Bus detection for an autonomous vehicle
Methods and systems are provided that may allow an autonomous vehicle to discern a school bus from image data. An example method may include receiving image data indicative of a vehicles operating in an environment. The image data may depict sizes of the vehicles. The method may also include, based on relative sizes of the vehicles, determining a vehicle that is larger in size as compared the other vehicles. The method may additionally include comparing a size of the determined vehicle to a size of a school bus and based on the size of vehicle being within a threshold size of the school bus, comparing a color of the vehicle to a color of the school bus. The method may further include based on the vehicle being substantially the same color as the school bus, determining that the vehicle is representative of the school bus.
1. A method comprising:
using, by one or more processors of a computing device, image data indicative of one or more vehicles in an environment to select a vehicle from the one or more vehicles to be a candidate school bus;
based on the selection of the vehicle to be the candidate school bus, the one or more processors making a comparison of one or more features of the vehicle to one or more known features of a school bus;
based on the comparison, determining a confidence level representing a likelihood that the vehicle is a school bus; and
based on the confidence level exceeding a predetermined threshold, the one or more processors determining that the vehicle is a school bus.
2. The method of claim 1 , wherein the one or more features of the vehicle include an estimated size of the vehicle,
wherein the one or more known features of a school bus include a known size of a school bus,
wherein the comparison includes a comparison of the estimated size of the vehicle to the known size of a school bus, and
wherein determining the confidence level based on the comparison comprises increasing the confidence level based at least in part on the comparison indicating that the estimated size of the vehicle is within a threshold size of the size of the school bus.
3. The method of claim 1 , wherein the one or more features of the vehicle include a color of the vehicle,
wherein the one or more known features of a school bus include a known color of a school bus,
wherein the comparison includes a comparison of the color of the vehicle to the known color of a school bus, and
wherein determining the confidence level based on the comparison comprises increasing the confidence level based at least in part on the comparison indicating that the color of the vehicle is substantially the same color as the known color of a school bus.
4. The method of claim 1 , wherein the one or more features of the vehicle include a color of the vehicle,
wherein the one or more known features of a school bus include a known substantially orange color of a school bus,
wherein making the comparison of the one or more features of the vehicle to the one or more known features of a school bus comprises:
determining a portion of the image data that depicts the vehicle;
dividing the portion of the image data into a plurality of sections;
for respective sections of the plurality of sections, determining a value of difference between a color of the respective section and the known substantially orange color; and
determining a number of sections of the respective sections of which the value of difference is less than a difference threshold, and
wherein determining the confidence level comprises increasing the confidence level based at least in part on the number of sections of the respective sections being greater than a number threshold.
5. The method of claim 1 , further comprising:
the one or more processors determining a portion of the image data that depicts the vehicle; and
the one or more processors determining that the portion of the image data includes data indicative of a stop-sign,
wherein determining the confidence level comprises increasing the confidence level based at least in part on the portion of the image data including data indicative of the stop-sign.
6. The method of claim 1 , further comprising:
the one or more processors determining a portion of the image data that depicts the vehicle; and
the one or more processors determining that the portion of the image data includes data indicative of a school-sign,
wherein determining the confidence level comprises increasing the confidence level based at least in part on the portion of the image data including data indicative of the school-sign.
7. The method of claim 1 , further comprising:
the one or more processors receiving operational data indicative of an operation of the vehicle in the environment,
wherein determining that the vehicle is a school bus is further based on the operational data.
8. The method of claim 7 , wherein the operational data indicative of the operation of the vehicle in the environment comprises data indicative of a stop pattern of the vehicle, and
wherein determining that the vehicle is a school bus based on the operational data comprises determining that the vehicle is a school bus based on the data indicative of the stop pattern of the vehicle.
9. The method of claim 8 , wherein the data indicative of the stop pattern of the vehicle includes data indicative of the vehicle operating a component of the vehicle to extend from a side portion of the vehicle as the vehicle stops.
10. The method of claim 1 , wherein the computing device resides in an autonomous vehicle operating in the environment, the method further comprising:
based on the vehicle being a school bus, the one or more processors providing instructions to control the autonomous vehicle in the environment.
