Representing navigable surface boundaries of lanes in high definition maps for autonomous vehicles

Patent metadata

Field	Value
Publication number	US-10474164-B2
Application number	US-201715853633-A
Country	US
Kind code	B2
Filing date	Dec 22, 2017
Priority date	Dec 30, 2016
Publication date	Nov 12, 2019
Grant date	Nov 12, 2019

Abstract

A system generates a high definition map for an autonomous vehicle to travel from a source location to a destination location. The system determines a low resolution route and receives high definition map data for a set of geographical regions overlaying the low resolution route. The system uses lane elements within the geographical regions to form a set of potential partial routes. The system calculates the error between the potential partial route and the low resolution route and removes potential partial routes with errors above the threshold. Once completed, the system selects a final route and sends signals to the controls of the autonomous vehicle to follow the final route. The system determines whether surface areas adjacent to a lane that are not part of the road are safe for the vehicle to drive in case of emergency. The system stores information describing navigable surface areas with representations of lanes.

First claim

What is claimed is:

1. A non-transitory computer readable storage medium storing instructions for implementing a navigable surface boundary in a high definition map encoded thereon that, when executed by a processor, cause the processor to perform the steps including:
receiving a high definition map representation of a geographical region comprising three-dimensional representations of structures in a geographical region comprising a representation of a road including one or more lanes, each lane having a pair of lane boundaries, each lane boundary representing an edge of the lane;
identifying from the high definition map, one or more structures of the geographical region, each structure located outside the road;
for each of the identified structures, identifying a set of points on the structure based on their perpendicular distance from a lane boundary of a road, wherein each point of the set lies outside the road;
generating a polyline representation through the identified set of points;
storing in the high definition map representation of the geographical region, the polyline representation as a navigable surface boundary for the road, wherein the navigable surface boundary corresponds to a physical area that lies outside the road such that a vehicle may safely navigate within the physical area; and
providing the high definition map representation of the geographical region to an autonomous vehicle driving within the geographical region, wherein the vehicle makes a determination whether to drive over a navigable surface outside the road.

2. The non-transitory computer readable medium of claim 1 , wherein a structure represents at least one of the following:
a fence;
a safety barrier;
a series of posts;
a wall;
a curb;
a ditch or drainage depression;
a hill;
a building; or
a tree.

3. The non-transitory computer readable medium of claim 1 , wherein for each identified structure, a type of surface is determined, the types of surfaces comprising:
pavement;
gravel; or
dirt.

4. The non-transitory computer readable medium of claim 1 , wherein structures are identified from the high definition map created using on sensor data captured by sensors or one or more vehicles driving on the lane.

5. The non-transitory computer readable medium of claim 1 , wherein identifying one or more structures representing obstructions further comprises:
receiving a point cloud representation of the geographical region;
from the point cloud representation, identifying a plurality of three dimensional points having a height above a threshold level;
mapping the plurality of three dimensional points to an image; and
using a machine learning based object recognition method to classify a structure corresponding to the plurality of three dimensional points mapped to the image.

6. The non-transitory computer readable medium of claim 1 , wherein generating a polyline further comprises, for each identified structure:
determining a perpendicular distance between data points corresponding to the identified structure and the lane boundary;
selecting a data point with a smallest perpendicular distance from the lane boundary; and
determining a polyline passing through the selected data point and one or more neighboring data points.

7. The non-transitory computer readable medium of claim 1 , wherein the polyline representation can be generated in one or more of the following coordinate systems:
a two dimensional coordinate system; or
a three dimensional coordinate system.

8. The non-transitory computer readable medium of claim 1 , wherein the level of difficulty for the vehicle to travel over the navigable surface is determined, the determination comprising:
for each of a plurality of points along a polyline, determining a score representing a level of difficulty for a car to travel over the navigable surface associated with the polyline based on a classification of a structure associated with the navigable surface; and
for each of the plurality of points along the polyline, storing the score with the representation of the lane.

9. The non-transitory computer readable medium of claim 1 , wherein the level of difficulty for the vehicle to travel over the navigable surface is based on the type of navigable surface over which the vehicle would travel.

10. The non-transitory computer readable medium of claim 1 , wherein implementing the navigable surface boundary comprises:
for the lane element that the vehicle is currently traveling on, accessing the polylines corresponding to the navigable surface boundary;
receiving an indication of an emergency, the emergency representing an event forcing the vehicle out of the lane;
responsive to the indication of the emergency, determining that the vehicle can travel safely over the navigable surface, the determination based on the score representing the level of difficulty; and
sending one or more signals to a control of the vehicle, the signals causing the vehicle to travel within the navigable surface boundary.

11. A computer-implemented method comprising:
receiving a high definition map representation of a geographical region comprising representations of structures in a geographical region comprising representation of a road including of one or more lanes, each lane having a pair of lane boundaries, each lane boundary representing an edge of the lane;
identifying from the high definition map, one or more structures of the geographical region, each structure located outside of the road;
for each of the identified structures, identifying a set of points on the structure based on their perpendicular distance from a lane boundary of a road, wherein each point of the set lies outside the road;
generating a polyline representation through the identified set of points;
storing in the high definition map representation of the geographical region, the polyline representation as a navigable surface boundary for the road, wherein the navigable surface boundary corresponds to a physical area that lies beyond the road such that a vehicle may safely navigate within the physical area; and
providing the high definition map representation of the geographical region to an autonomous vehicle driving within the geographical region, wherein the vehicle makes a determination whether to drive over a navigable surface outside the road.

12. The computer-implemented method of claim 11 , wherein a structure represents at least one of the following:
a fence;
a wall;
a curb;
a hill;
a building; or
a tree.

13. The computer-implemented method of claim 11 , wherein for each identified structure, a type of surface is determined, the types of surfaces comprising:
pavement;
gravel; or
dirt.

14. The computer-implemented method of claim 11 , wherein structures are identified from the high definition map created using on sensor data captured by sensors or one or more autonomous vehicles driving on the lane.

15. The computer-implemented method of claim 11 , wherein identifying one or more structures representing obstructions further comprises:
receiving a point cloud representation of the geographical region;
from the point cloud representation, identifying a plurality of three dimensional points having a height above a threshold level;
mapping the plurality of three dimensional points to an image; and
using a machine learning based object recognition method to classify a structure corresponding to the plurality of three dimensional points mapped to the image.

16. The computer-implemented method of claim 11 , wherein generating a polyline further com

Assignees

• Deepmap Inc

Inventors

• Wheeler Mark Damon

Classifications

• G08G1/0969

  having a display in the form of a map · CPC title

• G08G1/09685

  where the complete route is computed only once and not updated · CPC title

• G08G1/0141

  for traffic information dissemination · CPC title

• G08G1/04

  using optical or ultrasonic detectors · CPC title

• G06F16/9537

  Spatial or temporal dependent retrieval, e.g. spatiotemporal queries · CPC title