How to evaluate two vehicle paths?

Hello,

I have two sets of xy-data which represent two vehicle trajectories. I want to evaluate how similar both paths are to each other and calculate a statistically meaningful value which describes the deviation of both paths.

This number should also be normalized based on path length or number of measurements to be able to compare a set of paths of my control loop.

I already thought about sum of squares. So for each x value delta_y² is computed, accumulated and at the end divided by the number of measurement points. Do you think this approach makes sense for my problem?

[Adam Danz]

We need to see some examples, a wide variety of examples, at least one where the two paths are similar, and a few where the two paths are quite different. There is no standard approach to this solution but that doesn't mean a solution isn't available. I have quite a lot of experience in this area -- my phd research involved non-human primates steering in a virtual environment -- and I've seen others in the forum who are interested in this topic as well. But before people invest time in explaining suggestions, you should provide several diverse examples.