System and method for determining optimal lidar placement on autonomous vehicles
In one embodiment, example systems and methods related to a manner of optimizing LiDAR sensor placement on autonomous vehicles are provided. A range-of-interest is defined for the autonomous vehicle that includes the distances from which the autonomous vehicle is interested in collecting sensor data. The range-of-interest is segmented into multiple cubes of the same size. For each LiDAR sensor, a shape is determined based on information such as the number of lasers in each LiDAR sensor and the angle associated with each laser. An optimization problem is solved using the determined shape for each LiDAR sensor and the cubes of the range-of-interest to determine the locations to place each LiDAR sensor to maximize the number of cubes that are captured. The optimization problem may further determine the optimal pitch angle and roll angle to use for each LiDAR sensor to maximize the number of cubes that are captured.
1. A system for optimizing LiDAR sensor placement for an autonomous vehicle, comprising:
one or more processors; and
a memory communicably coupled to the one or more processors and storing:
a region of interest module including instructions that when executed by the one or more processors cause the one or more processors to:
receive information about an autonomous vehicle, wherein the information comprises a plurality of LiDAR sensor locations on the autonomous vehicle for a plurality of LiDAR sensors;
generate a range-of-interest for the autonomous vehicle, wherein the autonomous vehicle is centered in the range-of-interest; and
segment the range-of-interest into a plurality of units;
a shape module including instructions that when executed by the one or more processors cause the one or more processors to determine a shape for each LiDAR sensor of the plurality of LiDAR sensors based on information about each LiDAR sensor; and
a selection module including instructions that when executed by the one or more processors cause the one or more processors to select, for each LiDAR sensor of the plurality of LiDAR sensors, a LiDAR sensor location from the plurality of LiDAR sensor locations of the autonomous vehicle, wherein the selected LiDAR sensor locations maximize a number of units of the plurality of units that are captured by the shapes determined for each LiDAR sensor of the plurality of LiDAR sensors.
2. The system of claim 1 , wherein the information about a LiDAR sensor comprises a number of associated lasers and an angle associated with each laser.
3. The system of claim 2 , wherein generating the shape for the LiDAR sensor comprises generating a cone for each laser associated with the LiDAR sensor based on the angle associated with the laser.
4. The system of claim 1 , wherein the selection module further includes instructions to select, for each LiDAR sensor of the plurality of LiDAR sensors, a LiDAR sensor location from the plurality of LiDAR sensor locations of the autonomous vehicle and a pitch angle for the LiDAR sensor, wherein the selected LiDAR sensor locations and pitch angles maximize the number of units of the plurality of units that are captured by the shapes determined for each LiDAR sensor of the plurality of LiDAR sensors.
5. The system of claim 1 , wherein the selection module further includes instructions to select, for each LiDAR sensor of the plurality of LiDAR sensors, a LiDAR sensor location from the plurality of LiDAR sensor locations of the autonomous vehicle and a roll angle for the LiDAR sensor, wherein the selected LiDAR sensor locations and roll angles maximize the number of units of the plurality of units that are captured by the shapes determined for each LiDAR sensor of the plurality of LiDAR sensors.
6. The system of claim 1 , wherein the selection module further includes instructions to select, for each LiDAR sensor of the plurality of LiDAR sensors, a LiDAR sensor location from the plurality of LiDAR sensor locations of the autonomous vehicle, a roll angle for the LiDAR sensor, and a pitch angle for the LiDAR sensor, wherein the selected LiDAR sensor locations, roll angles, and pitch angles maximize the number of units of the plurality of units that are captured by the shapes determined for each LiDAR sensor.
7. The system of claim 1 , wherein the selection module maximizes the number of units of the plurality of units that are captured by the shapes determined for each LiDAR sensor of the plurality of LiDAR sensors using an object function.
