The Nursebot Project, a multidisciplinary and multi-institutional collaboration, has produced two separate prototype robots for older adults: a mobile robotic personal assistant and a robotic walker. These robots are intended to provide cognitive and physical assistance that augment rather than replace human care giving and support. We are proposing a two-phase investigation. In Phase I, we will conduct 6 randomly-ordered experiments with 48 adults (Mini-Mental Status Examination [MMSE] score of 20+) age 65 and older who will be videotaped walking 100 feet with each variation in several design features of the robotic walker (shape of handle bars, height of handle bars, responsiveness of the haptic interface, speed of movement, and user's freedom to deviate from a pre-planned path) and the mobile robotic personal assistant (speed of movement, proximity dudng ambulation, user's freedom to deviate from a pre-planned path, visual display, and auditory guidance). For each walk, we will observe participants' ambulation behavior (steadiness, gait, grip force, time to move from sitting to standing and from standing to walking, time to complete the walk, frequency of stops, whether conversation was initiated or engaged in, and whether the destination was reached). We will determine participants' preference for the variations in each design feature being evaluated, as well as the perceptions of participants and blinded and unblended observers regarding walking performance with each robot (ease of operation, level of comfort, and sense of security, stability, and fatigue). In Phase II, we will ask 24 adults age 65+ with an MMSE of 20+ who need ambulation assistance or guidance to complete 6 randomly-ordered, 150-foot walks under varying distraction (high vs. low) and walking conditions (receiving usual assistance or guidance vs. using the robotic walker vs. accompanied by the mobile robotic personal assistant). In addition to Phase I measures, for each walk we will ascertain perceptions of distractibility, frustration, and confidence reaching the destination from participants and blinded and unblinded observers. Summary descriptive statistics will be computed for each phase. Due to the pilot nature of this investigation, analyses will be strictly exploratory, using ANOVA to explore within- and between-subject differences in ambulation behavior and perceptions. These statistics will be used for effect and sample size estimation for future studies.
