Constrained mobility mapping
A method of constrained mobility mapping includes receiving from at least one sensor of a robot at least one original set of sensor data and a current set of sensor data. Here, each of the at least one original set of sensor data and the current set of sensor data corresponds to an environment about the robot. The method further includes generating a voxel map including a plurality of voxels based on the at least one original set of sensor data. The plurality of voxels includes at least one ground voxel and at least one obstacle voxel. The method also includes generating a spherical depth map based on the current set of sensor data and determining that a change has occurred to an obstacle represented by the voxel map based on a comparison between the voxel map and the spherical depth map. The method additional includes updating the voxel map to reflect the change to the obstacle.
1. A method comprising:
obtaining, at data processing hardware, a voxel map generated based on a first set of sensor data, wherein the voxel map indicates a height associated with a particular location;
obtaining, at the data processing hardware, a depth map generated based on a second set of sensor data, wherein the voxel map represents the first set of sensor data in a first format and the depth map represents the second set of sensor data in a second format that is different from the first format, wherein the depth map indicates a range of heights associated with the particular location;
performing, by the data processing hardware, a comparison of the height indicated by the voxel map to the range of heights indicated by the depth map based on the first format and the second format;
identifying, by the data processing hardware, a modification of an object of the voxel map based on the comparison of the height indicated by the voxel map to the range of heights indicated by the depth map;
updating, by the data processing hardware, the voxel map based on the modification of the object of the voxel map to obtain an updated voxel map; and
instructing, by the data processing hardware, navigation by a robot according to the updated voxel map.
2. The method of claim 1 , wherein the data processing hardware comprises data processing hardware of the robot.
3. The method of claim 1 , further comprising obtaining at least one of the first set of sensor data or the second set of sensor data from at least one sensor.
4. The method of claim 1 , wherein the first set of sensor data is an original set of sensor data and the second set of sensor data is a current set of sensor data.
5. The method of claim 1 , wherein the first set of sensor data is received by the data processing hardware prior to the data processing hardware receiving the second set of sensor data.
6. The method of claim 1 , wherein the comparison of the height indicated by the voxel map to the range of heights indicated by the depth map comprises a comparison of one or more columns of the voxel map to one or more columns of the depth map.
7. The method of claim 1 , wherein the depth map comprises a spherical representation of an environment of the robot.
8. The method of claim 1 , wherein the voxel map comprises a representation of an environment of the robot as a plurality of voxels.
9. The method of claim 1 , wherein obtaining the voxel map comprises generating the voxel map, and wherein obtaining the depth map comprises generating the depth map.
10. The method of claim 1 , wherein instructing navigation by the robot according to the updated voxel map comprises:
identifying one or more locations for placement of a foot of the robot based on the updated voxel map; and
instructing movement by the robot according to the one or more locations.
11. The method of claim 1 , wherein the robot is a legged robot, and wherein the updated voxel map indicates one or more objects that may interfere with movement of the legged robot.
12. The method of claim 1 , wherein updating the voxel map based on the modification of the object of the voxel map comprises removing, from the voxel map, one or more voxels corresponding to the object.
13. The method of claim 1 , wherein the depth map comprises a representation of the second set of sensor data, the representation comprising one or more structures defined by two or more points of the second set of sensor data.
14. The method of claim 1 , wherein the voxel map comprises a three-dimensional grid.
15. The method of claim 1 , wherein the robot is a legged robot, and wherein the updated voxel map indicates one or more objects that may interfere with movement of the legged robot.
16. The method of claim 1 , further comprising:
identifying one or more first voxels based on identifying the modification of the object of the voxel map; and
identifying, using one or more heuristics, one or more second voxels associated with the object,
wherein updating the voxel map comprises:
removing the one or more first voxels and the one or more second voxels from voxel map.
17. Non-transitory computer-readable media including computer-executable instructions that, when executed by data processing hardware of a computing system, cause the computing system to:
obtain a voxel map generated based on a first set of sensor data, wherein the voxel map indicates a height associated with a particular location;
obtain a depth map generated based on a second set of sensor data, wherein the voxel map represents the first set of sensor data in a first format and the depth map represents the second set of sensor data in a second format that is different from the first format, wherein the depth map indicates a range of heights associated with the particular location;
perform a comparison of the height indicated by the voxel map to the range of heights indicated by the depth map based on the first format and the second format;
identify a modification of an object of the voxel map based on the comparison of the height indicated by the voxel map to the range of heights indicated by the depth map;
update the voxel map based on the modification of the object of the voxel map to obtain an updated voxel map; and
instruct navigation by a robot according to the updated voxel map.
18. The non-transitory computer-readable media of claim 17 , wherein the computing system comprises a computing system of the robot.
19. The non-transitory computer-readable media of claim 17 , wherein execution of the computer-executable instructions by the data processing hardware of the computing system, further causes the computing system to obtain at least one of the first set of sensor data or the second set of sensor data from at least one sensor.
20. The non-transitory computer-readable media of claim 17 , wherein the first set of sensor data is an original set of sensor data and the second set of sensor data is a current set of sensor data.
21. A computing system comprising:
memory; and
one or more processing devices coupled to the memory and configured to:
obtain a voxel map generated based on a first set of sensor data, wherein the voxel map indicates a height associated with a particular location;
obtain a depth map generated based on a second set of sensor data, wherein the voxel map represents the first set of sensor data in a first format and the depth map represents the second set of sensor data in a second format that is different from the first format, wherein the depth map indicates a range of heights associated with the particular location;
perform a comparison of the height indicated by the voxel map to the range of heights indicated by the depth map based on the first format and the second format;
identify a modification of an object of the voxel map based on the comparison of the height indicated by the voxel map to the range of heights indicated by the depth map;
update the voxel map based on the modification of the object of the voxel map to obtain an updated voxel map; and
instruct navigation by a robot according to the updated voxel map.
22. The computing system of claim 21 , wherein the computing system comprises a computing system of the robot.