an intelligent control system. This robotic rehabilitation system will be a multipurpose and cost-effective platform while being adaptive to various therapeutic practices.

The objective of the project is the development of the Haptic3D, an interactive robot with 3 degrees of freedom (DOF), for the rehabilitation of movements of the upper limb. The haptic capabilities are added to this device through adaptive impedance/admittance-based control. The system is expandable to incorporate more sensing and actuation modalities, and its integration with VR technology.

In this project, we want to take a closer look at the adaptation of motor learning when an environmental disturbance occurs, particularly when the individual is subjected to a force field applied by a 6-DOF robot. We are interested to learn how the cognitive-motor processes evolve in an environment where the force field evolves. 

This work uses a continuous actor-critic Reinforcement Learning method for optimizing the control of an assistive robot using myoelectric interface. The RL agent uses a sparse human-delivered training signal, without requiring detailed knowledge about the task domain. This reinforcement-based machine learning framework is well suited for use by both patients and clinical staff, and may be easily adapted to different application domains and the needs of individual amputees. 

HapticMaster is a device used especially for the rehabilitation of the upper limb in patients with neuromotor deficiencies. This device can only control the hand translations, but not the wrist rotations. We have developed a 3 DOF mechanism that can be used with HapticMaster or any other arm/hand rehabilitation devices to enable the wrist/finger rotations. 

In this project, a customizable EMG acquisition sleeve has been developed using four EMG sensors to detect upper-limb muscle activation, filter them and record them in real-time to be used for robotic systems.