Humans are unique in our ability to create and use advanced tools. Today's exoskeletons and rehabilitation robots are very advanced tools, which require the real-time computer control of systems of interconnected components from mechanical, electrical, and other physical domains. Human dynamics and motor control are even more complex; understanding the dynamic synergies between humans and robots is a long-term goal of our research.

To tackle this challenge, we are developing unified computer models of biomechatronic systems that include both the neuro-musculoskeletal dynamics of the human and the physics of assistive and rehabilitative robots.  These integrated models are used to create model-based controllers for the devices, with the aim of optimizing the human performance and the robot design.

In this talk, I will discuss the application of unified system models, nonlinear model-predictive control (NMPC), and deep machine learning to the predictive dynamic simulation and control of biomechatronic systems, including rehabilitation, sports, and wearable robots.

Professor John McPhee is the Canada Research Chair in Biomechatronic System Dynamics at the University of Waterloo.  His research team (morg.uwaterloo.ca) develops dynamic models, machine learning algorithms, and model-based controllers for human-machine systems including exoskeletons, automobiles, sports, and rehabilitation robots.