Georeferenced trajectory estimation system
A georeferenced trajectory estimation system receives data generated by vehicle sensors from a fleet of vehicles and identifies sets of sensor measurements for a geographic region taken at different times. The sets of sensor measurements include environment geometry data and unaligned trajectory data in local coordinate frames. The georeferenced trajectory estimation system aligns the environment geometry data between sets of sensor measures, and, based on the alignment of the environment geometry data, transforms the identified corresponding trajectory data into a common coordinate frame.
1. A georeferenced trajectory system for vehicles, the georeferenced trajectory system comprising:
a memory to store instructions; and
one or more processors using the instructions stored in the memory to perform operations including:
accessing a first set of sensor measurements for a geographic region taken by a first vehicle over a first period of time, the first set of sensor measurements comprising first environment geometry data for the geographic region and first trajectory data for the first vehicle referenced to a first local reference frame;
accessing a second set of sensor measurements for the geographic region taken by a second vehicle over a second period of time, the second set of sensor measurements comprising second environment geometry data for the geographic region and second trajectory data referenced to a second local reference frame different than the first local reference frame;
aligning the first environment geometry data with the second environment geometry data;
based on the alignment of the first environment geometry data with the second environment geometry data, transforming the first trajectory data and the second trajectory data into a common coordinate frame;
generating a map based on the common coordinate frame; and
navigating a vehicle using the map based on the common coordinate frame.
2. The georeferenced trajectory system of claim 1 , wherein the first trajectory data comprises a first sequence of poses over the first period of time for the first vehicle, and the second trajectory data comprises a second sequence of poses over the second period of time for the second vehicle.
3. The georeferenced trajectory system of claim 1 , wherein the first trajectory data and the second trajectory data comprise measurements of vehicle position, orientation, force, or motion taken from at least one of an inertial measurement unit, a wheel speed encoder, and a GPS unit.
4. The georeferenced trajectory system of claim 1 , wherein the first environment geometry data and the second environment geometry data are taken from one or more LIDAR units, radar units, sonar units, or cameras.
5. The georeferenced trajectory system of claim 1 , wherein the sets of sensor measurements include LIDAR beam intensity values for the geographic region, and wherein the memory stores further instructions to perform operations including transforming the first trajectory data and the second trajectory data into the common coordinate frame based on an alignment of the LIDAR beam intensity values.
6. The georeferenced trajectory system of claim 1 , wherein the memory stores further instructions to perform operations including aligning environment geometry data from further sets of sensor measurements to transform further trajectory data into the common coordinate frame.
7. The georeferenced trajectory system of claim 1 , wherein the first trajectory data and the second trajectory data are transformed non-rigidly.
8. The georeferenced trajectory system of claim 1 , wherein the first vehicle and the second vehicle are the same vehicle.
9. A method of vehicle trajectory mapping, the method being implemented by one or more processors and comprising:
accessing a first set of sensor measurements for a geographic region taken by a first vehicle over a first period of time, the first set of sensor measurements comprising first environment geometry data for the geographic region and first trajectory data for the first vehicle referenced to a first local reference frame;
accessing a second set of sensor measurements for the geographic region taken by a second vehicle over a second period of time, the second set of sensor measurements comprising second environment geometry data for the geographic region and second trajectory data referenced to a second local reference frame different than the first local reference frame;
aligning the first environment geometry data with the second environment geometry data;
based on the alignment of the first environment geometry data with the second environment geometry data, transforming the first trajectory data and the second trajectory data into a common coordinate frame;
generating a map based on the common coordinate frame; and
navigating a vehicle using the map based on the common coordinate frame.
10. The method of claim 9 , wherein the identifying steps further comprise sensing as the first trajectory data a first sequence of poses over the first period of time for the first vehicle and as the second trajectory data a second sequence of poses over the second period of time for the second vehicle.
11. The method of claim 9 , wherein the identifying steps further comprise measuring vehicle position, orientation, force, or motion taken from at least one of an inertial measurement unit, a wheel speed encoder, and a GPS unit as the first trajectory data and the second trajectory data.
12. The method of claim 9 , wherein the identifying steps further comprise taking the first environment geometry data and the second environment geometry data from one or more LIDAR units, radar units, sonar units, or cameras.
13. The method of claim 9 , wherein the sets of sensor measurements include LIDAR beam intensity values for the geographic region, and wherein transforming the first trajectory data and the second trajectory data into the common coordinate frame is based on an alignment of the LIDAR beam intensity values.
14. The method of claim 9 , further comprising aligning environment geometry data from further sets of sensor measurements to transform further trajectory data into the common coordinate frame.
15. The method of claim 9 , wherein transforming the first trajectory data and the second trajectory data into the common coordinate comprises transforming the first trajectory data and the second trajectory data non-rigidly.
16. The method of claim 9 , wherein the first vehicle and the second vehicle are the same vehicle.
17. A non-transitory computer-readable medium that stores instructions, executable by one or more processors, to cause the one or more processors to perform operations comprising:
accessing a first set of sensor measurements for a geographic region taken by a first vehicle over a first period of time, the first set of sensor measurements comprising first environment geometry data for the geographic region and first trajectory data for the first vehicle referenced to a first local reference frame;
accessing a second set of sensor measurements for the geographic region taken by a second vehicle over a second period of time, the second set of sensor measurements comprising second environment geometry data for the geographic region and second trajectory data referenced to a second local reference frame different than the first local reference frame;
aligning the first environment geometry data with the second environment geometry data;
based on the alignment of the first environment geometry data with the second environment geometry data, transforming the first trajectory data and the second trajectory data into a common coordinate frame;
generating a map based on the common coordinate frame; and
navigating a vehicle using the map based on the common coordinate frame.
18. The medium of claim 17 , further comprising instructions to cause the one or more processors to perform operations including sensing as the first trajectory data a first sequence of poses over the first period of time for the first vehicle and as the second trajectory data a second sequence of poses over the second period of time for the second vehicle.
19. The medium of claim 17 , wherein the sets of sensor measurements include LIDAR beam intensity values for the geographic region, further comprising instructions to cause the one or more processors to perform operations including transforming the first trajectory data and the second trajectory data into the common coordinate frame based on an alignment of the LIDAR beam intensity values.
20. The medium of claim 17 , further comprising instructions to cause the one or more processors to perform operations including aligning environment geometry data from further sets of sensor measurements to transform further trajectory data into the common coordinate frame.