Car black box

Abstract

Automated car accident detection can save lives by decreasing the time required for information to reach emergency responders. It has been shown that most car accidents were attributed to driver error, e.g. such as drowsiness. On the other hand, an event data recorder (EDR) is a device installed in some automobiles to record information related to vehicle crashes or accidents. However, as far as we know, there have been no attempt to combine the driver physiological data and EDR information together. In this project, we plan to develop a system that integrates driver physiological data and car event information on top of a smartphone. The smartphone can provide data much like that gathered by vehicular ECUs. The data received by the smartphone (i.e. from physiological sensor and ECU) can be stored in the internal memory of the smartphone or send to a server when 3G connection is available.

Introduction

Car accidents are a leading cause of death. Automated car accident detection can save lives by decreasing the time required for information to reach emergency responders. It has been shown that most car accidents were attributed to driver error, e.g. such as drowsiness. People with medical disease like diabetes or heart problem could potentially be at a higher risk of driving mishaps than those without the disease.

Using noninvasive sensors to detect driver physiological parameters are becoming more and more important. This approach is to measure the physiological changes of drivers from their bio-signals, and is especially useful to monitor driver state as driver-vehicle system is a safety-critical system. It is widely agreed that one of the contributing factors of vehicle accidents is drivers and, therefore, lots of research efforts have been focused on driver state/behavior. If a driver’s state can be detected online and then be fed back to the driver-vehicle system, this may lead to the development of a more intelligent driver assistance system. Some examples are: electrocardiogram (ECG) [1] was used as driver state measure as well as an effective indicator of drivers’ fatigue. Electroencephalogram (EEG) [2] reflects the activities of human brain and provides psycho-physiological information about a driver’s physical or mental stress state. Electromyographic (EMG) [3] measures the activity of the muscles of certain parts of the human body, and it is used for assessing the stress level of humans.

On the other hand, an event data recorder or EDR [4] is a device installed in some automobiles to record information related to vehicle crashes or accidents, similar to the "black box" found on airplanes. EDRs are triggered by electronically sensed problems in the engine (often called faults), or a sudden change in wheel speed. One or more of these conditions may occur because of an accident. Information from these devices can be collected after a crash and analyzed to help determine what the vehicles were doing before, during and after the crash or event. A common use of EDR data is to determine the responsibility of car accidents.

However, as far as we know, there have been no attempt to combine the driver physiological data and EDR information together. Integration of such two types of data could be very useful for hospital and police to understand the cause of accident. In addition, such integrated data could be very valuable for designing an early warning system before the car accident occurs, which is particularly useful for high-risk drivers like people with chronic diseases such as diabetics or heart problem. In addition, although EDR is useful in collecting information of car conditions before an accident happen, older cars normally do not equip with EDR.

Methodology

Recent advances in smartphone technologies are making it possible to detect car accidents in a more portable and cost effective manner than conventional in-vehicle solutions. In this project, we plan to develop a system that integrates driver physiological data and car event information on top of a smartphone (such as iPhone or Android phone).  The system architecture is shown as in Figure 1.

 


Figure 1.

 

In our architecture, the physiological sensors are connected to a gateway. The smart phone is fixed on the car and receives sensor information through the gateway wirelessly. Conventional in-vehicle accident detection systems rely on sensor networks throughout the car and direct interaction with the vehicle’s electronic control units (ECUs). These sensors detect acceleration/deceleration, airbag deployment, and vehicular rollover. The smartphone can interact with these internal sensors in a car through standard OBD-II [5] interface. However, while many new cars are equipped with OBD-II any smartphone application that required interaction with an onboard computer would be useless in cars that lacked one. Therefore, in addition to trying to obtain in-car sensor information, we also utilize sensor inside the smartphone such as accelerometer and gyro to detect car events such as sudden brake, turning and collision. In the event of an accident, the smartphone will experience the same forces and accelerations experienced by the occupants of the vehicle. Moreover, if the smartphone remains stationary relative to the vehicle during the collision, it is possible to use the data gathered from the smartphone to recreate and model the forces it experienced. In this case, the smartphone can provide data much like that gathered by vehicular ECUs. The data received by the smartphone (i.e. from physiological sensor and ECU) can be stored in the internal memory of the smartphone or send to a server when 3G connection is available.

Reference

[1] http://en.wikipedia.org/wiki/Electrocardiography

[2] http://en.wikipedia.org/wiki/Electroencephalography

[3] http://en.wikipedia.org/wiki/Electromyography

[4] http://en.wikipedia.org/wiki/Event_data_recorder

[5] http://en.wikipedia.org/wiki/On-board_diagnostics