Automatic robotically steered camera for targeted high performance perception and vehicle control

1. A method of controlling a vehicle, comprising:

capturing a first image of a scene with a first sensor;

detecting an object in the first image;

determining whether to improve an accuracy of the detection based on a resolution of the first image, wherein determining whether to improve the accuracy comprises:

determining one or more probabilities that the object in the first image has a particular object type of one or more object types,

determining that each of the one or more probabilities is below a probability threshold, and

in response to determining that each of the one or more probabilities is below the probability threshold, determining that the accuracy should be improved;

in response to a determination to improve the accuracy:

directing a second sensor having a field of view smaller than the first sensor, and determining an aim point of the second sensor to generate a second image representing a location of the identified object and having a higher resolution than the resolution of the first image,

identifying the object based on the second image, and

controlling the vehicle based on the identified object.

2. The method of claim 1 , further comprising:

determining the object in the scene is a sign;

failing to recognize text within the sign based on a resolution of the sign within the first image; and

determining to improve the accuracy also based on the failure to recognize the text within the sign.

3. The method of claim 1 , further comprising determining a first probability that the object in the scene is a first type of object in a predefined group of object types based on the first image, and determining a second probability that the object in the scene is a second type of object in the predefined group of object types based on the first image, and updating the first probability and the second probability based on identifying the object based on the second image.

4. The method of claim 3 , wherein the predefined group of objects comprises at least one of: (i) a motorcycle object type, (ii) a pedestrian object type, (iii) a plastic bag object type, (iv) a concrete block object type, (v) a cyclist object type, (vi) a car object type, (vii) a truck object type, (viii) a bus object type, (ix) a temporary traffic control device object type, (x) an animal object type, (xi) a traffic light object type, or (xii) a sign object type.

5. The method of claim 1 , further comprising:

determining a region of interest based on the identified object;

determining the region of interest is larger than a field of view of the second sensor, wherein the second image excludes a first portion of the region of interest;

directing the second sensor to capture the first portion of the region of interest in a third image; and

generating a fourth image by stitching together the second and third images, wherein the identifying is based on the fourth image.

6. The method of claim 1 , wherein the second sensor is configured with an active illumination device that is synchronized to illuminate based on the second sensor capturing an image, the active illumination device configured with a field of view corresponding to the field of view of the second sensor.

7. The method of claim 1 , further comprising:

capturing another image with a third sensor; and

fusing the other image with the first image, wherein the detecting of the object is based on the fusing of the other image with the first image.

8. A system for controlling a vehicle, comprising:

hardware processing circuitry;

a hardware memory storing instructions that are executable by the hardware processing circuitry to configure the hardware processing circuitry to perform operations comprising:

capturing a first image of a scene with a first sensor;

detecting an object in the first image;

determining whether to improve an accuracy of the detection based on a resolution of the first image, wherein determining whether to improve the accuracy comprises:

determining one or more probabilities that the object in the first image has a particular object type of one or more object types,

determining that each of the one or more probabilities is below a probability threshold, and

in response to determining that each of the one or more probabilities is below the probability threshold, determining that the accuracy should be improved;

in response to a determination to improve the accuracy:

directing a second sensor having a field of view smaller than the first sensor, and determining an aim point of the second sensor to generate a second image representing a location of the identified object and having a higher resolution than the resolution of the first image,

identifying the object based on the second image, and

controlling the vehicle based on the identified object.

9. The system of claim 8 , the operations further comprising:

determining the object in the scene is a sign;

failing to recognize text within the sign based on a resolution of the sign within the first image; and

determining to improve the accuracy also based on the failure to recognize the text within the sign.

10. The system of claim 8 , the operations further comprising determining a first probability that the object in the scene is a first type of object in a predefined group of object types based on the first image, and determining a second probability that the object in the scene is a second type of object in the predefined group of object types based on the first image, and updating the first probability and the second probability based on identifying the object based on the second image.

11. The system of claim 10 , wherein the predefined group of objects comprises at least one of: (i) a motorcycle object type, (ii) a pedestrian object type, (iii) a plastic bag object type, (iv) a concrete block object type, (v) a cyclist object type, (vi) a car object type, (vii) a truck object type, (viii) a bus object type, (ix) a temporary traffic control device object type, (x) an animal object type, (xi) a traffic light object type, or (xii) a sign object type.

12. The system of claim 8 , the operations further comprising:

determining a region of interest based on the identified object;

determining the region of interest is larger than a field of view of the second sensor, wherein the second image excludes a first portion of the region of interest;

directing the second sensor to capture the first portion of the region of interest in a third image; and

generating a fourth image by stitching together the second and third images, wherein the identifying is based on the fourth image.

13. The system of claim 8 , wherein the second sensor is configured with an active illumination device that is synchronized to illuminate based on the second sensor capturing an image, the active illumination device configured with a field of view corresponding to the field of view of the second sensor.

14. The system of claim 8 , the operations further comprising:

capturing another image with a third sensor; and

fusing the other image with the first image, wherein the detecting of the object is based on the fusing of the other image with the first image.

15. The system of claim 8 , wherein the second sensor has a higher pixel density than the first sensor.

16. A method of controlling a vehicle, comprising:

capturing a first image of a scene with a first sensor;

detecting an object in the first image;

determining whether to improve an accuracy of the detection based on a resolution of the first image;

in response to a determination to improve the accuracy:

directing a second sensor having a field of view smaller than the first sensor, and determining an aim point of the second sensor to generate a second image representing a location of the identified object and having a higher resolution than the resolution of the first image,

identifying the object based on the second image,

controlling the vehicle based on the identified object;

determining a region of interest based on the identified object;

determining the region of interest is larger than a field of view of the second sensor, wherein the second image excludes a first portion of the region of interest;

directing the second sensor to capture the first portion of the region of interest in a third image; and

generating a fourth image by stitching together the second and third images, wherein the identifying is based on the fourth image.

17. A system for controlling a vehicle, comprising:

hardware processing circuitry;

a hardware memory storing instructions that are executable by the hardware processing circuitry to configure the hardware processing circuitry to perform operations comprising:

capturing a first image of a scene with a first sensor;

detecting an object in the first image;

determining whether to improve an accuracy of the detection based on a resolution of the first image;

in response to a determination to improve the accuracy:

directing a second sensor having a field of view smaller than the first sensor, and determining an aim point of the second sensor to generate a second image representing a location of the identified object and having a higher resolution than the resolution of the first image,

identifying the object based on the second image;

controlling the vehicle based on the identified object;

determining a region of interest based on the identified object;

determining the region of interest is larger than a field of view of the second sensor, wherein the second image excludes a first portion of the region of interest;

directing the second sensor to capture the first portion of the region of interest in a third image; and

generating a fourth image by stitching together the second and third images, wherein the identifying is based on the fourth image.