Curve Fitting with RBS Functional Networks

This paper concerns the problem of discrete curve fitting: given a cloud of data points, the goal is to obtain the curve that fits these data points better. This is a classical problem that has received much attention in the literature because of its outstanding applications in several domains. Among the myriad of methods to tackle this issue, those based on artificial intelligence techniques have received increasing attention during the last few years. In this context, we introduce the RBS functional networks, a new type of functional networks based on weighted B-spline basis functions that reproduces the functional structure of NURBS curves. In this paper, we apply RBS functional networks to the curve fitting problem. Firstly, the method is described in detail. Then, some illustrative examples aimed at showing the good performance of this method for sets of 2D and 3D points are also given.