In-Vehicle Situation Monitoring for Potential Threats Detection Based on Smartphone Sensors

Abstract

This paper presents an analysis of modern research related to potential threats in a vehicle cabin, which is based on situation monitoring during vehicle control and the interaction of the driver with intelligent transportation systems (ITS). In the modern world, such systems enable the detection of potentially dangerous situations on the road, reducing accident probability. However, at the same time, such systems increase vulnerabilities in vehicles and can be sources of different threats. In this paper, we consider the primary information flows between the driver, vehicle, and infrastructure in modern ITS, and identify possible threats related to these entities. We define threat classes related to vehicle control and discuss which of them can be detected by smartphone sensors. We present a case study that supports our findings and shows the main use cases for threat identification using smartphone sensors: Drowsiness, distraction, unfastened belt, eating, drinking, and smartphone use.

Threats Detection during Human-Computer Interaction in Driver Monitoring Systems

“…Table 1 represents the technologies underlying the stages of the threat detection sequence of the developed method, collected via the literature analysis. The first column identifies the type of the threat (defined in our previous paper [ 4 ]), and the other columns correspond to the threat detection sequence stages. Three types of threats are considered: fatigue, inattention, and irritation.…”

“…Irritation detection was proposed in [ 4 ]. However, it is still under research, and no particular technologies can be mentioned.…”

“…Furthermore, the detected vulnerabilities may be ranked and organized using some external criteria to prioritize countermeasures (e.g., [ 7 ]). We performed a similar analysis for a vehicle control scenario in our earlier paper [ 4 ] and found that abnormal driver states pose severe safety risks, both to the driver and to the whole transportation system. Therefore, a method to identify such driver states needs to be developed.…”

“…We cover this gap in the case of a human–computer interaction in a driver monitoring system. The originality of our approach is that we frame the problem of the threat detection in a holistic way: we build on the driver–ITS system analysis performed in an earlier paper [ 4 ], and generalize existing methods for driver state analysis into a threat detection method, covering the identified threats. In the paper, we concentrate on the threats related to the human (driver and operator).…”

“…In the paper, we concentrate on the threats related to the human (driver and operator). In our previous paper we proposed a threat classification in the area of intelligent transportation systems [ 4 ]. In this paper, we propose a threat detection method that considers how to detect threats in vehicle cabin, as well as a reference model for operator–computer interaction.…”