8. A method for optimizing LiDAR sensor placement for an autonomous vehicle, the method comprising:
receiving information about an autonomous vehicle, wherein the information comprises a plurality of LiDAR sensor locations on the autonomous vehicle;
generating a range-of-interest for the autonomous vehicle, wherein the autonomous vehicle is centered in the range-of-interest;
segmenting the range-of-interest into a plurality of units;
receiving information about each LiDAR sensor of a plurality of LiDAR sensors, wherein the information about a LiDAR sensor comprises a number of lasers associated with the LiDAR sensor and an angle associated with each laser;
for each LiDAR sensor of the plurality of LiDAR sensors, determining a shape for the LiDAR sensor based on the information about the LiDAR sensor; and
selecting, for each LiDAR sensor of the plurality of LiDAR sensors, a LiDAR sensor location from the plurality of LiDAR sensor locations of the autonomous vehicle, wherein the selected LiDAR sensor locations maximize a number of units of the plurality of units that are captured by the shapes determined for each LiDAR sensor of the plurality of LiDAR sensors.
9. The method of claim 8 , wherein the received information about the autonomous vehicle further comprises dimensions of the autonomous vehicle.
10. The method of claim 9 , wherein the range-of-interest is generated based on the dimensions of the vehicle.
11. The method of claim 8 , wherein the plurality of units are cubes.
12. The method of claim 8 , wherein generating the shape for the LiDAR sensor comprises generating a cone for each laser associated with the LiDAR sensor based on the angle associated with the laser.
13. The method of claim 8 , further comprising placing the LiDAR sensors of the plurality of LiDAR sensors on the autonomous vehicle at the selected LiDAR sensor locations.
14. The method of claim 8 , further comprising selecting, for each LiDAR sensor of the plurality of LiDAR sensors, a LiDAR sensor location from the plurality of LiDAR sensor locations of the autonomous vehicle and a pitch angle for the LiDAR sensor, wherein the selected LiDAR sensor locations and pitch angles maximize the number of units of the plurality of units that are captured by the shapes determined for each LiDAR sensor of the plurality of LiDAR sensors.
15. The method of claim 8 , further comprising selecting, for each LiDAR sensor of the plurality of LiDAR sensors, a LiDAR sensor location from the plurality of LiDAR sensor locations of the autonomous vehicle and a roll angle for the LiDAR sensor, wherein the selected LiDAR sensor locations and roll angles maximize the number of units of the plurality of units that are captured by the shapes determined for each LiDAR sensor of the plurality of LiDAR sensors.
16. The method of claim 8 , further comprising selecting, for each LiDAR sensor of the plurality of LiDAR sensors, a LiDAR sensor location from the plurality of LiDAR sensor locations of the autonomous vehicle, a roll angle for the LiDAR sensor, and a pitch angle for the LiDAR sensor, wherein the selected LiDAR sensor locations, roll angles, and pitch angles maximize the number of units of the plurality of units that are captured by the shapes determined for each LiDAR sensor.
17. A non-transitory computer-readable medium for optimizing LiDAR sensor placement for an autonomous vehicle and including instructions that when executed by one or more processors cause the one or more processors to:
receive information about an autonomous vehicle, wherein the information comprises a plurality of LiDAR sensor locations on the autonomous vehicle;
receive a range-of-interest for the autonomous vehicle, wherein the autonomous vehicle is centered in the range-of-interest and the range-of-interest is segmented into a plurality of units;
receive information about each LiDAR sensor of a plurality of LiDAR sensors, wherein the information about a LiDAR sensor comprises a number of lasers associated with the LiDAR and an angle associated with each laser;
for each LiDAR sensor of the plurality of LiDAR, determine a shape for the LiDAR sensor based on the information about the LiDAR sensor; and
selecting, for each LiDAR sensor of the plurality of LiDAR sensors, a LiDAR sensor location from the plurality of LiDAR sensor locations of the autonomous vehicle, a roll angle for the LiDAR sensor, and a pitch angle for the LiDAR sensor, wherein the selected LiDAR sensor locations, roll angles, and pitch angles maximize a number of units of the plurality of units that are captured by the shapes determined for each LiDAR sensor of the plurality of LiDAR sensors.
18. The non-transitory computer-readable medium of claim 17 , further comprising placing the LiDAR sensors of the plurality of LiDAR sensors on the autonomous vehicle at the selected LiDAR sensor locations.
19. The non-transitory computer-readable medium of claim 17 , wherein generating the shape for the LiDAR sensor comprises generating a cone for each laser associated with the LiDAR sensor based on the angle associated with the laser.
20. The non-transitory computer-readable medium of claim 17 , wherein there are at least two LiDAR sensors in the plurality of LiDAR sensors.