System and method for validation of lidar-generated horizon estimation
A system and method are described for assessing accuracy of a virtual horizon generated by a vehicle-based lidar system as the vehicle traverses a route of travel. Travel route topology data is obtained by a vehicle vehicle-mounted global positioning system-assisted inertial measurement unit (GPS/IMU) data. A reference virtual horizon at points along the route of travel is obtained from the GPS/IMU data and is compared with the virtual horizon generated by the lidar system for the same points to assess the accuracy of the virtual horizon generated by the lidar system.
1. A method of determining accuracy of a horizon determination made by a vehicle-mounted lidar system comprising:
recording, with a vehicle-mounted global positioning system-assisted inertial measurement unit (GPS/IMU) system, vehicle position and vehicle orientation data at particular points in time as the vehicle traverses a route;
generating a route topology based on the vehicle position and the vehicle orientation data;
recording a first horizon estimation by the lidar system corresponding to a first horizon angle relative to a neutral angle for at least one of the particular points in time;
determining a relative lidar system position and a relative lidar system orientation based on a relative location of the lidar system with respect to the GPS/IMU system;
calculating a second horizon estimation for the lidar system corresponding to a second horizon angle relative to a neutral angle for the at least one of the particular points in time based on the route topology and the relative lidar system position and relative lidar system orientation; and
assessing a horizon estimation accuracy of the lidar system that is able to generate the first horizon estimation faster than utilizing the GPS/IMU system to augment the calculation of the second horizon estimation including by comparing for the at least one of the particular points in time the first horizon estimation including the first horizon angle with the second horizon estimation including the second horizon angle to generate a horizon accuracy metric.
2. The method of claim 1 , wherein: each of the particular points in time is associated with a particular frame of data captured by the lidar system.
3. The method of claim 1 , wherein the route topology is based on the location of the lidar system on the vehicle.
4. The method of claim 1 , wherein the route topology is a road surface route topology and the generation of the road surface route topology is further based on a location of the GPS/IMU system on the vehicle relative to the road surface.
5. The method of claim 1 , wherein the GPS/IMU system vehicle position and vehicle orientation data is further based on real time kinematic (RTK) system correction data.
6. The method of claim 1 , wherein the GPS/IMU system and the lidar system share a common time signal.
7. The method of claim 1 , wherein each first horizon estimation comprises a first vertical angle corresponding to a first distance on a road surface ahead of the vehicle and each second horizon estimation comprises a second vertical angle corresponding to a second distance on the road surface ahead of the vehicle.
8. The method of claim 7 wherein the first and second distances on the road surface are located between 60 and 120 meters ahead of the vehicle.
9. The method of claim 7 wherein the first and second distances on the road surface are related to at least one of road speed limits, and road traffic levels.
10. The method of claim 1 , wherein the particular points in time occur at fixed time intervals.
11. The method of claim 1 , wherein the particular points in time occur at fixed distance intervals travelled by the vehicle.
12. The method of claim 1 , wherein the particular points in time correspond to distance traveled by the vehicle and to the route topology.
13. The method of claim 12 , wherein the particular points in time occur at shorter distance intervals for irregular road surface topology than distance intervals for flat road surface topology.
14. The method of claim 1 , wherein the vehicle orientation data is based in part on vehicle-mounted gyroscope data.
15. A method of determining accuracy of a horizon determination made by a vehicle-mounted lidar system comprising:
recording, with a vehicle-mounted global positioning system-assisted inertial measurement unit (GPS/IMU) system, vehicle position and vehicle orientation data at particular points in time as the vehicle traverses a route;
generating a route topology based on the vehicle position and the vehicle orientation data;
recording a first upper limit horizon estimation and a first lower limit horizon estimation by the lidar system for at least one of the particular points in time;
determining a relative lidar system position and a relative lidar system orientation based on a relative location of the lidar system with respect to the GPS/IMU system;
calculating a second upper limit horizon estimation and a second lower limit horizon estimation for the lidar system for at least one of the particular points in time based on the route topology and the relative lidar system position and relative lidar system orientation; and
assessing an upper limit horizon estimation accuracy and a lower limit horizon estimation accuracy of the lidar system that is able to generate the first upper limit horizon estimation and the first lower limit horizon estimation faster than utilizing the GPS/IMU system to augment the calculation of the second upper limit horizon estimation and the second lower limit horizon estimation including by comparing for the at least one of the particular points in time the first upper limit horizon estimation with the second upper limit horizon estimation and comparing the first lower limit horizon estimation with the second lower limit horizon estimation to generate horizon accuracy metrics.
16. The method of claim 15 , wherein the horizon accuracy metrics include upper horizon accuracy metrics and lower horizon accuracy metrics.
17. The method of claim 15 , wherein: each of the particular points in time is associated with a particular frame of data captured by the lidar system.
18. The method of claim 15 , wherein the route topology is based on the location of the lidar system on the vehicle.
19. The method of claim 15 , wherein the route topology is a road surface route topology and the generation of the road surface route topology is further based on a location of the GPS/IMU system on the vehicle relative to the road surface.
20. The method of claim 15 , wherein the GPS/IMU system vehicle position and vehicle orientation data is further based on real time kinematic (RTK) system correction data.
21. The method of claim 15 , wherein the GPS/IMU system and the lidar system share a common time signal.