Encoding lidar scanned data for generating high definition maps for autonomous vehicles
Embodiments relate to methods for efficiently encoding sensor data captured by an autonomous vehicle and building a high definition map using the encoded sensor data. The sensor data can be LiDAR data which is expressed as multiple image representations. Image representations that include important LiDAR data undergo a lossless compression while image representations that include LiDAR data that is more error-tolerant undergo a lossy compression. Therefore, the compressed sensor data can be transmitted to an online system for building a high definition map. When building a high definition map, entities, such as road signs and road lines, are constructed such that when encoded and compressed, the high definition map consumes less storage space. The positions of entities are expressed in relation to a reference centerline in the high definition map. Therefore, each position of an entity can be expressed in fewer numerical digits in comparison to conventional methods.
1. A method of generating high definition map, comprising:
receiving, from one or more sensors mounted on a vehicle, sensor data describing an environment around the vehicle captured by the one or more sensors;
determining based on the received sensor data, a plurality of sensor orientation datasets;
constructing one or more image representations from the plurality of sensor orientation datasets, each image representation comprising a plurality of pixels, wherein each pixel stores sensor orientation data;
compressing each of the one or more image representations to generate compressed codes representing the plurality of sensor orientation datasets; and
transmitting the compressed codes for generation of a high definition map, the high definition map comprising a plurality of geographical regions, wherein the high definition map is sent to an autonomous vehicle for navigation.
2. The method of claim 1 , wherein the sensor orientation dataset comprises information describing a point, the information comprising, a pitch value, a yaw value, a range value, and an intensity value.
3. The method of claim 1 , wherein an image representation comprises pixel values representing one of the range value or the intensity value from the sensor orientation dataset.
4. The method of claim 3 , wherein each pixel location of the image representation corresponds to a combination of a pitch value and a yaw value of a sensor orientation dataset.
5. The method of claim 1 , wherein the one or more image representations comprise an image representation, wherein each pixel of the image representation represents a most significant byte of a range value of a point.
6. The method of claim 1 , wherein the one or more image representations comprise an image representation, wherein each pixel of the image representation represents a least significant byte of a range value of a point.
7. The method of claim 1 , wherein the one or more image representations comprise an image representation, wherein each pixel of the image representation represents a byte of an intensity value of a point.
8. The method of claim 1 , wherein the one or more image representations comprise a first image representation and a second image representation, wherein each pixel of the first image representation is a most significant byte of a range value of a sensor orientation dataset and each pixel of the second image representation is a least significant byte of the range value of a sensor orientation dataset, the method further comprising
encoding the first image representation using a lossless encoding process; and
encoding the second image representation using a lossy encoding process.
9. The method of claim 1 , wherein a number of rows of each of the one or more image representations is dependent on a number of lasers of the vehicle sensor.
10. The method of claim 1 , wherein a number of columns of each of the one or more image representations is dependent on a minimum angular resolution of the vehicle sensor.
11. A non-transitory computer readable storage medium comprising computer instructions that, when executed by a processor, cause the processor to perform steps comprising:
receiving, from one or more sensors mounted on a vehicle, sensor data describing an environment around the vehicle captured by the one or more sensors;
determining based on the received sensor data, a plurality of sensor orientation datasets;
constructing one or more image representations from the plurality of sensor orientation datasets, each image representation comprising a plurality of pixels, wherein each pixel stores sensor orientation data;
compressing each of the one or more image representations to generate compressed codes representing the plurality of sensor orientation datasets; and
transmitting the compressed codes for generation of a high definition map, the high definition map comprising a plurality of geographical regions, wherein the high definition map is sent to an autonomous vehicle for navigation.
12. The non-transitory computer readable storage medium of claim 11 , wherein the sensor orientation dataset comprises information describing a point, the information comprising, a pitch value, a yaw value, a range value, and an intensity value.
