Unmarked crosswalks for autonomous vehicles
Aspects of the disclosure provide for controlling an autonomous vehicle having an autonomous driving mode. For instance, sensor data generated by a perception system of a vehicle may be received. The sensor data identifying objects in an environment of the vehicle. A trajectory corresponding to a future path of the vehicle may be received. That the vehicle is approaching an unmarked crosswalk situation may be determined based on the trajectory and the sensor data. In response to the determination, an area for an unmarked crosswalk is identified. The vehicle may be controlled in the autonomous driving mode based on the determination and the area.
1. A method for controlling a vehicle having an autonomous driving mode, the method comprising:
receiving, by one or more processors of the vehicle, sensor data generated by a perception system of a vehicle, the sensor data identifying objects in an environment of the vehicle;
receiving, by the one or more processors of the vehicle, a trajectory corresponding to a future path of the vehicle;
determining, by the one or more processors of the vehicle, that the vehicle is approaching an unmarked crosswalk situation based on the trajectory and the sensor data;
in response to the determination, identifying, by the one or more processors of the vehicle, an area corresponding to an unmarked crosswalk based on a type of the unmarked crosswalk situation; and
controlling, by the one or more processors of the vehicle, the vehicle in the autonomous driving mode based on the determination and the area.
2. The method of claim 1 , wherein the unmarked crosswalk situation corresponds to a situation in which there is no marked crosswalk in the environment.
3. The method of claim 1 , wherein the unmarked crosswalk situation corresponds to a situation in which there is no marked crosswalk in any pre-stored map information used to control the vehicle in the autonomous driving mode.
4. The method of claim 1 , wherein determining that the vehicle is approaching an unmarked crosswalk situation includes determining whether the environment includes a school zone.
5. The method of claim 1 , wherein determining that the vehicle is approaching an unmarked crosswalk situation includes determining whether the environment includes a school bus.
6. The method of claim 5 , wherein identifying the area is based on a location of the school bus.
7. The method of claim 1 , wherein determining that the vehicle is approaching an unmarked crosswalk situation includes determining whether the environment includes a construction zone.
8. The method of claim 7 , wherein identifying the area is based on a polygon of the construction zone.
9. The method of claim 1 , wherein determining that the vehicle is approaching an unmarked crosswalk situation includes determining whether the environment includes one or more persons directing traffic.
10. The method of claim 9 , wherein identifying the area is based on a location of the one or more persons directing traffic.
11. The method of claim 1 , wherein determining that the vehicle is approaching an unmarked crosswalk situation includes determining whether the environment includes a traffic accident.
12. The method of claim 11 , wherein identifying the area is based on a polygon for an accident zone of the accident.
13. The method of claim 1 , wherein determining that the vehicle is approaching an unmarked crosswalk situation includes determining whether the environment includes one or more emergency vehicles.
14. The method of claim 13 , wherein identifying the area is based on a location of a first of the one or more emergency vehicles.
15. The method of claim 14 , further comprising identifying a second area based on a location of a second of the one or more emergency vehicles, and wherein controlling the vehicle is further based on the second area.
16. The method of claim 1 , wherein controlling the vehicle includes giving a pedestrian in an area precedence for crossing a road with respect to the vehicle.
17. The method of claim 1 , wherein controlling the vehicle includes yielding to a pedestrian on a side of a road outside of the area.
18. The method of claim 1 , wherein controlling the vehicle includes using occlusion reasoning for the area in order to predict one or more potentially occluded pedestrians.
19. The method of claim 1 , wherein controlling the vehicle includes allowing the vehicle's computing devices to increase allocation of computing resources for detecting and responding to pedestrians as the vehicle approaches the area.
20. A vehicle having an autonomous driving mode, the vehicle comprising:
a perception system; and
a computing device including one or more processors configured to:
receive sensor data generated by the perception system, the sensor data identifying objects in an environment of the vehicle;
receive a trajectory corresponding to a future path of the vehicle;
determine that the vehicle is approaching an unmarked crosswalk situation based on the received trajectory and the sensor data;
in response to the determination, identify an area corresponding to an unmarked crosswalk based on a type of the unmarked crosswalk situation; and
control the vehicle in the autonomous driving mode based on the determination and the area.
21. The vehicle of claim 20 , wherein the unmarked crosswalk situation corresponds to a situation in which there is no marked crosswalk in any pre-stored map information used to control the vehicle in the autonomous driving mode.