Consider this scenario - you've just collided with the back of another car because you weren't paying attention. However, you won't be able to claim that you were braking hard if an electronic record shows that you didn't begin to slow down until the moment of impact. Or perhaps you were speeding? Once again, the electronic record will reveal all to crash investigators.

Convicted by your car? - it's more than just a possibility, with one such case having already occurred in the US. There, a driver involved in a double fatality claimed he had been travelling at about 100km/h. However, the electronic record logged by his vehicle's airbag showed that his speed just five seconds before impact was, in fact, 184km/h!

So what data is logged and why is it recorded? Do all airbag-equipped cars have this facility? How can you read it? And who owns the information?

The implications - not only for drivers but also for insurance companies, the police, car rental companies and fleet owners - are profound. But if the thought of your car logging your driving behaviour horrifies you, here's a let-off - at least for the time being. At this stage, General Motors in the US appears to be the only car company that's wholeheartedly embracing the technology.

In fact, GM is publicly releasing details on their systems and also working with a third party provider to make available a dedicated data reader for general purchase.

The potential benefits of Event Data Logging (EDL) has also resulted in strong US Government support for adopting universal standards for such systems. In other words, due to the influence of US legislation on car makers, it's probably only a matter of time before all cars have Event Data Logging recorded in a standard format that can be easily read.

Automotive logging

About 20 years ago, the fuel and ignition control in cars started a move from mechanical systems (carburettors and points) to electronic systems (EFI and electronically controlled ignition). These electronic systems rely on sensors to measure various parameters, such an engine airflow, engine speed and throttle position, with an Electronic Control Unit (ECU) then making decisions about the fuel injection pulse width and ignition timing. Most of these systems have the ability to detect and store faults in memory so that they can be later read out and diagnosed.

It comes as no surprise then that the airbag control system not only has the ability to store data but also uses a wide variety of sensors as part of its decision making process. However, the use of the controller as an Event Data Recorder (EDR) goes a step further - not only are fault codes stored but in some systems, the outputs from a variety of sensors are also continually logged.

Early development

So how did this come about? The story goes back to the early 1970s, when the US National Transportation Safety Board recommended that vehicle manufacturers gather information on vehicle crashes using on-board collision sensing and recording devices. As a result, since 1974, General Motors (GM) systems have recorded data for impacts that resulted in the triggering of the airbag (a "deployment event"), while other systems were also introduced that could additionally record "near deployment" events.

Subsequently, in 1999, GM introduced a system that could also record pre-crash data - ie, data is recorded to a buffer on a continuous basis and overwriting ceases immediately if a crash occurs. Ford in the US started installing EDRs in one model in 1997 and by 1999 nearly all its US models were so equipped. A range of other manufacturers either admit to some data recording or are looking to implement such strategies.

Rather than use airbag control systems to record crash and pre-crash data, some US-manufactured heavy trucks use the engine's ECU instead. For example, Cummins, Detroit Diesel and Caterpillar all use electronic control systems on their diesel engines which also log driving data.

The GM airbag system

The information recorded by GM airbag systems includes data for both deployment and near deployment events.

A near deployment event (ie, one where the airbag doesn't inflate) is defined as an event that's severe enough to "wake up" the algorithm within the control unit (an algorithm is used to analyse the severity of the crash pulse; ie, the control unit uses the shape and magnitude of the deceleration pulse it is undergoing before deciding whether or not to fire the airbag).

Two different systems are used by GM; one stores data on the near deployment event which had the greatest change in road speed, while the other stores the most recent near deployment event. In both cases, the following data is recorded:

