Learning cognitive models for assistive wearable robots
SECOND CALL
APPLICATION ID: ALL3
What we are looking for:
Proposals oriented at pushing the state of the art in combining variable impedance control with comfortable sensing systems and/or muscle electrostimulation in lower limb exoskeletons for people with mobility impairments.
The context:
Contemporary lower limb exoskeletons often fall short in their adaptability to diverse walking environments, hindering their practicality and overall effectiveness. Variable Impedance Control (VIC) approaches are likely to provide users with a seamlessly responsive experience, allowing for more natural and adaptable movement. VIC, coupled with reinforcement learning and adaptation techniques will enable end-users to walk more naturally across various scenarios, accounting for factors such as walking pace or slope.
The problem to address:
Adrià Colome (IRII, CSIC-UPC) and Juan Moreno (CAR, CSIC-UPM) are joining efforts to address the use of variable impedance approaches and adaptive controllers in terms of adaptability for wearable robotic exoskeletons.
Objectives:
- Innovative solutions for monitoring muscular fatigue during fusion of robot actuation with electrostimulation.
- Combination of more classical sensory inputs (motion, joint forces, electromyography, etc) with such new monitored indicator.
Qualifications:
- Required profile: Control Engineering or Robotics
- Desirable skills/interests: Machine learning – HW/SW systems integration – electrical stimulation – impedance control
Expected Outcomes:
- Contribution to adoption and transformative impact of wearable technologies on mobility assistance and rehabilitation.
- Qualitative Improvements on the user’s perception of using the new technologies versus prior approaches.
