Limitations. Our haptic steering interface was tested in a simulator and not in a real vehicle. However, for designing a suitable interface, a simulator has a number of benefits as it allows for accurately measuring the position of the vehicle on the track and we don’t have to worry about serious consequences when a driver crashes their vehicle. Because we were interested in testing whether prior driving experience would affect haptic steering performance, subject recruitment was restricted to recruiting blind subjects without any driving experience, which we found were difficult to recruit. Our study did not involve controlling the speed of the vehicle or detecting obstacles. Driving a vehicle on a track without any obstacles is a task where an autonomous vehicle most likely can outperform a human driver, but currently humans are better at driving a vehicle in traffic. As we only focus on steering performance, one could argue that our solution merely has a blind driver rely instructions from the computer to the steering wheel, which to an extent is true. Nevertheless a usable and effective steering interface with a study that presents quantitative data is an important step towards letting blind people drive vehicles independently, where in future work the presence of other traffic will be conveyed using non-visual means.