Authors: Alexander Broad, Todd Murphey, Brenna Argall
We present a novel approach to shared control of human-machine systems. Our method assumes no a priori knowledge of the system dynamics. Instead, we learn both the dynamics and information about the user’s interaction from observation through the use of the Koopman operator. Using the learned model, we define an optimization problem to compute the optimal policy for a given problem, and compare the user input to the optimal input. We demonstrate the efficacy of our approach with a user study. We also analyze the individual nature of the learned system dynamics and user interaction by comparing the effectiveness of our approach when using a model learned from the user’s own data to one learned from a consensus of other users’ data and also one learned from a domain expert’s data. Positive results include statistically significant improvements on task metrics when comparing a user only control paradigm with our shared control paradigm. Surprising results include findings that suggest that individualizing the model based on a user’s own data does not matter in learning a well-matched dynamic system. We explore this tension as it relates to developing human-in-the-loop systems further in the discussion.