Every detail counts: Why research plays an important role in the development of safety systems.

Experiment vs. reality. Controlled laboratory conditions vs. realistic environment. Two different ways of looking at or exploring a topic. In the end however, the result is not necessarily the same. But is it really always an either-or situation?

In our case, the answer is no. Because in order to have a detailed basis for further developments in the area of safety functions, we always have to combine both sides.

A good example is a study from Dhaval. Together with a small team, he investigated the extent to which the activity of the muscles from the pelvis down has an influence on the degree of injury in a pedestrian-vehicle collision.

Starting point for the study was a noticeable difference between experiments and real accident data. For example, there are current (laboratory) studies in which a large number – around 90 percent – of knee injuries (medial collateral ligament (MCL)) are observed when postmortem human subjects (PMHS) hit a vehicle. However, the analysis of various accident databases suggests that the MCL is not injured as often in accidents as in laboratory experiments. Where does this difference come from?

If you would work only with results from experiments, you would probably draw the wrong conclusions. Misinterpretations could – in extreme cases – lead to potentially dangerous, undesirable developments of functions and systems. We want to avoid that.

The existing data from various experiments that imitate the “interaction” between a vehicle and pedestrian are of course indispensable for developing and/or validating pedestrian dummies/ impactors and computational human body models (HBMs). At the same time however, it is essential to use real (accident) data in a second step and to compare both. That’s how possible misinterpretations can be eliminated and results that are as realistic as possible can be generated. And that’s exactly our goal.

Among the parameters, the muscle contractions have been suspected to have a relevant role in reducing knee ligament injuries. To check this, Dhaval and his team investigated the effect of these muscle contractions on MCL elongation and lateral knee bending. For this they used the Explicit Finite Element Solver Visual Performance solution. How they proceeded exactly and what interesting conclusion they came to is summarized in a paper and was presented at the IRCOBI by Dhaval. Curious? It’s definitely worth reading it: click here, to get to the paper.

Assumptions are never enough for us. Not in the field of biomechanics and not in any of our other fields. We don’t just accept numbers and data. Our goal is to question critically and develop an in-depth understanding of the topic and to determine the data and findings that are as close as possible to the real world. This is the only way we can achieve a maximum of safety for autonomous driving. For us, 360° vehicle safety means the all-round package – from beginning to end. Dhaval’s results can now be incorporated when developing safety features. Because every detail counts. Every bit of research is important for our future ideas, solutions and projects.