11. An autonomous vehicle comprising:
a computer system configured to:
use image data indicative of one or more vehicles in an environment to select a vehicle from the one or more vehicles to be a candidate school bus;
based on the selection of the vehicle to be the candidate school bus, make a comparison of one or more features of the vehicle to one or more known features of a school bus;
based on the comparison, determine a confidence level representing a likelihood that the vehicle is a school bus; and
based on the confidence level exceeding a predetermined threshold, determine that the vehicle is a school bus.
12. The autonomous vehicle of claim 11 , further comprising at least one sensor configured to sense the environment of the autonomous vehicle, wherein the computer system is further configured to:
before using the image data, use the at least one sensor to determine the image data.
13. The autonomous vehicle of claim 11 , wherein the computer system is further configured to:
determine a portion of the image data that depicts the vehicle; and
determine that the portion of the image data includes data indicative of one or both of a stop-sign and a school-sign,
wherein determining the confidence level comprises increasing the confidence level based at least in part on the portion of the image data including the data indicative of one or both of the stop-sign and the school-sign.
14. The autonomous vehicle of claim 11 , wherein the computer system is further configured to:
receive operational data indicative of a stop pattern of the vehicle,
wherein the stop pattern includes data indicative of the vehicle operating a component of the vehicle to extend from a side portion of the vehicle as the vehicle stops,
wherein determining that the vehicle is a school bus is further based on the operational data.
15. The autonomous vehicle of claim 11 , wherein using the image data indicative of the one or more vehicles in the environment to select the vehicle from the one or more vehicles to be the candidate school bus comprises:
using the image data to determine that the vehicle substantially fills a predetermined bounding box without exceeding the predetermined bounding box; and
based on the determination that the vehicle substantially fills the predetermined bounding box without exceeding the predetermined bounding box, selecting the vehicle to be the candidate school bus.
16. The autonomous vehicle of claim 11 , wherein the one or more features of the vehicle include an estimated size of the vehicle,
wherein the one or more known features of a school bus include a known size of a school bus,
wherein the comparison includes a comparison of the estimated size of the vehicle to the known size of a school bus, and
wherein determining the confidence level based on the comparison comprises increasing the confidence level based at least in part on the comparison indicating that the estimated size of the vehicle is within a threshold size of the size of the school bus.
17. The autonomous vehicle of claim 11 , wherein the one or more features of the vehicle include a color of the vehicle,
wherein the one or more known features of a school bus include a known color of a school bus,
wherein the comparison includes a comparison of the color of the vehicle to the known color of a school bus, and
wherein determining the confidence level based on the comparison comprises increasing the confidence level based at least in part on the comparison indicating that the color of the vehicle is substantially the same color as the known color of a school bus.
18. A non-transitory computer readable medium having stored therein instructions, that when executed by a computer system, cause the computer system to perform functions comprising:
using image data indicative of one or more vehicles in an environment to select a vehicle from the one or more vehicles to be a candidate school bus;
based on the selection of the vehicle to be the candidate school bus, making a comparison of one or more features of the vehicle to one or more known features of a school bus;
based on the comparison, determining a confidence level representing a likelihood that the vehicle is a school bus; and
based on the confidence level exceeding a predetermined threshold, determining that the vehicle is a school bus.
19. The non-transitory computer readable medium of claim 18 , wherein the one or more features of the vehicle include an estimated size of the vehicle and a color of the vehicle,
wherein the one or more known features of a school bus include a known size of a school bus and a known color of a school bus,
wherein making the comparison of the one or more features of the vehicle to the one or more known features of a school bus comprises making a first comparison of the estimated size of the vehicle to the known size of a school bus and further comprises making a second comparison of the color of the vehicle to the known color of a school bus,
wherein determining the confidence level based on the comparison comprises increasing the confidence level to exceed the predetermined threshold based on the first comparison indicating that the estimated size of the vehicle is within a threshold size of the size of the school bus and further based on the second comparison indicating that the color of the vehicle is substantially the same color as the known color of a school bus.
20. The non-transitory computer readable medium of claim 18 , wherein using the image data indicative of the one or more vehicles in the environment to select the vehicle from the one or more vehicles to be the candidate school bus comprises:
using the image data to determine that the vehicle substantially fills a predetermined bounding box without exceeding the predetermined bounding box; and
based on the determination that the vehicle substantially fills the predetermined bounding box without exceeding the predetermined bounding box, selecting the vehicle to be the candidate school bus.