Investigations on Driver Unique Identification from Smartphone’s GPS Data Alone

Abstract

Driver identification is an emerging area of interest in vehicle telematics, automobile control, and insurance. Recent body of works indicates that it may be possible to uniquely identify a driver using multiple dedicated sensors. In this paper, we present an approach for driver identification using smartphone GPS data alone. For our experiments, we collected data from 38 drivers for two months. We quantified the driver's natural style by extracting a set of 137 statistical features from data generated for each completed trip. The analysis shows that, for the "driver identification" problem, an average accuracy of 82.3% is achieved for driver groups of 4-5 drivers. This is comparable to the state of the arts where mostly a multisensor approach has been taken. Further, it is shown that certain behavioral attributes like high driving skill impact identification accuracy. We observe that Random Forest classifier offers the best results. These results have great implications for various stakeholders since the proposed method can identify a driver based on his/her naturalistic driving style which is quantified in terms of statistical parameters extracted from only GPS data.

Public Transport Driver Identification System Using Histogram of Acceleration Data

“…In this section, we thoroughly discuss the researches that utilize data obtained from a single sensor to perform driver identification. Chowdhury et al proposed a driver identification method using only GPS data [20]. Other data such as speed, acceleration, jerk, and jerk energy were computed from the GPS data.…”

“…The former is the research using multiple sensors [7][8][9][10][11][12][13][14][15][16][17]. The latter is the research using a single sensor [18][19][20].…”

“…Moreover, using multiple external sensors suffers from installation, data processing, implementation, and computation. Therefore, a number of recent approaches have been proposed to use only data obtained from a single sensor [18][19][20]. In [20], GPS data from a smartphone is used to identify the driver by deriving other data from the location such as acceleration, angular speed, and jerk.…”

“…Therefore, a number of recent approaches have been proposed to use only data obtained from a single sensor [18][19][20]. In [20], GPS data from a smartphone is used to identify the driver by deriving other data from the location such as acceleration, angular speed, and jerk. A Random Forest was used to identify the driver from statistical features extracting from those derived data.…”

Decision-Making Framework for Validation of Data Collection Process in a Survey with GPS Data

“…Characteristics of several methods can be seen in the table 9. [11], [15], [16], [17] [10], [11], [12], [13], [14], [15], [16] [15], [16], [17], [18], [19] Accuracy After the method was chosen, the rest of our research will continue with previous research, find something that is a contribution from research comparison of several studies with research on the validation of the enumeration process in a survey can be seen in table 10. Table 10 shows comparison of several study in decision-making with GPS data to analyze human behavior.…”