System and method for evaluating the perception system of an autonomous vehicle

Patent metadata

Field	Value
Publication number	US-9911030-B1
Application number	US-201715587680-A
Country	US
Kind code	B1
Filing date	May 5, 2017
Priority date	Oct 5, 2010
Publication date	Mar 6, 2018
Grant date	Mar 6, 2018

Abstract

A method and apparatus are provided for optimizing one or more object detection parameters used by an autonomous vehicle to detect objects in images. The autonomous vehicle may capture the images using one or more sensors. The autonomous vehicle may then determine object labels and their corresponding object label parameters for the detected objects. The captured images and the object label parameters may be communicated to an object identification server. The object identification server may request that one or more reviewers identify objects in the captured images. The object identification server may then compare the identification of objects by reviewers with the identification of objects by the autonomous vehicle. Depending on the results of the comparison, the object identification server may recommend or perform the optimization of one or more of the object detection parameters.

First claim

The invention claimed is:

1. A method for optimizing object detection performed by autonomous vehicles, the method comprising:
receiving, by one or more computing devices having one or more processors, sensor data captured by a sensor of an autonomous vehicle, wherein the sensor data includes a first object labels applied by the autonomous vehicle, each of the first object labels corresponding to a computed speed of a particular object of the sensor data;
receiving, by the one or more computing devices, a one or more second object labels not applied by the autonomous vehicle for the sensor data corresponding to a computed speed of the particular object;
determining, by the one or more computing devices, differences in speed between corresponding objects of the first object labels and one or more second object labels;
determining, by the one or more computing devices, a number and value of differences in speed based on the determined differences; and
generating, by the one or more computing devices, a recommendation for an adjustment of at least one object detection parameter of at least one of a plurality of sensors based on the determined number and value of differences in speed in order to optimize object detection.

2. The method of claim 1 , wherein the one or more second object labels include labels generated at least in part by one or more processors of a second computing device.

3. The method of claim 1 , further comprising, for each given object of the sensor data, determining a computed speed for an object label by determining a distance that the given object has traveled in a series of images based on a rate at which the series of images were captured.

4. The method of claim 3 , wherein the series of images includes laser point cloud images.

5. The method of claim 1 , further comprising displaying on a display device the recommendation.

6. The method of claim 1 , further comprising reconfiguring the autonomous vehicle according to the recommendation.

7. The method of claim 1 , further comprising:
determining a difference between a first value indicating a quantity of the first object labels applied by the autonomous vehicle and a second value indicating a quantity of the one or more second object labels; and
determining whether the autonomous vehicle has met a predetermined missed object threshold corresponding to a maximum allowable quantity of object labels not applied by the autonomous vehicle by comparing the difference to the threshold, and wherein the recommendation is generated further based on the determination of whether the vehicle has met the predetermined missed object threshold.

8. The method of claim 7 , further comprising for each given one of the first object labels, determining an object label ratio corresponding to an area of intersection between a given one and a given one or more second object labels and an area of union between the given one and the given one or more second object labels, and wherein the recommendation is generated further based on the ratio.

9. The method of claim 1 , further comprising for each given one of the first object labels, determining an object label ratio corresponding to an area of intersection between a given one and a given one or more second object labels and an area of union between the given one and the given one or more second object labels, and wherein the recommendation is generated further based on any determined ratios.

10. The method of claim 9 , further comprising:
determining an average object label ratio value for each of the determine object label ratios; and
comparing the average object label ratio value to an object label ratio threshold to determine whether the object label ratio threshold has been met, and wherein the recommendation is generated further based on the determination of whether the object label ratio threshold has been met.

11. A system for optimizing object detection performed by autonomous vehicles, the system comprising one or more computing devices having one or more processors configured to:
receive sensor data captured by a sensor of an autonomous vehicle, wherein the sensor data includes first object labels applied by the autonomous vehicle, each of the first object labels corresponding to a computed speed of a particular object of the sensor data;
receive one or more second object labels not applied by the autonomous vehicle for the sensor data corresponding to a computed speed of the particular object;
determine differences in speed between corresponding objects of the first object labels and one or more second object labels;
determining, by the one or more computing devices, a number and value of differences in speed based on the determined differences; and
generate a recommendation for an adjustment of at least one object detection parameter of at least one of a plurality of sensors based on the determined number and value of differences in speed in order to optimize object detection.

12. The system of claim 11 , wherein the one or more second object labels include labels generated at least in part by one or more processors of a second computing device.

13. The system of claim 11 , wherein the one or more processors are further configured to, for each given object of the sensor data, determine a computed speed for an object label by determining a distance that the given object has traveled in a series of images based on a rate at which the series of images were captured.

14. The system of claim 11 , further comprising the vehicle.

15. The system of claim 11 , further comprising a display device, and wherein the one or more processors are further configured to display the recommendation on the display device.

16. The system of claim 11 , wherein the one or more processors are further configured to reconfigure the autonomous vehicle according to the recommendation.

17. The system of claim 11 , wherein the one or more processors are further configured to:
determine a difference between a first value indicating a quantity of the first object labels applied by the autonomous vehicle and a second value indicating a quantity of the one or more second object labels;
determine whether the autonomous vehicle has met a predetermined missed object threshold corresponding to a maximum allowable quantity of object labels not applied by the autonomous vehicle by comparing the difference to the threshold; and
generate the recommendation further based on the determination of whether the vehicle has met the predetermined missed object threshold.

18. The system of claim 17 , wherein the one or more processors are further configured to:
for each given one of the first object labels, determine an object label ratio corresponding to an area of intersection between a given one and a given one or more second object labels and an area of union between the given one and the given one or more second object labels; and
generate the recommendation further based on the ratio.

19. The system of claim 11 , wherein the one or more processors are further configured to:
for each given one of the first object labels, determine an object label ratio corresponding to an area of intersection between a given one and a given one or more second object labels and an area of union between the given one and the given one or more second object labels; and
generate the recommendation further based on any determined ratios.

20. The system of claim 19 , wherein the one or more processors are further configured to:
determine an average object

Assignees

• Waymo Llc

Inventors

• Zhu Jiajun
• Urmson Christopher Paul
• Haehnel Dirk
• Fairfield Nathaniel
• Smith Russell Leigh

Classifications

• B60W2552/05

  Type of road, e.g. motorways, local streets, paved or unpaved roads · CPC title

• B60W2556/50

  of positioning data, e.g. GPS [Global Positioning System] data · CPC title

• B60W2555/60

  Traffic rules, e.g. speed limits or right of way · CPC title

• G01S17/86

  Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders · CPC title

• G06T7/74

  involving reference images or patches · CPC title