A Question of Speed

This collision, it was a motorcycle, 2012 Honda Shadow, um, colliding with the side of a minivan, and it caused that minivan to rotate about 100 degrees just due to the impact. And in addition to that, we had a 90-foot tire mark leading up to the impact from the motorcycle. When you just combine all those things in your head, uh, from my perspective as a reconstructionist, a speed at the speed limit there just doesn't make sense. Sometimes things just don't pass the sniff test, right? We were initially approached by a client who had previously had a reconstructionist look at this case. That reconstructionist had determined that the motorcycle was going about the speed limit, which was 55 miles an hour. So she wanted a second set of eyes, someone to come in, look at the evidence, and determine what the motorcycle speed was. We were fortunate in this case that there was a wealth of physical evidence that included a tire mark from the motorcycle.

Uh, it included good documentation of the rest positions of the vehicles, uh, and also a surveillance video from a nearby business that showed the collision and the lead up to that collision. Uh, we also had EDR data from the struck vehicle here that allowed us to analyze the collision. So that laid a really good foundation for us to be able to come in and do our own analysis. Typically, we'll do an initial look at the video, we'll determine frame rates, uh, evaluate its usefulness if we were going to try to track something through the frame. Um, in this case, it was a variable frame rate, so that produces its own set of complications. A constant frame rate would mean that there is a consistent amount of time between every single frame of a video. So if we're talking 30 frames per second, then between each frame, there's one 30th of a second.

Now, if we're talking variable frame rate, then we could still have 30 frames per second, but the interval between frames may not be the same. It may be changing every time. So we may have one, uh, where there's a 10th of a second between the frame and the next one there's two tenths of a second, and then the next one, there's. 05 seconds before the next frame. So it's varying frame to frame. So we're looking at all those things prior to us going out to the scene.

We did some testing of the camera itself. So we had a rental car, we attached a little amber flashing light, a beacon to the top of our car, drove it through the same direction the motorcycle was traveling a couple of times while we were logging our own speed of our rental car, uh, and then we were able to obtain the video of our drive through and de- determine what the frame rate of the video was that way. And then in the subject video, we also had some people that were walking through the parking lot leaving this business, and we were able to determine what kind of the average walking speed of a human is and determine the frame rate based on having those people travel at that average speed. In this particular case, because it was a nighttime crash, uh, it was very dark and the motorcycle was a, a long ways away from the camera, the best methodology for us to use was, uh, it's kind of known as a gate analysis.

So we're picking some features that we can see in the video and we're drawing lines from the camera through those features that are easily distinguishable and then through the travel lane where the motorcycle was traveling. And, uh, we can put those on a top-down diagram and determine which frames the motorcycle would be passing by those reference lines we've drawn, calculate a distance based on where those lines are at, and then we know the time interval, we get speed.

The police really did a good job documenting the evidence here, and those photos are critical because using the scene data that we collected during our site inspection, we can use a process called camera matching, which essentially lets us reconstruct the position from where some of these photographs were taken and then place objects into a three-dimensional scene where they were at at the time of the crash. So that's how we were able to locate and diagram the rest position of the Dodge of the motorcycle and the position of that tire mark leading up to impact. There was a reconstructionist on the case prior to our involvement. They had already inspected the 2018 Grand Caravan that was the rideshare vehicle, and as part of that inspection, they had downloaded the EDR data. And the data showed the pre-impact movement of the Dodge, uh, and then it also included the Delta-V during the crash itself, and that's front to back change in velocity and side to side, lateral change in velocity during the impact.

So there are methods in reconstruction where we can use that Delta-V of the struck vehicle to determine the speed of the striking vehicle. One thing that we think the prior reconstructionists likely overlooked in their analysis was the location of the airbag control module, which is where the EDR data is recorded and where the accelerometers that are recording this Delta-V are located in the vehicle. So because there was such a substantial amount of rotation in this case, about 100 degrees between the time of impact and where it came to rest at, uh, that rotation is going to drastically impact the lateral Delta-V that's recorded at the ACM. So the location of that ACM along the length of the vehicle makes a big, big difference. If that airbag control module was located near where the vehicle was struck, so in this case, that's the rear axle, then we would see a lateral Delta-V that's about what we would expect.

The airbag control module in this case was located underneath the center radio area, underneath the dashboard, so it was a good distance in front of where the vehicle was struck. And so as the vehicle's rotating there, it's gonna be experiencing less of a lateral acceleration than the center of gravity of the vehicle is. So we have to adjust that lateral Delta-V to be accounting for the distance between the airbag control module and where the center of gravity of the vehicle's at. When we do that, we get a much, much higher acceleration.

The speed limit on this road that the motorcycle was traveling on was 55 miles an hour, and that's very close to what the original accident reconstruction expert on this case was getting as a result of his analysis, and we ended up determining that the motorcycle was traveling 90 miles per hour prior to braking and 80 miles per hour at the time of impact. I think this case is really unique because it allowed us to use three different methodologies, and those methodologies all ended up in agreement with one another. First methodology was video analysis. We were able to use the surveillance video to track the headlight of the motorcycle as it approached the area of the collision, and that led us to, to a speed of 90 miles per hour for the motorcycle prior to braking. The second methodology was a simulation in a software called PC Crash, and in that simulation, we were trying to determine the impact speed that was necessary for the motorcycle to rotate the rideshare taxi to its rest position.

That gave us an impact speed of 80 miles per hour, and when we accounted for the pre-impact tire mark, it took us, again, very close to 90 miles per hour pre-braking. And then the final methodology here was an analysis of the EDR data on the Dodge Grand Caravan, which, uh, we used the Delta-V at the airbag control module, adjusted to the center of gravity of the vehicle, uh, and we determined that our PC crash simulation was consistent with that Delta-V on the Dodge. It also gave our client a high degree of confidence going into mediation, knowing that the information we gave them is right. We proved it using three different methodologies.