Vehicle Automation–Other Road User Communication and Coordination: Theory and Mechanisms

Joshua E Domeyer,Heishiro Toyoda,John D Lee

doi:10.1109/access.2020.2969233

Abstract

When automobiles were first introduced in the early 1900s, poor communication and unsafe interactions between drivers and other road users generated resistance. This created a need for new infrastructure, vehicle design, and social norms to mitigate their negative effects on society. Vehicle automation may lead to similar challenges as drivers are supplanted by machines, potentially eliminating social behaviors that serve to smooth on-road communication and coordination. Through a review of communication, robotics, and traffic engineering literature, we explore the mechanisms that allow people to communicate on the road. We show the sensitivity of road users to signals that are sent through vehicle motion, suggesting a need to design vehicle automation kinematics for communication and not just external lighting signals. The framework further points to interdependence in communication where road users modulate their behaviors concurrently to exchange information and develop common ground. Designing automation to support common ground may smooth negotiations by generating interpretable signals in ambiguous situations. We propose a process to make automation observable and directable for other road users by considering vehicle motion during development of algorithms, interfaces, and interactions. Road users will be incidental users of vehicle automation—users whose goals are not directly supported by the technology—and poor communication with them may undermine the safety and acceptance of vehicle automation. As the reach of automation grows, communication among humans and machines may fundamentally change social interactions, requiring a framework to guide the process of making automation interactions smooth and natural.

Highlights

Vehicle automation may be the largest disruption to the transportation system since the original introduction of automobiles in the early 1900s [1]
This paper focuses on how vehicle automation will interact with its incidental users—other road users
The concepts described here do not invalidate the need for research on lighting and other explicit signals, it suggests that behavior is the basis for on-road communication and that vehicle automation communication designs that incorporate this may be more tolerable than those that do not

Summary

INTRODUCTION

Vehicle automation may be the largest disruption to the transportation system since the original introduction of automobiles in the early 1900s [1]. Resolving the challenge of interdependence for both primary and incidental users may be necessary for vehicle automation to be accepted Such system-wide effects inform a network perspective of people who interact with automation [13]. High profile crashes as well as low initial trust may result in interactions as incidental users having a larger effect on risk perception, making them critical for the acceptance of vehicle automation [22], [24], [25]. The observability and trialability of a technology may reduce perceived risk because they enhance the sense of control or agency provided by the technology This theory suggests that vehicle automation communication that improves an interaction, fits with the goals of incidental users, is easy to understand, and that can be tested, will be more successful than alternatives. With these tools as our foundation, we can consider more complex topics, such as joint action theory

COMMUNICATING THROUGH ACTION

EXISTING ROAD USER COMMUNICATION

Findings

DISCUSSION