Developing body gesture interaction guidelines with passenger elicitation for adjusting highly automated vehicle dynamics

Abstract

Highly Automated Vehicles (HAVs) could provide better safety, convenience, and eco-friendliness. However, realizing those benefits depends on not only the technical breakthrough but also the extent of people’s usage, which is significantly influenced by whether HAV driving styles match passengers’ preference. Therefore, this research studies user-elicited whole-body gestures for communicating the intention of adjusting vehicle dynamics in HAVs to provide design implications for the corresponding human-machine interaction (HMI). The study was based on user-elicitation gesture design method that immersed participants in HAV riding scenarios with a virtual reality (VR) simulator and elicited their gesture design for adjusting vehicle dynamics in HAV. The HAV driving scenarios, stemming from the literature on future HAV use cases and HAV ride plots, consist of three different road profiles and 15 discomfort-inducing plots. Participants were required to perform gesture interaction when they felt unsatisfied with the vehicle dynamics while experiencing the scenarios, report their interaction intentions and rationale of their gesture design after experiencing the scenarios, and draw down their interface need if there was. The user test (N=12) produced five kinds of intentions, at least one gesture design accompanied by explanations for each intention from each participant, and 12 sets of HMI design sketches. Based on the analysis of collected data, a taxonomy of whole-body gesture interaction for adjusting HAV dynamics was proposed. It was demonstrated that consensus existed among the participants on the gesture design. According to the consensus extent, an end-user generated gesture set was constructed. This paper highlights the implications of this work to the design of HAV HMI that assists passengers with communicating their intention of adjusting vehicle dynamics.