• Driver's Seat Belt: this is recorded as buckled or unbuckled. However, this may be recorded incorrectly if power to the unit is lost during the crash.
• SIR Warning Lamp: the on/off status of the Supplemental Inflatable Restraint warning lamp is recorded.
• Change in Forward Velocity: this is determined by integrating the average of four 312ms acceleration samples and is recorded in RAM every 10ms. Depending on the module, either 300ms or 150ms of this data is available.
• Time To Deployment: the time in milliseconds between the start of the event (ie, enabling of the algorithm which requires two consecutive acceleration samples of over 2g) and the command for the airbag deployment.
• Time Between Events - the time in seconds between a deployment event and a near deployment event, if that time is less than five seconds.
• Vehicle Speed: the pre-crash speed, recorded every second for five seconds prior to any event. This information is derived from the vehicle speed sensor.
• Engine RPM: engine speed, as derived from the engine management system. As with vehicle speed, it is recorded every second for five seconds prior to any event.
• Throttle Opening: the percentage that the throttle is open, where 100% is wide open. This information is sent by the engine management system along with engine and vehicle speeds and is again recorded every second for 5s prior to any event.
• Brake Status: brakes on/off, as derived from the ABS or engine management unit every second for 5s prior to any event. Braking intensity is not recorded.
• Data Validity: a check that none of the four pre-crash parameters (vehicle speed, engine rpm, throttle opening or brake status) is out of range or has logged faults.

Pre-Crash Graph GM Airbag Module   Post-Crash Graph GM Airbag Module

In addition, the number of ignition key cycles at the time of the events and at the time of download is logged, as is whether or not the passenger-side front airbag has been manually switched off.

One of the two GM EDR units is designed so that 150ms after the deployment algorithm has been enabled, all the data stored in the memory is permanently written to EEPROM. It then cannot be erased, cleared or altered, so this type of device must be replaced after an airbag deployment.

As a matter of interest, the Ford system records both longitudinal and lateral acceleration, the deployment strategy for the dual-stage airbag, seat-belt use, pretensioner operation and the fore-aft position of the driver's seat.

One reason that data from the GM system is being widely used in crash research is that the company licensed the Vetronix Corporation to build a data retrieval tool for their EDR as far back as 1999. Ford subsequntly followed suit for their own EDR system.

The Vetronix Crash Data Retrieval (CDR) tool consists of both hardware and software. The hardware component comprises an interface between the vehicle's diagnostic connector (or the EDR itself where the vehicle wiring has been damaged) and a PC. In operation, the CDR system reads the hexadecimal code stored in the EDR and converts it to engineering units, making it available in both tabular and graphical forms.

And the cost of this unit? - about $US2500.

Data usefulness

EDRs improve crash analyses, both by simplifying and improving the accuracy of the reconstruction process. This results in more detailed and more accurate conclusions. Tables 1 & 2 summarise the information available to crash investigators with and without EDRs.

Before EDR, crash investigators could only rely on vehicle damage and other obvious physical signs like skid marks (less likely with ABS) in order to make major judgements. So logged data on vehicle speed and other parameters can be very useful.

Data validity

So how good is the data collected via an EDR? The answers to that question are surprisingly broad; certainly there is plenty of information available for someone who wants to fight EDR evidence in a court of law. However, on the other side of the fence, if used carefully, the data gained from an EDR is invaluable when it comes to determining the events that occurred before and during the crash.

So just what are the potential problems? They are as follows:

• Problem 1: vehicle speed, engine rpm, throttle opening and brake status are logged only once per second - a sampling frequency that's much too low when analysing many types of crashes. For example, did the driver brake at 3.1 or 3.9 seconds before impact? The difference is major. Additionally, this data is not synchronised with the start of the crash data and is potentially offset from the crash data by up to one second.
• Problem 2: the recorded data goes back only five seconds before the algorithm enable event occurs. There is no record of vehicle behaviour earlier than this - behaviour which might show erratic driver inputs, for example.
• Problem 3: the use of only five data points for each of the speed, rpm, throttle opening and brake status parameters can give a false impression; eg, if the data is plotted on a graph, with the various points connected by a straight line. In reality, the true values of any of these parameters might have been quite different between the discrete points, compared to the values indicated by the graphs.
• Problem 4: most EDRs record speed only in a longitudinal direction. However, many accidents also involve lateral as well longitudinal movement and so the speed recording may give a false impression of the events that occurred. No current original equipment EDRs record vertical accelerations.
• Problem 5: where the crash does not involve a major deceleration - eg, when a large truck hits a small car or when a pedestrian is run over - the EDR may not record the event at all.
• Problem 6: vehicle speed, engine rpm, throttle opening and brake status all depend for their accuracy on sensors and/or switches. However, vehicle speed and throttle position sensors can vary by up to 10% in accuracy, a point that seems to have been overlooked by some researchers.

