[Martin Taylor 2016.08.19.16.14]

Martin Taylor 2016.08.19.14.00]

...

To test the model you MUST use a reference track in which the along-track velocity is independent of the local curvature. It's easy to do. Here's anothe method to add to the ones suggested previously. You could have a "squiggle-like" column listing some arbitrarily varying local velocity in x and in y, and integrate that to produce the reference squiggle.

That is wrong. The squiggle shape must be independent of the velocity with which the target point moves along it, and this construction makes the shape depend on the velocity. To correct this, the squiggle shape must be generated by whatever mechanism is suitable (e.g. hand tracing or, to make an ellipse, parabolae in x and y), and an arbitrary velocity as a function of distance along the squiggle used to determine where on he squiggle the target point is at time "t".

Sorry about that.

Martin