Automatic robotically steered sensor for targeted high performance perception and vehicle control
Disclosed are methods, systems, and non-transitory computer readable media that control an autonomous vehicle via at least two sensors. One aspect includes capturing an image of a scene ahead of the vehicle with a first sensor, identifying an object in the scene at a confidence level based on the image, determining the confidence level of the identifying is below a threshold, in response to the confidence level being below the threshold, directing a second sensor having a field of view smaller than the first sensor to generate a second image including a location of the identified object, further identifying the object in the scene based on the second image, controlling the vehicle based on the further identification of the object.
1. A method of controlling a vehicle, the method comprising:
capturing a first image with a first sensor, wherein the first image includes an object;
forecasting motion of the object in the first image;
determining a capture location of the object for obtaining a second image of the object by a second sensor, wherein the capture location of the object is based on the forecasting;
directing the second sensor to capture the second image of the object at the capture location, wherein the second image has a different field of view than the first image;
identifying the object based on the second image; and
controlling the vehicle based on the identifying the object.
2. The method of claim 1 , further comprising:
determining a movement vector of the object, wherein the movement vector is indicative of the motion of the object across a plurality of images captured by the first sensor.
3. The method of claim 1 , further comprising:
determining a delay in positioning the second sensor based on a current position of the second sensor and the motion of the object; and
determining the capture location of the object based on the motion of the object and the delay.
4. The method of claim 1 , wherein the capture location is based on a position of the vehicle and a time when the first image is captured.
5. The method of claim 1 , wherein directing the second sensor to capture the second image of the object at the capture location further comprises:
determining a distance from a current position of the second sensor to a capture position of the second sensor, wherein the capture position is indicative of the capture location; and
adjusting a zoom of the second sensor based on the distance from the current position of the second sensor to the capture position of the second sensor.
6. The method of claim 5 , further comprising:
determining a vertical difference between the current position of the second sensor and the capture position of the second sensor; and
adjusting a pitch of the second sensor, wherein adjusting the pitch compensates for the vertical difference between the current position of the second sensor and the capture position of the second sensor.
7. The method of claim 5 , further comprising:
determining a horizontal difference between the current position of the second sensor and the capture position of the second sensor; and
adjusting a yaw of the second sensor, wherein adjusting the yaw compensates for the horizonal difference between the current position of the second sensor and the capture position of the second sensor.
8. The method of claim 1 , wherein forecasting the motion of the object comprises determining a direction and a speed of the object.
9. The method of claim 1 , wherein the second image has a higher resolution than the first image.
10. A computing system for controlling a vehicle, comprising:
one or more hardware processors;
one or more tangible, non-transitory, computer readable media that store instructions executable by the one or more hardware processors to cause the one or more hardware processors to perform operations, the operations comprising:
capturing a first image with a first sensor, wherein the first image includes an object;
forecasting motion of the object in the first image;
determining a capture location of the object for obtaining a second image of the object by a second sensor, wherein the capture location of the object is based on the forecasting;
directing the second sensor to capture the second image of the object at the capture location, wherein the second image has a different field of view than the first image;
identifying the object based on the second image; and
controlling the vehicle based on the identifying the object.
11. The computing system of claim 10 , the operations further comprising:
determining a movement vector of the object, wherein the movement vector is indicative of the motion of the object between a plurality of images captured by the first sensor.
12. The computing system of claim 10 , the operations further comprising:
determining a delay in positioning the second sensor based on a current position of the second sensor and the motion of the object; and
determining the capture location of the object based on the motion of the object and the delay.
13. The computing system of claim 10 , wherein the second sensor includes one or more positioning motors used to traverse the second sensor at a traverse rate in a horizonal dimension and a traverse rate in a vertical dimension.
14. The computing system of claim 10 , wherein directing the second sensor to capture the second image of the object at the capture location further comprise:
determining a distance from a current position of the second sensor to a capture position of the second sensor, wherein the capture position is indicative of the capture location; and
adjusting a zoom of the second sensor based on the distance from the current position of the second sensor to the capture position of the second sensor.
15. The computing system of claim 10 , wherein forecasting the motion of the object comprises determining a direction and a speed of the object.
16. The computing system of claim 10 , wherein the second image has a higher resolution than the first image.
17. A vehicle comprising:
vehicle controls for controlling the vehicle;
a first sensor and a second sensor;
one or more processors;
a memory storing instructions that are executable by the one or more processors to cause the one or more processors to:
capture a first image with the first sensor, wherein the first image includes an object;
forecasting motion of the object in the first image;
determining a capture location of the object for obtaining a second image of the object by a second sensor, wherein the capture location of the object is based on the forecasting;
directing the second sensor to capture the second image of the object at the capture location, wherein the second image has a different field of view than the first image;
identifying the object based on the second image; and
controlling the vehicle based on the identifying the object.
18. The vehicle of claim 17 , wherein the instructions further cause the one or more processors to:
determine a movement vector of the Object, wherein the movement vector is indicative of the motion of the object between a plurality of images captured by the first sensor.
19. The vehicle of claim 17 , wherein the instructions further cause the one or more processors to:
determine a delay in positioning the second sensor based on a current position of the second sensor and the motion of the object; and
determine the capture location of the object based on the motion of the object and the delay.
20. The vehicle of claim 17 , wherein the second sensor includes one or more positioning motors used to traverse the second sensor at a traverse rate in a horizonal dimension and a traverse rate in a vertical dimension.