Bathing robots have the potential to foster the independence of older adults who require assistance with bathing. Making human-robot interaction (HRI) for older persons as easy, effective, and user-satisfying as possible is, however, a major challenge in the development of such robots. The study aimed to evaluate the effectiveness (coverage, step effectiveness) and user satisfaction (After-Scenario Questionnaire, ASQ) with three operation modes (autonomous operation, shared control, tele-manipulation) for the HRI with a bathing robot in potential users. Twenty-five older adults who require bathing assistance tested these operation modes in a water rinsing task for the upper back. Autonomous operation led to maximum effectiveness (100%), which was significantly worse in the shared control (51.6-79.4%, p ≤ 0.001) and tele-manipulation mode (43.9-64.4%, p < .001). In the user-controlled modes, effectiveness decreased with decreasing robot assistance (shared control: 51.6-79.4% vs. tele-manipulation: 43.9-64.4%, p = 0.009-0.016). User satisfaction with the autonomous operation (ASQ: 2.0 ± 1.0pt.) was higher than with the tele-manipulation mode (ASQ: 3.0 ± 1.4pt., p = 0.003) and in trend also than with the shared control mode (ASQ: 2.5 ± 1.5pt., p = 0.071). Our study suggests that for an effective and highly satisfying HRI with a bathing robot in older users, operation modes with high robot autonomy requiring a minimum of user input seem to be necessary.
Keywords: Activities of daily living; aged; baths; human-robot interaction; robotics.