Cloud-based Driver Monitoring and Vehicle Diagnostic with OBD2 Telematics

Cloud-based Driver Monitoring and Vehicle Diagnostic with OBD2 Telematics

Malintha Amarasinghe (Department of Computer Science and Engineering, University of Moratuwa, Moratuwa, Sri Lanka), Shashika Ranga Muramudalige (Department of Computer Science and Engineering, University of Moratuwa, Moratuwa, Sri Lanka), Sasikala Kottegoda (Department of Computer Science and Engineering, University of Moratuwa, Moratuwa, Sri Lanka), Asiri Liyana Arachchi (Department of Computer Science and Engineering, University of Moratuwa, Moratuwa, Sri Lanka), H. M. N. Dilum Bandara (Department of Computer Science and Engineering, University of Moratuwa, Moratuwa, Sri Lanka) and Afkham Azeez (WSO2 Lanka Inc., Dehiwela, Sri Lanka)
Copyright: © 2015 |Pages: 18
DOI: 10.4018/IJHCR.2015100104
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

A cloud based, vehicular data acquisition and analytics system for real-time driver behavior monitoring, trip analysis, and vehicle diagnostics is presented. The system consists of an On Board Diagnostics (OBD) port to Bluetooth dongle, an app running on a smartphone, and a cloud-based backend. Based on OBD data, Complex Event Processors (CEPs) at both the smartphone and backend detects and notifies unsafe and anomalous events in real time. For example, CEP at the smartphone alerts drivers about rising coolant temperature and rapid fuel drops. It also provides a trip log and filter out messages to be sent to the backend, saving both the bandwidth and power. Backend CEP detects reckless driving in real time. Backend also uses historical data to detect driving anomalies and predict impending sensor failures. The system is tested on actual vehicles and tests demonstrate that the computing, bandwidth, and power consumption of the smartphone are reasonable.
Article Preview

Introduction

A car is no more a luxurious belonging of a person. It has rather become an integral part of a modern family. Use of vehicles all over the World has drastically increased during the last decade. For example, over 60 million passenger cars have been manufactured in the year 2012 (International Organization of Motor Vehicle Manufacturers, 2012). This rapid increase of vehicles has led to many concerns for all stakeholders. For example, all parties such as drivers, fleet vehicle managers, insurance companies, and law enforcement authorities are concerned about reckless driving and driver anomalies. Moreover, drivers and fleet vehicle managers are also concerned about fuel economy and maintenance. Furthermore, buyers and sellers are concerned about the condition of the vehicle, its maintenance, and accident records.

On Board Diagnostics (OBD) is a standard which allows assessing the status of various vehicle subsystems through the attached sensors. Some of the frequently accessed sensors include speed, engine RPM, coolant temperature, fuel rate and oxygen. OBD also reports various error states to help the diagnostic process. OBD2 (ELM Electronics, 2015), is the latest version of OBD and is implemented in most of the vehicles manufactured after 1996. OBD data can be read via the OBD port, and several adapters are commercially available to read the data from OBD2 port. ELM327, which is used in the proposed system, is one such adapter where the data read from the OBD2 port are transmitted via Bluetooth upon pairing.

Given the potential benefits of vehicular data analysis and the availability of technologies such as OBD, several vehicle monitoring and intelligent transport systems have been proposed. The vehicle diagnosis program proposed by Kim et al. (2010) provides diagnosis of different kinds of vehicle malfunctions within the navigation system. It displays the data collected through the OBD2 port in a human readable manner. To see this information the driver has to select the vehicle information menu of the navigation pane. While much of this information is displayed on the dashboard by default, the driver has to use his/her own experience/expertise to derive any meaning out of the sensor readings. Besides, a driver cannot be staring into the navigation pane while driving, as it distracts the driver. Many related work have been carried out in the area of vehicle monitoring through a server. While the main focus of most of these work is on tracking vehicles (Huang et al., 2013; Aljaafreh et al. (2011); Tahat et al., 2012), fault detection has also gained considerable attention (Huang et al., 2013; Tahat et al., 2012). However, in almost all the proposed systems, there has been either simple or no processing of the data gathered from the Engine Control Unit (ECU) prior to display. Hence, it is hard to alert and/or predict undesired outcomes such as reckless driving, accidents, and sensor failures, as they require both real-time and long-term analysis of data regarding the driving habits and the vehicle condition. Hence, there is a need for a single unified system which is capable of analyzing multiple sensor readings in real time and correlating them with past behaviors to provide more meaningful and accurate alerts. Moreover, a single unified platform makes it easier to collect a set of sensor readings and analyze them in different angles to cater the needs of multiple and competing stakeholders. For example, a driver’s acceleration and deceleration profile could be used by an insurance company and fleet manager to determine the risk taking of a driver, while the driver or fleet manager can use it to determine the impact on fuel economy. Aggregation of OBD and GPS data from multiple vehicles to the same platform also enables inferring traffic conditions and typical driving patterns in a given geographic region. The authors’ objective is to develop a system that demonstrates the feasibility of such a unified, driver monitoring and vehicle diagnostic system, and its ability to cater to the needs of multiple stakeholders. Through this process, authors also present several solutions to address some of the research challenges.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 8: 4 Issues (2017): 2 Released, 2 Forthcoming
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing