Path sensing using structured lighting

1. A method of sensing a physical feature of or along a path of a vehicle, comprising:

a. projecting a structured light pattern from a vehicle onto a path upon which said vehicle may travel, so as to generate either a plurality of light lines or a plurality of light spots on said path, wherein at least two of said plurality of light lines or said plurality of light spots are at different down-range locations relative to said vehicle;

b. capturing from said vehicle at least one image of said structured light pattern;

c. comparing at least one image-space location of at least one light line or light spot of said plurality of light lines or said plurality of light spots with a corresponding at least one reference image-space location of a set of a plurality of predetermined reference image-space locations;

d. determining if a light spot of said plurality of light spots, or a portion of a light line or said plurality of light lines, exists for which a corresponding said at least one image-space location is substantially different from said corresponding at least one reference image-space location; and

e. associating a reference world-space down-range location of said corresponding at least one reference image-space location with said light spot or said portion of said light line or said plurality of light lines, wherein said at least one reference image-space location and said reference world-space down-range location are stored for each of said plurality of light lines or said plurality of light spots, and each said at least one reference image-space location and said reference world-space down-range location corresponds to when a corresponding said at least one light line or light spot, or a similar light line or light spot, is projected from said vehicle onto a nominal relatively flat surface in world-space.

2. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 1 , further comprising detecting the presence of either an object on or a disturbance of said path responsive to said at least one image-space location that is substantially different from said corresponding at least one reference image-space location, wherein a location of said object on or said disturbance of said path relative to said vehicle is responsive to a relatively closest down-range location for which said at least one image-space location is substantially different from said corresponding at least one reference image-space location.

3. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 1 , wherein the operation of comparing at least one said image-space location of at least one light spot of said plurality of light spots with said corresponding at least one reference image-space location of said set of said plurality of predetermined reference image-space locations comprises sequentially comparing for different down-range locations relative to said vehicle, beginning with a relatively farthest down-range location and continuing with successively closer down-range locations until comparing for a relatively closest down-range location.

4. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 1 , wherein the operation of comparing at least one said image-space location of at least one light spot of said plurality of light spots with said corresponding at least one reference image-space location of said set of said plurality of predetermined reference image-space locations comprises sequentially comparing for different down-range locations relative to said vehicle, beginning with a relatively closest down-range location and continuing with successively farther down-range locations until comparing for a relatively farthest down-range location.

5. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 1 , wherein the operation of comparing at least one said image-space location of at least one light spot of said plurality of light spots with said corresponding at least one reference image-space location of said set of said plurality of predetermined reference image-space locations comprises:

a. sequentially comparing for different down-range locations relative to said vehicle, beginning with a relatively farthest down-range location and continuing with successively closer down-range locations until comparing for a relatively closest down-range location, then

b. sequentially comparing for different down-range locations relative to said vehicle, beginning with a relatively closest down-range location and continuing with successively farther down-range locations until comparing for a relatively farthest down-range location, then

c. performing steps a and b at least once.

6. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 1 , further comprising, for each light spot of said plurality of light spots:

a. locating a plurality of pixels associated with said light spot in said at least one image; and

b. determining a location of said light spot in said at least one image responsive to said plurality of pixels associated with said light spot in said at least one image.

7. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 6 , wherein the operation of locating said plurality of pixels associated with said light spot in said at least one image for each light spot of said plurality of light spots comprises searching a predetermined region-of-interest in said at least one image associated with said light spot.

8. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 7 , further comprising determining a pitch angle of said vehicle responsive to at least one image-space separation of at least one pair of said plurality of light spots, wherein each said light spot of said at least one pair of said plurality of light spots are at different world-space down-range locations and are at a substantially common world-space cross-range location, and a location of said predetermined region-of-interest in said at least one image is modified responsive to said pitch angle.

9. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 6 , wherein the operation of determining said location of said light spot in said at least one image comprises successively calculating centroids of each of a plurality of subsets of said plurality of pixels associated with said light spot in said at least one image so as to generate a single centroid that is representative of said location of said light spot in said at least one image.

10. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 6 , wherein the operation of determining said location of said light spot in said at least one image comprises:

a. using an adaptive gradient method to determine an edge profile of said light spot from said plurality of pixels associated with said light spot in said at least one image; and

b. calculating said location of said light spot in said at least one image responsive to said edge profile.

11. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 10 , wherein the operation of calculating said location of said light spot in said at least one image comprises calculating either a centroid of said edge profile, an average pixel location of said edge profile, or a median pixel location of said edge profile.

12. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 10 , further comprising:

a. determining a best-fit ellipse associated with said edge profile; and

b. determining at least one of a pitch angle or a roll angle of said vehicle responsive to an orientation of said best-fit ellipse in said at least one image.

13. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 6 , wherein the operation of determining said location of said light spot in said at least one image comprises successively binning said plurality of pixels associated with said light spot in said at least one image so as to generate a single binned pixel that is representative of said location of said light spot in said at least one image.

14. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 6 , wherein the operation of determining said location of said light spot in said at least one image comprises:

a. locating a plurality of edge points of said light spot, wherein each edge point of said plurality of edge points is located responsive to a Savitzky-Golay filtering process of said plurality of pixels along a corresponding polar direction relative to a nominal center of said plurality of pixels, and

b. determining said location of said light spot in said at least one image from a centroid of said plurality of edge points.

15. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 1 , further comprising:

a. determining a first aggregate metric responsive to a composite displacement of said plurality of light spots above said nominal relatively flat surface;

b. determining a second aggregate metric responsive to a composite displacement of said plurality of light spots below said nominal relatively flat surface; and

c. responsive to said first and second aggregate metrics, determining whether or not there is likely a substantial object on said path or a substantial disturbance of said path.

16. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 1 , further comprising:

a. fitting at least one first polynomial to a corresponding at least one row of said plurality of light spots in image space; and

b. determining a first aggregate metric of said at least one first polynomial responsive to a degree to which said at least one first polynomial mathematically fits said corresponding at least one row of said plurality of light spots in said image space.

17. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 16 , wherein said at least one first polynomial is quadratic.

18. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 1 , further comprising:

a. fitting at least one second polynomial to a corresponding at least one column of said plurality of light spots in image space; and

b. determining a second aggregate metric of said at least one second polynomial responsive to a degree to which said at least one second polynomial mathematically fits said corresponding at least one column of said plurality of light spots in said image space.

19. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 18 , wherein said at least one second polynomial is cubic.

20. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 1 , further comprising:

a. fitting at least one first polynomial to a corresponding at least one row of said plurality of light spots in image space;

b. determining a first aggregate metric of said at least one first polynomial responsive to a degree to which said at least one first polynomial mathematically fits said corresponding at least one row of said plurality of light spots in said image space;

c. fitting at least one second polynomial to a corresponding at least one column of said plurality of light spots in image space;

d. determining a second aggregate metric of said at least one second polynomial responsive to a degree to which said at least one second polynomial mathematically fits said corresponding at least one column of said plurality of light spots in said image space; and

e. responsive to said first and second aggregate metrics, determining whether or not there is likely a substantial object on said path or a substantial disturbance of said path.

21. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 1 , further comprising:

a. fitting at least one first polynomial to a corresponding at least one row of said plurality of light spots in image space;

b. determining a first aggregate metric of said at least one first polynomial responsive to a degree to which said at least one first polynomial mathematically fits said corresponding at least one row of said plurality of light spots in said image space;

c. fitting at least one second polynomial to a corresponding at least one column of said plurality of light spots in image space;

d. determining a second aggregate metric of said at least one second polynomial responsive to a degree to which said at least one second polynomial mathematically fits said corresponding at least one column of said plurality of light spots in said image space;

e. determining at least one feature related to a corresponding at least one tile, wherein said at least one tile is bounded by said at least one first polynomial for corresponding to a first row of said plurality of light spots, said at least one first polynomial corresponding to a second row of said plurality of light spots, said at least one second polynomial corresponding to a first column of said plurality of light spots, and said at least one second polynomial corresponding to a second column of said plurality of light spots, wherein said first and second rows of said plurality of light spots are adjacent to one another, and said first and second columns of said plurality of light spots are adjacent to one another; and

f. detecting or discriminating a physical feature of or along said path in said at least one image responsive to said at least one feature.

22. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 21 , wherein said at least one feature is selected from the group consisting of an area of said at least one tile, a set of four interior angles of said at least one tile, and a pair of vertical slopes of said at least one tile.

23. The method of sensing a physical feature of or along a path of a vehicle as recited in claim 1 , wherein said structured light pattern comprises said plurality of light lines, each line of said plurality of light lines is oriented substantially transverse to said path of said vehicle, and each said line of said plurality of light lines is separated in down-range from one another.