Making a Delivery Robot Streetwise Using Expressive Movement

Abstract

Last-mile delivery robots intended to operate on the sidewalk still require supervision because they are only semi-autonomous. If left to their own devices, they tend to cause undesired interactions. Our project proposes a design philosophy that can reduce the amount and severity of these interactions, improving the overall acceptance of sidewalk delivery robots. Our proposed design philosophy uses the power of expressive movement to create delivery robots with a streetwise attitude. Being streetwise means possessing behavioural qualities that allow you to deal with the potential difficulties of an urban environment. Using creative association, we narrowed these qualities down to confidence, pride, calmness, stability and smoothness. A design probe study confirmed that the streetwise qualities can be used to differentiate between desirable and undesirable events. Next, we propose a methodology to combine our streetwise qualities with movement theories. We verify the efficacy of the method in a 2D movement study. The results suggest that our interpretation of the movement theories is on the right track, but more iterations are necessary to improve the method's reliability. During our final evaluation, we combine all our findings in a 3D video prototype scenario to confirm if they hold up in a more realistic and relatable context. We showed participants two different versions of this scenario to compare a streetwise delivery robot with a non-streetwise delivery robot in terms of desirability. Both versions turned out to have positives and negatives, but overall, the streetwise robot appeared more favourable.