Puddle occupancy grid for autonomous vehicles
Aspects of the disclosure relate to generating a puddle occupancy grid including a plurality of cells. For instance, a first probability value for a puddle being located at a first location generated using sensor data from a first sensor may be received. A second probability value for a puddle being located at a second location generating using sensor data from a second sensor different from the first sensor may be received. A first cell may be identified from the plurality of cells using the first location. The first cell may also be identified using the second location. A value for the cell may be generated using the first probability value and the second probability value.
1. A method of generating a puddle occupancy grid including a plurality of cells, the method comprising:
receiving, by one or more processors, a first probability value for puddling at a first location generated using sensor data from a first sensor;
receiving, by the one or more processors, a second probability value for puddling at a second location generating using sensor data from a second sensor different from the first sensor;
identifying, by the one or more processors, a first cell from the plurality of cells using the first location;
identifying, by the one or more processors, the first cell from the plurality of cells using the second location;
generating, by the one or more processors, a value for the first cell using the first probability value and the second probability value; and
using, by the one or more processors, the value for the first cell to control a vehicle in an autonomous driving mode.
2. The method of claim 1 , wherein the first sensor is a LIDAR sensor, and the second sensor is a camera sensor.
3. The method of claim 1 , wherein generating the value for the first cell includes first incorporating the first probability value and subsequently incorporating the second probability value.
4. The method of claim 3 , wherein incorporating the first probability value includes using an inverse logistic regression function.
5. The method of claim 4 , wherein incorporating the second probability value includes using the inverse logistic regression function.
6. The method of claim 1 , wherein using the value for the first cell includes using a Sigmoid function to convert the value to a probability of puddling at the first cell.
7. The method of claim 1 , further comprising, sending the value for the first cell to one or more remote computing devices.
8. The method of claim 7 , further comprising, receiving values for cells of the occupancy grid from the one or more remote computing devices, and wherein controlling the vehicle is further based on the received values.
9. The method of claim 1 , wherein using the value for the first cell includes clustering cells of the occupancy grid, and wherein controlling the vehicle is further based on the clustered cells.
10. The method of claim 1 , wherein the first probability value and the second probability value are generated at different points in time.
11. A system for populating a puddle occupancy grid including a plurality of cells, the system comprising one or more processors configured to:
receive a first probability value for puddling being located at a first location generated using sensor data from a first sensor;
receive a second probability value for puddling being located at a second location generating using sensor data from a second sensor different from the first sensor;
identify a first cell of the plurality of cells using the first location;
identify the first cell of the plurality of cells using the second location;
generate a value fir the first cell using the first probability value and the second probability value; and
use the value for the first cell to control a vehicle in an autonomous driving mode.
12. The system of claim 11 , wherein the first sensor is a LIDAR sensor, and the second sensor is a camera sensor.
13. The system of claim 11 , wherein generating the value for the first cell includes first incorporating the first probability value and subsequently incorporating the second probability value.
14. The system of claim 13 , wherein the one or more processors are further configured to incorporate the first probability value by using an inverse logistic regression function.
15. The system of claim 14 , wherein the one or more processors are further configured to incorporate the second probability value by using the inverse logistic regression function.
16. The system of claim 11 , wherein the one or more processors are further configured to use the value for the first cell by using a Sigmoid function to convert the value to a probability of puddling at the first cell.
17. The system of claim 11 , the one or more processors are further configured to send the value for the first cell to one or more remote computing devices.
18. The system of claim 17 , the one or more processors are further configured to receive values for cells of the occupancy grid from the one or more remote computing devices, and wherein controlling the vehicle is further based on the received values.
19. The system of claim 11 , wherein the one or more processors are further configured to use the value for the first cell by clustering cells of the occupancy grid, and wherein controlling the vehicle is further based on the clustered cells.
20. The system of claim 11 , wherein the first probability value and the second probability value are generated at different points in time.