For the sensorimotor system to complete motor tasks it controls the body, it controls objects that the sensorimotor system acts upon within the environment, and it anticipates future states of the environment. The sensorimotor system is known to adapt and improve in performance with practice in response to predictable phenomena. The literature explains motor adaptation and performance improvement in terms of models, called internal models, of future loads. The theory of internal models has been investigated in the neuroscience and human motor behavior communities, where electrophysiological data and motor performance experiments have yielded rich data in support of the role of predictive modeling.

Internal models can be divided into two types: internal models of the plant and internal models of exogenous processes. While internal models of the plant have a rich history and have been studied extensively, literature on internal models of exogenous processes is less developed. This dissertation introduces the Internal Model Principle (IMP) as a tool for modeling internal models of exogenous processes. This dissertation further extends the usefulness of the IMP for modeling human motor control by extending the model to handle sensorimotor tasks that feature signal blanking.

Haptic feedback can be considered as an exogenous signal (a disturbance) whose features can be predicted because they are produced by the plant under control. Haptic feedback is an information signal providing the receiver feedback about the state of the system. However, haptic feedback is also a power signal; sufficient force due to haptic feedback can backdrive the biomechanics of a participant. In this dissertation these topics are explored in two studies, one in the context of driving oscillations in a spring-mass system and the other in the context of shared control design for semi-autonomous vehicles.

Date: Monday, August 9, 2021
Time: 2:00 PM
Zoom: https://umich.zoom.us/j/97211413457
Chair: Brent Gillespie