Object detection and classification for autonomous vehicles

• US 8,195,394 B1
• Filed: 07/13/2011
• Issued: 06/05/2012
• Est. Priority Date: 07/13/2011
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Zhu, Jiajun, Montemerlo, Michael Steven, Urmson, Christopher Paul, Chatham, Andrew
• Primary Examiner(s): Tran, Dalena
• Application Number:
  US13/181,999
• Time in Patent Office: 328 Days
• Field of Search: 701/514, 701/400, 701408-410, 701/422, 701428-430, 701/433, 701/437, 701/468, 701/444, 701/446, 701/458, 701/461, 701/469, 701/523, 340/435, 340/904, 340/988, 340/995.14, 34099517-99519, 340/995.22, 340/995.24, 340/995.25, 348/113, 348/116, 348/118, 348/119, 348/142, 348/135, 348/148, 348/149, 382/103, 382/104, 382/107
• US Class Current: 701/514
• CPC Class Codes: G01C 21/26 specially adapted for navigation in a road network
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G08G 1/09626 where the origin of the information is within the own vehicle, e.g. a local storage device, digital map
• CPC Class Codes: G08G 1/096725 where the received information generates an automatic action on the vehicle control
• CPC Class Codes: G08G 1/096741 where the source of the transmitted information selects which information to transmit to each vehicle
• CPC Class Codes: G08G 1/096775 where the origin of the information is a central station

Abstract

Aspects of the disclosure relate generally to safe and effective use of autonomous vehicles. More specifically, objects detected in a vehicle'"'"'s surroundings may be detected by the vehicle'"'"'s various sensors and identified based on their relative location in a roadgraph. The roadgraph may include a graph network of information such as roads, lanes, intersections, and the connections between these features. The roadgraph may also include the boundaries of areas, including for example, crosswalks or bicycle lanes. In one example, an object detected in a location corresponding to a crosswalk area of the roadgraph may be identified as a person. In another example, an object detected in a location corresponding to a bicycle area of the roadgraph and identified as a bicycle. By identifying the type of object in this way, an autonomous vehicle may be better prepared to react to or simply avoid the object.

Claims

1. A method for identifying an object, the method comprising:

determining a current location of a vehicle;

accessing roadgraph data based on the current location of the vehicle, the roadgraph data including information about a roadway including one or more tagged areas of the roadway;

collecting information about the vehicle'"'"'s surroundings using an object detection device;

detecting a moving object and a geographic location of the moving object based on the collected information;

comparing, by a processor, the geographic location of the moving object to the roadgraph data; and

identifying, by the processor, a type of the moving object based on whether the geographic location of the moving object corresponds to a tagged area of the roadgraph.

2. The method of claim 1, wherein determining the current location of the vehicle comprises:

receiving data from a geographic position component;

identifying map information based on the received data; and

correcting the current location of the vehicle based on the identified map information.

3. The method of claim 1, wherein:

the object detection device includes a laser configured to collect range and intensity data; and

the range and intensity data is used to detect the moving object and the location of the moving object.

4. The method of claim 1, wherein the corresponding tagged area of the roadgraph is a crosswalk and the identified type of the moving object is a pedestrian.

5. The method of claim 1, wherein a tag of the corresponding tagged area of the roadgraph identifies the corresponding tagged area as a crosswalk.

6. The method of claim 1, wherein the corresponding tagged area of the roadgraph is a crosswalk and the identified type of the moving object is a bicyclist.

7. The method of claim 1, wherein a tag of the corresponding tagged area of the roadgraph identifies the corresponding tagged area as a bicycle lane.

8. The method of claim 1, further comprising the processor causing the vehicle to maintain a minimum distance between the vehicle and the moving object based on the identification of the moving object.

9. The method of claim 1, wherein the vehicle is an autonomous vehicle and the method further comprises maneuvering the autonomous vehicle along the roadway, by the processor, without input from a human driver.

10. The method of claim 1, wherein the roadgraph graphically stores features of the roadway and relationships of the features of the roadway.

11. The method of claim 1, further comprising, if the moving object is not within a corresponding tagged area of the roadgraph, using at least one image matching technique to identify the type of the moving object.

12. A device for identifying an object, the device comprising:

memory storing roadgraph data including information about a roadway including one or more tagged areas of the roadway;

an object detection device for collecting information about the device'"'"'s surroundings;

a processor coupled to the memory and the object detection device, the processor being operable to;

maneuver a vehicle along a roadway;

determine a current location of the vehicle;

access the roadgraph data based on the current location of the vehicle;

receive collected information from the object detection device;

detect a moving object and a location of the moving object based on the collected information;

compare the location of the moving object to the roadgraph data; and

identify a type of the moving object based on whether the location of the moving object corresponds to a tagged area of the roadgraph.

13. The device of claim 12, wherein the corresponding tagged area of the roadgraph is a crosswalk and the identified type of the moving object is a pedestrian.

14. The device of claim 12, wherein the corresponding tagged area of the roadgraph is a bicycle lane and the identified type of the moving object is a bicycle.

15. The device of claim 12, wherein the processor is further operable to maintain a minimum distance between the vehicle and the moving object based on the identification of the moving object.

16. The device of claim 12, wherein the processor is further operable to, if the moving object is not within a corresponding tagged area of the roadgraph, identify the type of the moving object using at least one image matching technique.

17. A tangible, non-transitory, computer-readable storage medium on which computer readable instructions of a program are stored, the instructions, when executed by a processor, cause the processor to perform a method of identifying an object, the method comprising:

maneuvering a vehicle along a roadway;

determining a current location of the vehicle;

accessing roadgraph data based on the current location of the vehicle, the roadgraph data including information about the roadway including one or more tagged areas of the roadway;

collecting information about the vehicle'"'"'s surroundings using an object detection device;

detecting a moving object and a location of the moving object based on the collected information;

comparing the location of the moving object to the roadgraph data; and

identifying a type of the moving object based on whether the location of the moving object corresponds to a tagged area of the roadgraph.

18. The tangible, non-transitory, computer-readable storage medium of claim 17, wherein the corresponding tagged area of the roadgraph is a crosswalk and the identified type of the moving object is a pedestrian.

19. The tangible, non-transitory, computer-readable storage medium of claim 17, wherein the corresponding tagged area of the roadgraph is a bicycle lane and the identified type of the moving object is a bicycle.

20. The tangible, non-transitory, computer-readable storage medium of claim 17, wherein the method further comprises, if the moving object is not within a corresponding tagged area of the roadgraph, using at least one image matching technique to identify the type of the moving object.