Versatile Bio-inspired Edge-computing Devices

In recent years, thanks to significant advancements in developing bio-electronic systems, monitoring the relevant features related to human motion (e.g., skeletal muscle activity, body kinematics) is opening novel possibilities in assistive and rehabilitation technologies.
In this context, the core of this research activity is to design and develop low-power intelligent wearable devices to assess and process body movements during daily-life actions. Indeed, the sensors’ fusion of surface ElectroMyoGraphy (sEMG) and Inertial Tracking (IMU) data implemented on edge-computing modules, combined with bio-inspired processing techniques (e.g., the event-driven Average Threshold Crossing – ATC), allows the realization of custom multi-channel systems featuring a robust detection of human activities with minimal energy requirements, thus extending their functional operating time up to weeks.
Starting from these systems, different biomedical applications associated with human movement are possible. On one side, the computing capabilities of the edge nodes can be involved in body gesture recognition, constituting a human-machine interface able to control different actuators and capable of being embedded into game-rehabilitation scenarios. On the other hand, the low power consumption and small design of the devices allow them to be worn under normal clothing, thus spreading their involvement in rehabilitation and long-term monitoring scenarios, both in a lab and outdoor environments