A Study on the Relationship between In-Vehicle Systems and Driving Safety Based on Drivers’ Visual Features

Abstract

The use of in-vehicle systems has become a potential factor in traffic accidents, resulting in increased driving workload. The driver's visual features could be used to represent the change of driving workload. Therefore, this study examined the influence of in-vehicle systems on drivers’ visual features through a real-vehicle test on urban roads. Devices including eye trackers and navigation systems were installed in the test vehicle, and four urban routes were selected to conduct the on-vehicle experiment. A comparison was made regarding the difference of subjects’ fixation and saccadic behavior with or without the condition of in-vehicle systems, with devices at different positions and different information release modes. Within-vehicle systems used, drivers’ saccadic times and duration, as well as small-scale saccadic behavior all increased significantly, and the average saccadic speed was greater. Besides, drivers’ fixation area and the average fixation proportion towards the road ahead both reduced significantly, with the fixation location closer to the ground, but the fixation duration remained unchanged. When the in-vehicle systems was installed above the shelf and provided voice broadcast information, drivers’ saccadic frequency, scope, and speed were low. The fixation distribution area and fixation times towards the road ahead changed the least. These results indicate that in-vehicle systems occupies more cognitive resources, resulting in more mental workload and less driving safety. The influence on driving safety could be mitigated by optimizing the position of in-vehicle systems and information release mode.