Autonomous actuated-interfaces provide a unique research opportunity for shared-control interfaces, as the human and the interface collaborate using the physical interaction modality, manipulating the same physical elements at the same time. Prior studies show that sharing control with physical modality interfaces often results in frustration and low sense-of-control. We designed and implemented adaptive behavior for shared-control actuated-interfaces that extends prior work by providing humans the ability to anticipate the autonomous action, and then accept or override it. Results from a controlled study with 24 participants indicate better collaboration in the Adaptive condition compared with the Non-adaptive one, with improved sense-of-control, feelings of teamwork, and overall collaboration quality. Our work contributes to shared-control tangible, shape-change, and actuated interfaces. We show that leveraging minimal non-verbal social cues to physically communicate the actuated-interface's intent, coupled with providing autonomy to the human to physically accept or override the shift-in-control, improves the shared-control collaboration.
https://dl.acm.org/doi/abs/10.1145/3491102.3517449
The ACM CHI Conference on Human Factors in Computing Systems (https://chi2022.acm.org/)