Automated vehicle road model definition system

Patent metadata

Field	Value
Publication number	US-10101746-B2
Application number	US-201615244474-A
Country	US
Kind code	B2
Filing date	Aug 23, 2016
Priority date	Aug 23, 2016
Publication date	Oct 16, 2018
Grant date	Oct 16, 2018

Abstract

A road-model-definition system suitable for an automated-vehicle includes a lidar-unit and a controller. The lidar-unit is suitable to mount on a host-vehicle. The lidar-unit is used to provide a point-cloud descriptive of an area proximate to the host-vehicle. The controller is in communication with the lidar-unit. The controller is configured to: select ground-points from the point-cloud indicative of a travel-surface, tessellate a portion of the area that corresponds to the travel-surface to define a plurality of cells, determine an orientation of each cell based on the ground-points within each cell, define a road-model of the travel-surface based on the orientation of the cells, and operate the host-vehicle in accordance with the road-model.

First claim

We claim:

1. A road-model-definition system suitable for an automated-vehicle, said system comprising:
a lidar-unit suitable to mount on a host-vehicle, said lidar-unit used to provide a point-cloud descriptive of an area proximate to the host-vehicle; and
a controller in communication with the lidar-unit, said controller configured to
select ground-points from the point-cloud indicative of a travel-surface,
tessellate a portion of the area that corresponds to the travel-surface to define a plurality of cells,
determine an orientation of each cell based on the ground-points within each cell,
divide a cell when a height-deviation from a plane of the cell by the ground-points within the cell is greater than a deviation-threshold,
combine adjacent cells when planes of the adjacent cells are parallel within a parallel-threshold,
designate a cell as a finalized-cell when the deviation from the plane of the cell by the ground-points within the cell is not greater than the deviation-threshold and the cell is not parallel to any adjacent cell within the parallel-threshold,
define the road-model of the travel-surface based on the orientation of finalized-cells when all cells are characterized as finalized-cells, and
operate the host-vehicle in accordance with the road-model.

2. The system in accordance with claim 1 , wherein the deviation is greater than the deviation-threshold when a height-difference to the plane from any ground-point is greater than the deviation-threshold.

3. The system in accordance to claim 1 , wherein the orientation is determined iteratively for each of the finalized-cells using fit-seeds selected from cloud-points within the finalized-cell characterized by a height-value less than a height-threshold.

4. The system in accordance with claim 1 , wherein the point-cloud is organized into a plurality of circumferential scan-lines of increasing radius centered around the lidar-unit, and the controller is further configured to determine a slope-value of a selected-cloud-point characterized as closest to a radial-line extending radially from the lidar-unit and associated with a selected-scan-line, said slope-value determined based on a height-change relative to a subsequent-cloud-point characterized as closest to the radial-line and associated with a subsequent-scan-line adjacent to the selected-scan-line, and define a line-segment indicative of a consecutive-group of cloud-points characterized as closest to the radial-line and characterized by slope-values that are equal within a slope-tolerance.

5. The system in accordance with claim 4 , wherein the line-segment is characterized by a segment-slope, and the cloud-points represented by the line-segment are classified as ground-points when the segment-slope is less than a slope-threshold.

6. The system in accordance with claim 1 , wherein the orientation of a cell is determined based on a plane defined by ground-points of the cell characterized by a height-value less than a height-threshold.

7. The system in accordance with claim 1 , wherein the orientation of a cell is determined using ground-points characterized as within a deviation-threshold of an estimated-plane.

8. The system is accordance to claim 1 , wherein the ground-points are determined using an iterative deterministic plane-fit using plane-fit-seeds from the point-cloud that were selected from a forward-portion located forward of the host-vehicle, and a cloud-point is classified as a ground-point when a fit-error is less than a plane-fit-threshold.

9. The system is accordance to claim 1 , wherein the ground-points are determined by dividing the points-cloud into a plurality of regions, a first plane is determined iteratively using a deterministic plane-fit on plane-fit-seeds from the point-cloud that were selected from an area located forward of the host-vehicle, a second plane is determined using an iterative deterministic plane-fit on plane-fit-seeds from the point-cloud that were selected from low height cloud-points in the first plane and the second plane, and a cloud-point is classified as a ground-point when a fit-error is less than a plane-fit-threshold.

Assignees

• Delphi Tech Inc

Inventors

• Izzat Izzat H
• Zermas Dimitris
• Pitambare Uday

Classifications

• G05D1/0212 Primary

  with means for defining a desired trajectory (involving a plurality of land vehicles G05D1/0287) · CPC title

• G05D1/0088

  characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours (using knowledge based models G06N5/00) · CPC title

• G06K9/00791

  Physics · mapped topic

• G01C21/30

  Map- or contour-matching · CPC title

• G06K9/00798

  Physics · mapped topic