D³: Abnormal driving behaviors detection and identification using smartphone sensors

Abstract

Real-time abnormal driving behaviors monitoring is a corner stone to improving driving safety. Existing works on driving behaviors monitoring using smartphones only provide a coarsegrained result, i.e. distinguishing abnormal driving behaviors from normal ones. To improve drivers' awareness of their driving habits so as to prevent potential car accidents, we need to consider a finegrained monitoring approach, which not only detects abnormal driving behaviors but also identifies specific types of abnormal driving behaviors, i.e. Weaving, Swerving, Sideslipping, Fast Uturn, Turning with a wide radius and Sudden braking. Through empirical studies of the 6-month driving traces collected from real driving environments, we find that all of the six types of driving behaviors have their unique patterns on acceleration and orientation. Recognizing this observation, we further propose a finegrained abnormal Driving behavior Detection and iDentification system, D 3 , to perform real-time high-accurate abnormal driving behaviors monitoring using smartphone sensors. By extracting unique features from readings of smartphones' accelerometer and orientation sensor, we first identify sixteen representative features to capture the patterns of driving behaviors. Then, a machine learning method, Support Vector Machine (SVM), is employed to train the features and output a classifier model which conducts fine-grained identification. From results of extensive experiments with 20 volunteers driving for another 4 months in real driving environments, we show that D 3 achieves an average total accuracy of 95.36%.

An object oriented Bayesian network approach for unsafe driving maneuvers prevention system

“…Proposed System the latest smartphone technology, the functionality of GPS, accelerometer, gyroscope, and magnetometer are bundled on the smartphone device. The use of smartphones accelerometer was explored by Chen et al [6] to identify a number of driving patterns including swerving, zigzag, swaying, sideslipping, fast turn, and fast U-turn respectively. However, since indirect sensing is used, there is a tradeoff between accuracy and information enrichment of vehicles operation.…”

“…In addition, it becomes the part of the system that will be seen directly by drivers. Johannsdottir and Herdman [6] to understand a relation between working memory and SA discussed how working memory assists driver awareness in a hazardous situation in which sometimes in-cabin conversation with passengers cannot be avoided. Besides impairing driving performance, the drivers secondary task can degrade SA as well [26,27].…”

“…So far, to minimize unsafe driving maneuvers, existing researches include (1) detecting unsafe driving events based on driving pattern, such as unsafe acceleration and lane change [5], swerving and weaving [6]; (2) predicting unsafe maneuvers based on drivers body movement and intent which observes drivers head, hand, and eye gaze [7][8][9]; (3) predicting unsafe driving maneuvers using probabilistic approaches from driving history data that can be used to estimate stop behavior at intersection [10] and unsafe lane change that will be made by drivers [11]. However, in helping drivers to make a good driving maneuvers decision, delivering situational information is very important.…”

Analysis and Development of a Novel Algorithm for the In-vehicle Hand-Usage of a Smartphone

“…Commonly employed drivers risk profile indexes are based on measures related to speeding, driving smoothness, harsh accelerations, brakes, swerves, and cornerning, [3], [4], [5], [6]. In [7], it is shown that all most dangerous driving behaviors share a unique pattern in terms of acceleration and orientation. The recognized patterns are used in an abnormal driving behavior detection systems, which perform a real-time abnormal driving behavior monitoring.…”

Cloud-Assisted Safe Driving Framework

“…In order to prevent potential accidents in advance for safer driving, a cloud-assisted safe driving framework is proposed in this paper. The framework leverages smartphones or dash cams in vehicles to acquire particular information of the front vehicles [7]. The information of the front vehicles (e.g., license plate number) will be acquired and recognized in the cloud environment within the safe driving framework.…”