Hybrid lane model

1. A method of hybrid lane modeling, comprising:

receiving a roadway image;

extracting a set of lane points from the roadway image;

fitting a polynomial line to the set of lane points;

determining a fitted error of the fitted polynomial line;

outputting the polynomial line if the fitted error is less than a predetermined threshold wherein the predetermined threshold is a characterization of lane complexity;

selecting a set of clean lane points from the set of lane points if the fitted error is greater than the predetermined threshold;

removing noisy lane points from the set of lane points to yield the set of clean lane points; and

interpolating a cubic spline line to the set of clean lane points;

and

wherein the set of clean line points is more complex than the polynomial line.

2. The method of hybrid lane modeling of claim 1 , further comprising outputting the cubic spline hoe.

3. The method of hybrid lane modeling of claim 1 , wherein the set of lane points is at least one of equidistant and equitemporal.

4. The method of hybrid lane modeling of claim 1 , further comprising clustering the set of lane points.

5. The method hybrid lane modeling of claim 1 , further comprising converting the set of lane pouts to a bird's eye view.

6. The method of hybrid lane modeling of claim 5 , wherein the bird's eye view is based on a focal length, an optical center, a pitch angle, a yaw angle and a camera height.

7. The method of hybrid lane modeling of claim 1 , wherein the outputted polynomial line is a modeled lane.

8. The method of hybrid lane modeling of claim 1 , wherein the interpolated cubic spline line is a modeled lane.

9. The method of hybrid lane modeling of claim 1 , wherein a predicted movement along the interpolated cubic spline line does not exceed a movement noise threshold.

10. The method of hybrid lane modeling of claim 1 , wherein the fitted error is computed using the following two formulas:

y′ i =ax i 3 +bx i 2 +cx i +d and Error=1/ nΣ i=1 n (| y′ i −y i |).