13. The non-transitory computer readable storage medium of claim 11 , wherein an image representation comprises pixel values representing one of the range value or the intensity value from the sensor orientation dataset.
14. The non-transitory computer readable storage medium of claim 13 , wherein each pixel location of the image representation corresponds to a combination of a pitch value and a yaw value of a sensor orientation dataset.
15. The non-transitory computer readable storage medium of claim 11 , wherein the one or more image representations comprise an image representation, wherein each pixel of the image representation represents a most significant byte of a range value of a point.
16. The non-transitory computer readable storage medium of claim 11 , wherein the one or more image representations comprise an image representation, wherein each pixel of the image representation represents a least significant byte of a range value of a point.
17. The non-transitory computer readable storage medium of claim 11 , wherein the one or more image representations comprise an image representation, wherein each pixel of the image representation represents a byte of an intensity value of a point.
18. The non-transitory computer readable storage medium of claim 11 , wherein the one or more image representations comprise a first image representation and a second image representation, wherein each pixel of the first image representation is a most significant byte of a range value of a sensor orientation dataset and each pixel of the second image representation is a least significant byte of the range value of a sensor orientation dataset, wherein the computer instructions further cause the processor to:
encoding the first image representation using a lossless encoding process; and
encoding the second image representation using a lossy encoding process.
19. The non-transitory computer readable storage medium of claim 11 , wherein a number of rows of each of the one or more image representations is dependent on a number of lasers of the vehicle sensor.
20. The non-transitory computer readable storage medium of claim 11 , wherein a number of columns of each of the one or more image representations is dependent on a minimum angular resolution of the vehicle sensor.
21. A computer-system comprising:
a processor; and
a non-transitory computer readable storage medium comprising computer instructions that, when executed by a processor, cause the processor to perform steps comprising:
receiving, from one or more sensors mounted on a vehicle, sensor data describing an environment around the vehicle captured by the one or more sensors;
determining based on the received sensor data, a plurality of sensor orientation datasets;
constructing one or more image representations from the plurality of sensor orientation datasets, each image representation comprising a plurality of pixels, wherein each pixel stores sensor orientation data;
compressing each of the one or more image representations to generate compressed codes representing the plurality of sensor orientation datasets; and
transmitting the compressed codes for generation of a high definition map, the high definition map comprising a plurality of geographical regions, wherein the high definition map is sent to an autonomous vehicle for navigation.
22. The computer-system of claim 21 , wherein the sensor orientation dataset comprises information describing a point, the information comprising, a pitch value, a yaw value, a range value, and an intensity value.
23. The computer-system of claim 21 , wherein an image representation comprises pixel values representing one of the range value or the intensity value from the sensor orientation dataset.
24. The computer-system of claim 23 , wherein each pixel location of the image representation corresponds to a combination of a pitch value and a yaw value of a sensor orientation dataset.
25. The computer-system of claim 21 , wherein the one or more image representations comprise an image representation, wherein each pixel of the image representation represents a most significant byte of a range value of a point.
26. The computer-system of claim 21 , wherein the one or more image representations comprise an image representation, wherein each pixel of the image representation represents a least significant byte of a range value of a point.
27. The computer-system of claim 21 , wherein the one or more image representations comprise an image representation, wherein each pixel of the image representation represents a byte of an intensity value of a point.
28. The computer-system of claim 21 , wherein the one or more image representations comprise a first image representation and a second image representation, wherein each pixel of the first image representation is a most significant byte of a range value of a sensor orientation dataset and each pixel of the second image representation is a least significant byte of the range value of a sensor orientation dataset, wherein the computer instructions further cause the processor to:
encoding the first image representation using a lossless encoding process; and
encoding the second image representation using a lossy encoding process.
29. The computer-system of claim 21 , wherein a number of rows of each of the one or more image representations is dependent on a number of lasers of the vehicle sensor.
30. The computer-system of claim 21 , wherein a number of columns of each of the one or more image representations is dependent on a minimum angular resolution of the vehicle sensor.