Human-automation interaction in automated vehicles: An innovative HMI design approach. The case of elderly and cyclists

Abstract

In recent years, significant progresses have been achieved in automated driving technologies and highly Automated Vehicles (AVs) are expected to become available to end-users within a decade. At the same time, many countries around the world are facing a demographic shift toward an aging society. Level 3 to 4 AVs will allow users to be released from the driving task for extended periods; however, they will be requested to take back control of the vehicle in specific situations. In the specific case of older road users, take over request stations could lead to significant complications in safety as well as mobility. Furthermore, communication needs to grant safe interactions between AVs and vulnerable road users, such as cyclists, should be deeply investigated and solutions should be proposed. This highlights the importance of designing AVs interfaces that are user friendly, safe, adaptable, and accessible for elderly drivers and for vulnerable road users. The present study aims at developing and testing an innovative framework for designing inclusive and adaptive HMIs, both addressing AVs users’ (elderlies) and other traffic participants (cyclists) basing on the Efficient Driver-Vehicle Cooperation Model proposed by Kraus et al. [1], integrating driver behavior models and user state assessment technologies. The model foresees that successful human-automation cooperation can be understood as the result of a relationship building process comparable to human relationships. Antecedents of safe and enjoyable interactions with highly AVs have been assessed and used as a basis for developing the framework, focusing on the psychological processes during the initial encounters with a system, in which system features interact with personality factors in building up beliefs and attitudes about a system affecting the further usage of the system. The proposed framework is integrating driver state assessment technologies and knowledge on human behavior to establish situation appropriate task function allocation between the driver and the vehicle. Our study is providing insights specific on vulnerable road users (e.g. elderly) characteristics and needs in designing interfaces.