Identification of Time Delay in Pointing Tasks for Human-Blimp Interactions

Abstract

In this paper, we explore the problem of estimating time delay parameters when humans interact with a Miniature Autonomous Blimp (MAB). As our earlier research has shown, the Vector Integration to Endpoint (VITE) model can be applied as a reset controller to efficiently minimize the blimp's motion overshoot and mimic the wand motions produced when people interact with it. Furthermore, our study shows that the closed-loop human-blimp dynamics are exponentially stable. However, dealing with the perceptual delay between the operator and the autonomous system is a key challenge in human-robot interaction experiments. In this paper, we use both experimental measurements and the Padé approximation method to model the time delay in human-blimp interactions. Through experiments, we estimate the parameters of the new VITE model after incor-porating the time delay. We also demonstrate that the new VITE model remains effective for simulating human-blimp interactions in 3D space, and that human intentions can be inferred from the trajectories of the blimp and pointer movements.