Professor Andrew Wallace and Dr Sen Wang, from Heriot-Watt University, chased rain, snow, and fog around the rural northern roads and the urban sprawl of Edinburgh to gather the data.
As part of the road trip, the two researchers kitted out a van with light detection and ranging (lidar) radar, stereo cameras, and geo-positioning devices. The technology produced a new dataset of three hours’ worth of radar images and 200,000 tagged road objects, including other vehicles and pedestrians.
Image credit: Heriot-Watt University
According to the researchers, the information will be valuable to manufacturers and researchers of driverless vehicles as most public data currently available is based on sunny, clear days. It has also relied mainly on data collected from optical sensors, which do not work...