|
In this project, we designed and evaluated a steering wheel interface that can physically transform to reflect the mode of vehicle automation. We prototyped this steering wheel to function in a high-fidelity driving simulator at Stanford. The steering wheel was programmed to receive commands from the simulation software and to transform during a transition of control. We evaluated the transforming steering wheel interface using a controlled experiment (N=56) to see whether it could improve the performance of human drivers after a transition of control in a partial automation driving context. In the scenario, we varied the behavior of the steering wheel (transforming and non transforming) and also the transition time (2 seconds and 5 seconds). The results were then compared with those of the benchmark Minimum Necessary Time Studies (which used a conventional steering wheel) that we previously conducted. From the results of our studies, we note that drivers who experienced the transforming steering wheel performed significantly better than those who experienced a non transforming steering wheel / conventional steering wheel. To better understand the user experience issues associated with the transforming steering wheel movement, we also conducted a qualitative study (N=14) with the interaction design experts. Using Wizard of Oz design improvisation, we varied the speed and style of the transformations to explore the properties of the motion afforded by this robotic transforming steering system. The findings of these two studies help demonstrate how a transforming steering wheel design can assist human drivers of partially autonomous vehicles in safety-critical situations. The robotic transforming steering wheel prototype was designed / built by Brian Mok in 2017. The study was also designed by Brian Mok in 2017. |
|
In this project, we conducted three studies benchmarking the minimum necessary time needed for human drivers to safely regain control of the vehicle in an unstructured (or unplanned) transition after a period of autonomous driving at the driving simulator at Stanford. In these 3 studies, we used a simulated driving environment where control of the vehicle alternated between the human drivers and the car’s automated driving system. Participants were given no secondary task (only monitoring the car) in Study #1 (N=30), passive secondary task (watching a video) in Study #2 (N=30) and active secondary task (playing a game) in Study #3 (N=30) to perform while the car’s automated driving system was in control of the vehicle. In our simulation scenario, the car was performing automated freeway driving when it came upon an unexpected road hazard. At that point, the human driver was expected to immediately disengage from the secondary task and retake control of the vehicle after receiving a notification from the car. We tested three different transition conditions, where the transition occurred 2 seconds, 5 seconds, or 8 seconds before the road hazard. This critical event was not only the likely motivation for the unstructured transition but also acted as the principal test of post-transition driving performance. The drivers needed to react and regain the control of vehicle so that they could successfully assess the situation and safely negotiate the imminent road hazard. From the results of these three studies conducted in the driving simulator at Stanford, few human drivers in the 2-second condition were able to safely negotiate the road hazard situation, while the majority of drivers in 5 or 8-second conditions were able to navigate the hazard safely. The study was designed by Brian Mok in 2015. |
|
While automated driving systems will become increasingly capable and common in the future, there will still be instances when human drivers want or need to make corrections to the car’s automated driving behavior. In this project, we conducted two studies exploring how driving interfaces could be designed to better execute the drivers’ intentions. In our first study, adult participants (N=40) experienced a simulated driving scenario that varied the behavior of the car’s automation (perfect driving and imperfect driving) and the intervention modalities (takeover and takeover + influence). At certain segments, the car’s automation would drive perfectly or weave within the lane. During those times, participants could intervene using the available modalities. When experiencing instances of imperfect driving, drivers who had the ability to takeover+influence intervened more often than drivers who were only given the option to takeover. As intervening would require them to resume full control, drivers in the takeover condition were more tolerant of the imperfect driving. Also, most drivers tried to intervene initially by influencing the car, even those drivers who were only given the ability to takeover. In our second study, we examined how participants (N=40) of different demographics (high school students and seniors) would respond when they were subjected to the imperfect driving scenarios. High school drivers intervened just as much as the adult drivers. However, senior drivers intervened far less. These two studies suggest that when intervention is necessary, human drivers have a desire for shared control, which allows them to act as supervisors rather than operators of automated vehicles. The study was designed by Brian Mok in 2016. |
|
In this project, we conducted an exploratory study to better understand the questions and concerns for human drivers of automated cars and to identify critical issues concerning the transitioning/sharing control between the drivers and the vehicle automation. With the help and feedback of interaction and interface design experts (N=12), we conducted a series of design improvisation sessions with our Wizard of Oz autonomous driving simulator system to determine essential aspects governing the driving experience in an automated vehicle. Through the feedback collected in these sessions, insights in five areas were discovered: drivers’ desire for shared control, transitions in driving mode, response latency, addressing requests, and drivers’ trust in the car. Additional examination yielded concepts that were implemented and tested in future work. The study was designed by Brian Mok in 2014. |