Neuro-muscular state estimation and control for physical human-robot interaction

Understanding the human agent in physical human-robot interaction (pHRI) is imperative. Nevertheless, design and control of such systems favours the robot; potentially resulting in low acceptance of pHRI technology. A contributing factor to this is complexity in understanding a human agent’s intent and physiological state i.e. mental and muscle activity. 

Dyspraxia/Developmental Coordination Disorder (DCD) is a common but under-recognised neurodevelopmental condition affecting approximatively 5% of the population. It impairs the development of motor control and coordination and persists into adulthood. This leads to struggles with basic fine motor skills, such as tying shoelaces and handwriting; and with gross motor skills including learning to ride a bike, drive, or engaging in physical activity. Dyspraxia remains under-researched, with an urgent need to enhance understanding, improve early identification, and develop effective support strategies for Dyspraxic individuals. 

This project is dual-faceted. The HRX-1 robot (https://www.humanrobotix.co.uk/) will be utilised in a fundamental tracking task whereby muscle and neural activity is monitored. Involving normative and Dyspraxic populations as participants, the project will investigate neuro-muscular state and human intent estimation from a fundamental perspective. Concurrently, novel robot-assisted motor-coordination assessment methods with clinical relevance will be developed. The scope of the PhD project is:

• Development of an experimental setup fusing a human-robot interface, virtual reality and EMG/EEG sensing;
• Signal processing/analysis of muscle/neural activity data obtained from user studies;
• Exploration of data-driven modelling techniques (Koopman operator/ergodic theory) to develop a neuro-muscular and human intent state estimator;
• Embed the estimator into the pHRI loop and assess efficacy in providing adaptive control assistance in human-robot scenarios; and
• Development of assessment methods for Dyspraxia identification and support strategies using pHRI. 

Overall, the project outcomes will promote the use of robotics to benefit everyday human life.

This project involves direct collaboration with Human RobotiX and King’s College London. 

Open to any UK or international candidates. Up to 30% of our UKRI funded studentships can be awarded to candidates paying international rate fees.

We welcome candidates with the following background:

• A strong undergraduate degree in Mechatronic/Robotic/Biomedical/Mechanical/Control Engineering, Human Movement, Neuroscience or Mathematical Sciences. Further postgraduate studies in the aforementioned areas will be favourably received;
• Significant knowledge (or willingness to learn) in as many of the following as possible: control theory/state estimation, dynamical systems modelling, linear algebra, calculus, optimisation, data-driven techniques, signal processing, biomechanics and neuromechanics;
• Strong programming skills with experience in at least Matlab/Simulink or Python. Prior experience with developing virtual scenes and/or with ROS2 is a bonus;
• Previous experience and familiarity in conducting user studies involving EMG and/or EEG;
• Strong interpersonal skills. A significant portion of this project involves conducting user studies with neurodivergent participants; hence, the desired candidate must demonstrate the ability to effectively communicate and work sensitively with special populations;
• Willingness to learn and administer standardised motor assessments (e.g. M-ABC);
• Experience with data collection and participant recruitment would be appreciated;
• Self-motivated and able to work independently to find innovative solutions; and
• Teamworking skills to work effectively with supervisors, colleagues, and interdisciplinary teams.