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 comprising:
generating one or more image representations based at least on sensor data corresponding to at least a portion of an environment and obtained using one or more sensors, respective sizes of the one or more image representations being based at least on one or more properties corresponding to at least one of the one or more sensors; and
compressing the one or more image representations to generate one or more compressed image representations, wherein the one or more compressed image representations are used for generating a map that is used by a vehicle for navigation.
2. The method of claim 1 , wherein the compressing of the one or more image representations includes generating one or more corresponding compression codes that respectively represent the one or more image representations.
3. The method of claim 1 , wherein the one or more image representations are generated based at least on sensor orientation data that is determined based at least on the sensor data.
4. The method of claim 3 , wherein the sensor orientation data includes one or more of: a pitch value, a yaw value, a range value, or an intensity value.
5. The method of claim 1 , wherein the one or more image representations respectively include a plurality of pixels in which one or more individual pixels represent a respective location within the environment.
6. The method of claim 5 , wherein one or more respective positions of the one or more individual pixels in the one or more image representations respectively correspond to one or more of a pitch value or a yaw value corresponding to the sensor data.
7. The method of claim 5 , wherein one or more respective values corresponding to the one or more individual pixels correspond to one or more respective sensor data values.
8. The method of claim 7 , wherein the one or more respective sensor (Original) data values correspond to one or more of a range value or an intensity value corresponding to Light Detection And Ranging (LiDAR) data.
9. The method of claim 1 , further comprising removing one or more pixels from the one or more image representations prior to the compressing of the one or more image representations, the removing of the one or more pixels being based at least on the one or more pixels not being populated with sensor data.
10. The method of claim 9 , further comprising generating a bitmap that indicates removal of the one or more pixels.
11. The method of claim 1 , wherein the one or more properties include one or more of:
a number of lasers of the one or more sensors;
a field of view of the one or more sensors; or
a resolution of the one or more sensors.
12. A system comprising:
one or more processing units to perform operations comprising:
generating one or more image representations based at least on sensor data corresponding to at least a portion of an environment, respective sizes of the one or more image representations being based at least on one or more properties corresponding to one or more sensors used to obtain the sensor data;
compressing the one or more image representations to generate one or more compressed image representations; and
transmitting the one or more compressed image representations for generation of a map that is used by a vehicle for navigation.
13. The system of claim 12 , wherein the map is used by a vehicle for navigation.
14. The system of claim 12 , wherein the one or more image representations are generated based at least on sensor orientation data that is determined based at least on the sensor data.
15. The system of claim 12 , wherein the one or more image representations respectively include a plurality of pixels in which one or more individual pixels represent a respective location within the environment.
16. The system of claim 15 , wherein one or more respective positions of the one or more individual pixels in the one or more image representations respectively correspond to one or more of: a pitch value or a yaw value corresponding to the sensor data.
17. A processor comprising processing circuitry to perform operations comprising:
receiving one or more compressed image representations respectively corresponding to one or more uncompressed image representations that are generated based at least on sensor data corresponding to at least a portion of an environment, respective sizes of the one or more image representations being based at least on one or more properties corresponding to one or more sensors used to obtain the sensor data; and
generating a map based at least on the compressed image representations, wherein the map is used by a vehicle for navigation.
18. The processor of claim 17 , wherein the one or more properties include one or more of:
a number of lasers of the one or more sensors;
a field of view of the one or more sensors; or
a resolution of the one or more sensors.
19. The system of claim 12 , wherein the one or more properties include one or more of:
a number of lasers of the one or more sensors;
a field of view of the one or more sensors; or
a resolution of the one or more sensors.
20. The system of claim 12 , wherein the operations further comprise removing one or more pixels from the one or more image representations prior to the compressing of the one or more image representations, the removing of the one or more pixels being based at least on the one or more pixels not being populated with sensor data.