Other research

A great deal of work has gone into testing the relationship between the data gathered from EDRs and that gained through other logging techniques. One approach is to measure the vehicle's change of velocity using the EDR and compare that figure with the crash test impact speed.

A series of Canadian tests has shown that there is usually fairly good agreement between the calculated and actual speeds - eg, an actual impact speed of 40.3km/h and an EDR-calculated speed of 42.4km/h. Typically, the EDR showed a slightly higher speed because it was affected by the car bouncing back off the barrier after the collision.

However, one test involving a 2000 Ford Taurus had a significantly greater difference between the actual (47.8km/h) and EDR (53.6km/h) speeds. The testers suggested that this discrepancy had been caused by a spike in the acceleration/time curve, caused by structural deformation in the area where the EDR was mounted. A major discrepancy also occurred in another test, where a 1988 Chevrolet Cavalier's EDR lost power during the crash. The independently measured test speed was 64.8km/h but the EDR showed 56.8km/h.

Away from the laboratory, the usefulness of the data - even with these reported inaccuracies - can be clearly demonstrated. In one case, an 83-year-old male driver of a 2000 Buick Century was negotiating a right-hand curve when he ran off the road, travelled down an embankment into brush and tall grass, then crossed a level section of lawn and a gravel driveway before finally colliding with two large rocks. The car came to rest approximately 140 metres from where it left the road.

Pre-crash data obtained from the EDR indicated that the driver wasn't operating the throttle or the brakes for at least five seconds prior to the impact with the rocks. At the crash scene, the driver was lethargic and he subsequently died in hospital. An autopsy showed that he had died from the results of a brain haemorrhage that had occurred while he was driving - a diagnosis well supported by the EDR data.

Conclusion

If the US success at implementing onboard diagnostics in cars is repeated with EDR, it's very likely that all new cars will have accident crash logging in 5-10 years.

So if you are ever involved in a car crash and there's some debate about the circumstances, think about the implications of an EDR. It may only be a matter of time before authorities can access such data.

Analysing An Accident

Potential Benefits of Event Data Recorders (1). Real Time Assistance: the use of EDR data in conjunction with Automatic Collision Notification systems would aid in quickly locating crashes and despatching emergency personnel with better crash information in advance. (2). Law Enforcement: obtaining impartial EDR data from a collision would help in more accurately determining the facts surrounding the incident. (3). Government Initiatives: the collection of EDR data would enable governments to introduce effective initiatives to help reduce fatalities, injuries and property loss. (4). Vehicle Design: EDRs allow manufacturers to collect accurate data to monitor system performance and improve vehicle design. (5). Highway Design: the use of EDR data can assist in assessing highway roadside safety and managing road systems. (6). Insurance/Legal: Additional objective data provided by EDRs advance quicker and fairer resolution of insurance and liability issues (7). Research: EDR data could provide objective data for researching driver behaviour and performance, as well as other research related topics. (8) Owners/Drivers: EDRs can help fleet owners and drivers monitor vehicle and driver performance, to ensure the safe and efficient movement of people and cargo. Canadian Multidisciplinary Road Safety Conference, 2001.

Who Owns The Logged Data? While the potential benefits of EDRs are highlighted by road safety researchers, many drivers and some vehicle manufacturers are concerned about the privacy implications. In fact, the US Federal Motor Carrier Safety Administration has stated that the following standards should apply to controlling access to EDR data: The vehicle's owner should also own the EDR data. Only the vehicle's owner, or another party having the owner's permission, may access the EDR data. Exceptions would include instances where a law enforcement official has a warrant for a crash investigation. One method of assuring that only owners have access is through the use of an EDR password. The storage and retrieval of EDR data must protect the privacy rights of the individual in accordance with law. At this stage, none of those points has been implemented, although truck owners can deactivate the EDR by setting the deceleration threshold inappropriately, giving them some measure of control over the data being collected. Certainly, there needs to be more public debate about the privacy issues involved with EDR.

Share this Article: 

Digg this story

Seed Newsvine

Post to Reddit

Save on del.icio.us

Email