Vehicle Automation Field Test: Impact on Driver Behavior and Trust

Walter Morales Alvarez,Nikita Smirnov,Elmar Matthes,Cristina Olaverri-Monreal

DOI: https://doi.org/10.1109/ITSC45102.2020.9294751

2022-01-10

Abstract:With the growing technological advances in autonomous driving, the transport industry and research community seek to determine the impact that autonomous vehicles (AV) will have on consumers, as well as identify the different factors that will influence their use. Most of the research performed so far relies on laboratory-controlled conditions using driving simulators, as they offer a safe environment for testing advanced driving assistance systems (ADAS). In this study we analyze the behavior of drivers that are placed in control of an automated vehicle in a real life driving environment. The vehicle is equipped with advanced autonomy, making driver control of the vehicle unnecessary in many scenarios, although a driver take over is possible and sometimes required. In doing so, we aim to determine the impact of such a system on the driver and their driving performance. To this end road users' behavior from naturalistic driving data is analyzed focusing on awareness and diagnosis of the road situation. Results showed that the road features determined the level of visual attention and trust in the automation. They also showed that the activities performed during the automation affected the reaction time to take over the control of the vehicle.

Human-Computer Interaction

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the impact of autonomous driving systems on drivers' behavior and trust. Specifically, through field tests, the researchers analyzed the behavioral changes of drivers when performing non - driving tasks in autonomous vehicles and their trust in the automation system. The main purpose of the study was to evaluate the following two hypotheses:

1. Road features (such as road bends) and non - driving tasks will not affect drivers' trust in automation.
2. The reaction time for drivers to take over vehicle control does not depend on the type of non - driving tasks they perform during the automation process.

To verify these hypotheses, the researchers designed a series of experiments. They let participants drive with the assistance of an autonomous driving system and perform different non - driving tasks during the driving process, such as reading text on their mobile phones or writing on their mobile phones. By analyzing the participants' gaze direction and frequency, the time of attention to the road, and the reaction time to takeover requests, the researchers reached relevant conclusions, indicating that the type of non - driving tasks and road features do indeed affect drivers' trust in the autonomous driving system and their reaction time.