Accident Data Reconstruction
What is Accident Data?
Once a vehicle collision occurs the accelerometer within the GO device will detect an ‘accident level event’ and record second-by-second data of the accident at 100 Hz frequency. This data is automatically re-transmitted to MyGeotab for interpretation. Latitude, longitude, speed, accelerometer, as well as various other pieces of information are included that are essential to reconstructing the accident event.
How do I interpret Accident Data?
There are various tools available within MyGeotab that can be utilized to interpret the raw data from the device. In this document we will review these different tools. For illustration purposes, the information below comes from reconstructions of two real accidents.
Reconstruction of Accident Data
Case 1: Side Collision
The first step to reconstructing accident data is to view the accident in the vehicle’s trips history. This can be done by clicking on Map > Trips History. Specifying the vehicle, and choosing the date and the time of when the accident occurred will show that trip on the map. The first caution is that the GPS data may be inaccurate for the initial few seconds after the ignition event as the device is still attempting to obtain GPS latch (locate 4 satellites). In the image below, this vehicle was hit from the side as it was trying to pull onto the main road. The map, however, should not be the only tool used as there are several additional pieces of information that create a more complete accident reconstruction.
Accident and Log Data
The Log Details and Accidents report shows the individual logs as they are reported by the device before they are interpreted on the map. Look for the timestamp of the ‘accident level detection’ and match that time on the map to get a better understanding of the vehicle location when the accident occurred. Note that, by default, accident level detection occurs whenever the GO device accelerometer reading exceeds 2.5 G in any direction. This tool can be used by clicking Drivers and Activity > View Accidents & Log Data. Choosing the appropriate vehicle and date range will show each log for that time period.
Next, look for the accident detection log as well as accelerometer logs. The accelerometer G value is proportional to the amount of force exerted on the vehicle during the collision. This can be used to get a basic understanding of how severe the accident was. The below image shows the accident detection log which is created when an accident occurs.
The accelerometer graph is a useful tool that can be used to determine the direction from which the vehicle was hit and the relative force exerted in each direction due to the collision. This information is available by going to Engine & Maintenance > Engine Data > Engine Measurements. From here, choosing the correct vehicle and time will show all the engine data that was logged for the specified device. Next, click one of: accelerometer side to side, or forward and back (depending on the collision direction).
This will show an interactive graph of the accelerometer values in the chosen directions. In this case, since the vehicle was hit on its side, ‘Acceleration Side to Side’ was selected. Below is the image of the graph.
When viewing ‘Acceleration Side to Side’ any negative values represent acceleration to the right and any positive values represent acceleration to the left. In this case it is evident that this vehicle was hit from the left which caused acceleration to the right at a rate of 54 m/s^2. (Alternatively, when viewing a graph of forward and back acceleration, any positive value represents forward acceleration and negative represents backwards acceleration.)
The Speed Profile is a great tool that shows vehicle speed vs. time. This can be used to accurately determine the speed of the vehicle in question at the time of the accident. This can be done by clicking on Activity> Speed Profile. Below is the Speed Profile of the vehicle before, during, and after the accident. In this case we can see that during the time of the accident the speed of the vehicle was zero meaning the vehicle went toward the intersection and then stopped (waiting for traffic to pass) before entering the road.
The RPM graph is important in determining if the driver was depressing the throttle of the vehicle. Engine RPM can supplement information in the ‘Speed Profile’. Keep in mind that normal engine idle RPM varies depending on the type of vehicle. A good way to interpret the RPM level is to compare the engine RPM value at ignition start with an engine at normal operating temperature. In the image below this Mazda 3 Sedan had an RPM of about 1500 at Ignition start which went down to about 1000. This vehicle will therefore idle at 1000-1500 RPM. Since the highest RPM point is just below 2000, this RPM value indicates very gentle acceleration.
Based on the above information we can conclude that this vehicle moved toward the intersection slowly and came to a stop near the intersection. While stopped, another vehicle from the main road collided with this vehicle from the left side causing this vehicle to move to the right with an acceleration of 54 m/s^2. The information however is not conclusive enough to show if the vehicle was or was not sticking out onto the main road due to the accuracy of the GPS latch being 2.5 meters.
Case 2: Rear End
In this case the driver reported to be on the highway during rush hour when he was struck from the rear by the vehicle behind him. We will use the same process outlined above to analyze the data collected for this collision and verify this driver’s statement.
The first step is to view the accident on the map. A quick look suggests that the accident did occur on the highway. A quick speed check around the area of the accident (mousing over the line representing the vehicle’s path) also shows that the vehicle was moving at a slow speed compared to the highway speed limit. This, in combination with the time of day (rush hour), indicates that the vehicle was likely driving in heavy traffic.
Accident and Log Data
The next step is to view the Accident and Log Data. More specifically, the next step is to look for the accident level accelerometer event. Below is an image of the Accident and Log Data for this vehicle.
Analyzing the accelerometer values, it is important to find the one that stands out. In this case we can see the Forward accelerometer recorded 3.46 G which stands out as the highest value recorded. Also, it is important to note that there is an upwards (forward) force but barely any side to side force. This shows that the vehicle with the GO device was impacted on the rear end.
To confirm the above data we can now look at the accelerometer values in a graph. Since the forward value was so large we will plot Acceleration forward or braking from Engine Data. The images below are of the Engine data and the acceleration graph.
In the graphs above, positive values represent forward acceleration whereas negative values represent reverse acceleration. We can see that at the time of the accident this vehicle had a large forward acceleration indicating that it was hit from the rear.
The Speed Profile can be used to view the speed of the vehicle right before the accident occurred. In this case based on the image below we can see the vehicle slowed down gradually, possibly due to traffic. Then, moments before the accident, they accelerated by a small amount before decelerating again. Next, based on the timeline, a third small acceleration corresponds to the rear end collision event.
This shows that the vehicle in question had to stop abruptly after that second small acceleration event. The vehicle behind them may not have noticed the sudden braking of the vehicle just prior to the collision.
In case 2, we did not require the use of a RPM Graph. This is because we had confirmed that the vehicle was being driven slowly (possibly due to heavy rush hour traffic) when it was rear ended. The RPM Graph would not have provide any extra insight to the accident reconstruction.
In this case, based on the above information, we can conclude that the driver had a harsh braking event prior to the collision (possibly due to traffic) and shortly thereafter was impacted from the rear by the vehicle following them.
Prior to performing accident analysis using GO device data the following should be verified;
a) There was accurate ignition detection
b) There was a valid GPS latch
c) There was good connectivity of the SIM card based on the device log time and server upload time. You will have to contact support via ticket to confirm this.
d) Accelerometer has not been disabled due to excessive logging. Usually caused by loose installs.
Sources of error for vehicle position, speed, and acceleration are dependent on the specific model of device; verify these with Geotab for your specific device. In cases where the GO device was not rigidly attached to the OBDII connector, the initial acceleration data from the device remains correct, the reaction (or backward) acceleration will be in error due to the flexing of the harness. Additionally, the actual acceleration measured by the Geotab device may not be identical to the acceleration of the center of gravity of the vehicle depending on the location of the device relative to the center of gravity of the vehicle.
Data for Geotab GO devices are stored in secure data centers only accessible by Geotab.
Always state the VIN (Vehicle Identification Number) for the vehicle ECM data collected by the Geotab GO device.
Gps Latch: Accurate to 2.5 Meters
Speed: Accurate to 0.52 m/s
Curve logic: Polls at a 300 Hz error rate and upon detection re-polls at a rate of 100 Hz
Please sign in to leave a comment.