Development of robotic mobility for infants: rationale and outcomes.

OBJECTIVES To assess the feasibility of a robotic mobility device for infants using alternative control interfaces aimed at promoting early self-initiated mobility, and to assess the effects of a training protocol and robot experience. DESIGN Observational and pre-post quantitative case studies. SETTING Standardised, research laboratory and day-care centres with toys and individuals familiar to infants. PARTICIPANTS Children with and without disabilities, aged 5 months to 3 years. INTERVENTIONS In each study, infants were seated over a Pioneer™ 3-DX mobile robot. Some infants controlled the directional movement of the robot by weight shifting their body on a Nintendo® Wii™ Balance Board (the WeeBot), while others used a modified joystick. Infants participated in five sessions over 2 to 5 weeks. Sessions consisted of administering a 10-minute training protocol preceded and followed by 2 to 3 minutes of free play. One child with motor impairment used a button switch array and a different experimental design. MAIN OUTCOME MEASURES From the videotaped free-play periods, goal-directed behaviours were coded and time in motion was measured. In the training period, a scoring system was developed to measure the infants' driving performance. RESULTS Preliminary outcomes indicate that infants without disabilities, aged 5 to 10 months, demonstrated significant improvement in driving performance and goal-directed movement using the WeeBot. Infants who used the joystick were less successful on all measures. Results for infants with disabilities using the WeeBot were mixed. CONCLUSIONS Mobile robots offer promise to enhance the development of early self-mobility. Novel types of interfaces, such as the WeeBot, warrant further investigation.