Systems and methods for vehicle guidance using depth map generation
This disclosure relates to systems and methods for vehicle guidance. Stereo images may be obtained at different times using a stereo image sensor. A depth image may be determined based on an earlier obtained pair of stereo images. The depth image may be refined based on predictions of an earlier stereo image and a later obtained stereo image. Depth information for an environment around a vehicle may be obtained. The depth information may characterize distances between the vehicle and the environment around the vehicle. A spherical depth map may be generated from the depth information. Maneuver controls for the vehicle may be provided based on the spherical depth map.
1. An aerial vehicle, comprising:
a sensor generating information from an environment around the aerial vehicle; and
a processing apparatus coupled to a memory storing instructions that when executed causes the processing apparatus to:
generate a depth map based on the environment, the depth map including maps cells corresponding to longitude angles and to latitude angles, wherein the depth map represents distances to closest surfaces of the environment around the aerial vehicle, and wherein the depth map is a function of longitude angles and latitude angles; and
provide controls for the aerial vehicle based on the depth map; and
wherein to generate the depth map comprises to:
determine distance values for the map cells based on the distances to the closest surfaces of the environment around the aerial vehicle, wherein a first distance value for a first map cell is determined based on a first distance to a first closest surface of the environment around the aerial vehicle at a first longitude angle and a first latitude angle.
2. The aerial vehicle of claim 1 , wherein the depth map includes a dead zone in a polar region, and the distance values are not determined for the map cells in the dead zone.
3. The aerial vehicle of claim 1 , wherein the determination of the distance values disregards distances to surfaces of the environment around the aerial vehicle at individual longitude angles and individual latitude angles that are greater than the distances to the closest surfaces of the environment around the aerial vehicle at the individual longitude angles and the individual latitude angles.
4. The aerial vehicle of claim 3 , wherein the distance values correspond only to the closest surfaces of the environment around the aerial vehicle at the individual longitude angles and the individual latitude angles.
5. The aerial vehicle of claim 1 , wherein the sensor is a stereo image sensor comprising:
a first image sensor configured to generate first visual output signals providing first visual information within a first field of view of the first image sensor; and
a second image sensor configured to generate second visual output signals providing second visual information within a second field of view of the second image sensor, wherein the depth map is generated by comparing the first visual information with the second visual information.
6. The aerial vehicle of claim 1 , wherein the processing apparatus comprises a hardware-implemented processor and a software-implemented processor.
7. The aerial vehicle of claim 6 , wherein the hardware-implemented processor is located remotely from the software-implemented processor.
8. The aerial vehicle of claim 1 , further including instructions that when executed causes the processing apparatus to:
responsive to detecting the aerial vehicle traveling a threshold distance, transform the depth map such that the center of the depth map coincides with a location of the aerial vehicle.
9. A method, comprising:
generating a depth map based on an environment of an unmanned aerial vehicle, the depth map including maps cells corresponding to longitude angles and to latitude angles, wherein the depth map represents distances to closest surfaces of the environment around the aerial vehicle, wherein the depth map is a function of longitude angles and latitude angles; and
providing controls for the unmanned aerial vehicle based on the depth map;
wherein to generate the depth map comprises to:
determine distance values for the map cells based on the distances to the closest surfaces of the environment around the aerial vehicle, wherein a first distance value for a first map cell is determined based on a first distance to a first closest surface of the environment around the aerial vehicle at a first longitude angle and a first latitude angle.
10. The method of claim 9 , wherein the depth map includes a dead zone in a polar region, and the distance values are not determined for the map cells in the dead zone.
11. The method of claim 9 , wherein the determination of the distance values disregards distances to surfaces of the environment around the aerial vehicle at individual longitude angles and individual latitude angles that are greater than the distances to the closest surfaces of the environment around the aerial vehicle at the individual longitude angles and the individual latitude angles.
12. The method of claim 9 , comprising:
generating first visual output signals conveying first visual information within a first field of view of a first image sensor of the unmanned aerial vehicle; and
generating second visual output signals conveying second visual information within a second field of view of a second image sensor of the unmanned aerial vehicle, wherein the depth map is generated by comparing the first visual information with the second visual information.
13. The method of claim 9 , comprising:
responsive to detecting the unmanned aerial vehicle traveling a threshold distance, transforming the depth map such that the center of the depth map coincides with a location of the unmanned aerial vehicle.
14. A non-transitory computer-readable medium including instructions that, when executed by one or more processors, cause the one or more processors to:
generate visual information from an environment around an unmanned aerial vehicle;
generate a depth map based on the environment, the depth map including maps cells corresponding to longitude angles and to latitude angles, wherein the depth map represents distances to closest surfaces of the environment around the aerial vehicle, wherein the depth map is a function of longitude angles and latitude angles;
providing controls for the unmanned aerial vehicle based on the depth map; and
determine distance values for the map cells based on the distances to the closest surfaces of the environment around the aerial vehicle, wherein a first distance value for a first map cell is determined based on a first distance to a first closest surface of the environment around the aerial vehicle at a first longitude angle and a first latitude angle.
15. The non-transitory computer-readable medium of claim 14 , wherein the depth map includes a dead zone in a polar region, and the distance values are not determined for the map cells in the dead zone.
16. The non-transitory computer-readable medium of claim 14 , wherein the determination of the distance values disregards distances to surfaces of the environment around the aerial vehicle at individual longitude angles and individual latitude angles that are greater than the distances to the closest surfaces of the environment around the aerial vehicle at the individual longitude angles and the individual latitude angles.
17. The non-transitory computer-readable medium of claim 16 , wherein the distance values correspond only to the closest surfaces of the environment around the aerial vehicle at the individual longitude angles and the individual latitude angles.
18. The non-transitory computer-readable medium of claim 14 , wherein the unmanned aerial vehicle includes a sensor that is a stereo image sensor comprising:
a first image sensor configured to generate first visual output signals providing first visual information within a first field of view of the first image sensor; and
a second image sensor configured to generate second visual output signals providing second visual information within a second field of view of the second image sensor, wherein the depth map is generated by comparing the first visual information with the second visual information.
19. The non-transitory computer-readable medium of claim 14 , wherein the one or more processors comprise a hardware-implemented processor and a software-implemented processor; and
wherein the hardware-implemented processor is located remotely from the software-implemented processor.
20. The non-transitory computer-readable medium of claim 14 , further including instructions that when executed causes the processing apparatus to:
responsive to detecting the aerial vehicle traveling a threshold distance, transform the depth map such that the center of the depth map coincides with a location of the aerial vehicle.