Road boundary detection based on radar and visual information

1 . A method for detecting a road-boundary of a road based on radar information and visual information, the method comprises:

estimating, based on the visual information obtained by a visual sensor of a vehicle, locations of road-boundary points up to a first distance from the visual sensor; wherein the visual information also comprises road-boundary points beyond to the first distance; wherein a distance ambiguity of visual based location determination of road-boundary points with the first distance range does not exceed an ambiguity threshold; wherein a distance ambiguity of visual based location determination of road-boundary points outside the first distance exceeds the ambiguity threshold; wherein a distance ambiguity of visual based location determination of a road-boundary point is an uncertainty associated with a determining of a location of the road-boundary point;

estimating, based at least on (i) the radar information, and (ii) angular-constrains regarding angular relationships between adjacent road-boundary points, wherein the angular constraints define a maximum allowed angular deviation between adjacent road-boundary points, the estimating uses the angular constraints to iteratively search for successive road-boundary segments within search regions defined by the angular constraints, to determine locations of road-boundary points outside the first distance range; wherein the angular constraints are based on physical limitations regarding road boundaries; and

determining the shape and position of the road-boundary based on the locations of the road-boundary points within the first range distance.

2 . The method according to claim 1 , wherein the estimating of the locations of the road-boundary points outside the first distance is executed in an iterative manner.

3 . The method according to claim 2 , wherein the iterative manner starts from the first distance.

4 . The method according to claim 1 , comprising obtaining a location of a current road-boundary segment; estimating a search region for searching for the next road-boundary segment, based at least in part on the angular-constrains; and searching the next road-boundary segment within the search region.

5 . The method according to claim 1 , wherein the first distance does not exceed few tens of meters.

6 . The method according to claim 1 , wherein the estimating of the locations of road-boundary points outside the first distance range is also based on visual information about locations of road-boundary points outside the first distance range.

7 . The method according to claim 1 , comprising responding to the determining of the shape and position of the road-boundary.

8 . The method according to claim 7 , wherein the responding comprises validating road-boundary information about the determining the shape and position of the road-boundary based on a map of an environment of the vehicle.

9 . The method according to claim 1 , comprising dynamically determining the ambiguity threshold thereby dynamically determining the first distance.

10 . The method according to claim 9 , wherein the dynamically determining of the ambiguity threshold is based on a current visibility of an environment of the vehicle.

11 . The method according to claim 9 , wherein the dynamically determining of the ambiguity threshold is based on an angle of an optical axis of the visual sensor to the road.

12 . A non-transitory computer readable medium for detecting a road-boundary of a road based on radar information and visual information, the non-transitory computer readable medium stores instructions for:

estimating, based on the visual information obtained by a visual sensor, and locations of road-boundary points up to a first distance from the visual sensor of a vehicle; wherein the visual information also comprises road-boundary points beyond to the first distance;

wherein a distance ambiguity of visual based location determination of road-boundary points with the first distance range does not exceed an ambiguity threshold; wherein a distance ambiguity of visual based location determination of road-boundary points outside the first distance exceeds the ambiguity threshold; wherein a distance ambiguity of visual based location determination of a road-boundary point is an uncertainty associated with a determining of a location of the road-boundary point;

estimating, based at least on (i) the radar information, and (ii) angular-constrains regarding angular relationships between adjacent road-boundary points, wherein the angular constraints define a maximum allowed angular deviation between adjacent road-boundary points, the estimating uses the angular constraints to iteratively search for successive road-boundary segments within search regions defined by the angular constraints, to determine locations of road-boundary points outside the first distance range; wherein the angular constraints are based on physical limitations regarding road boundaries;

determining the shape and position of the road-boundary based on the locations of the road-boundary points within the first range distance.

13 . The non-transitory computer readable medium according to claim 12 , wherein the estimating of the locations of the road-boundary points outside the first distance is executed in an iterative manner.

14 . The non-transitory computer readable medium according to claim 13 , wherein the iterative manner starts from the first distance.

15 . The non-transitory computer readable medium according to claim 12 , further storing instructions for obtaining a location of a current road-boundary segment; estimating a search region for searching for the next road-boundary segment, based at least in part on the angular-constrains; and searching the next road-boundary segment within the search region.

16 . The non-transitory computer readable medium according to claim 12 , wherein the first distance does not exceed few tens of meters.

17 . The non-transitory computer readable medium according to claim 12 , wherein the estimating of the locations of road-boundary points outside the first distance range is also based on visual information about locations of road-boundary points outside the first distance range.

18 . The non-transitory computer readable medium according to claim 12 , that further stores instructions for responding to the determining of the shape and position of the road-boundary.

19 . The non-transitory computer readable medium according to claim 16 , wherein the responding comprises validating road-boundary information about the determining the shape and position of the road-boundary based on a map of an environment of the vehicle.

20 . A system that comprises a processor that is configured to:

estimate, based on the visual information obtained by a visual sensor of a vehicle, locations of road-boundary points up to a first distance from the visual sensor; wherein a distance ambiguity of visual based location determination of road-boundary points with the first distance range does not exceed a ambiguity threshold; wherein a distance ambiguity of visual based location determination of road-boundary points outside the first distance exceeds a ambiguity threshold; wherein a distance ambiguity of visual based location determination of a road-boundary point is an uncertainty associated with a determining of a location of the road-boundary point; dynamically determining the ambiguity threshold thereby dynamically determining the first distance;

estimate, based at least on (i) the radar information, and (ii) angular-constrains regarding angular relationships between adjacent road-boundary points, wherein the angular constraints define a maximum allowed angular deviation between adjacent road-boundary points, the estimating uses the angular constraints to iteratively search for successive road-boundary segments within search regions defined by the angular constraints, to determine locations of road-boundary points outside the first distance range; wherein the angular constraints are based on physical limitations regarding road boundaries;

determine the shape and position of the road-boundary of a road based on the locations of the road-boundary points within the first range distance; and

dynamically determine the ambiguity threshold thereby dynamically determining the first distance.