Influence of legibility on perceived safety in a virtual human-robot path crossing task

Abstract

In the future robots will more and more enter our daily life. If we want to increase their acceptance it is necessary that people feel safe in the surrounding of robots. As a prerequisite we think that the robot's behavior has to be legible in order to achieve such a feeling of perceived safety. With our present experiment we assess the perceived safety participants feel when an autonomous robot is crossing their path. Therefore participants are presented with a video based scenario in first person perspective. The robot is moving with two different navigation algorithms which allows us to test whether the legibility has an influence on the perceived safety and whether the two navigation algorithms differ regarding their resulting legibility and thus perceived safety. Results show that legibility as defined here increases perceived safety of both navigation methods while the level of perceived safety differs between them.