Determining the stationary state of detected vehicles
Aspects of the disclosure relate to an autonomous vehicle that may detect other nearby vehicles and designate stationary vehicles as being in one of a short-term stationary state or a long-term stationary state. This determination may be made based on various indicia, including visible indicia displayed by the detected vehicle and traffic control factors relating to the detected vehicle. For example, the autonomous vehicle may identify a detected vehicle as being in a long-term stationary state based on detection of hazard lights being displayed by the detected vehicle, as well as the absence of brake lights being displayed by the detected vehicle. The autonomous vehicle may then base its control strategy on the stationary state of the detected vehicle.
1. A computer-implemented method comprising:
identifying, by one or more computing devices, a first vehicle using sensor data generated by one or more sensors of a second vehicle;
identifying, by the one or more computing devices, that the first vehicle is stationary;
determining, by the one or more computing devices, a stationary state of the first vehicle relating to a period of time that the first vehicle is expected to remain stationary based on map information and the sensor data; and
controlling, by the one or more computing devices, the second vehicle in accordance with a control strategy based on the determined stationary state.
2. The method of claim 1 , wherein the determined stationary state is further based on whether the sensor data indicates that the first vehicle emitting a particular type of light.
3. The method of claim 2 , wherein the particular type of light is at least one of hazard lights, emergency vehicle lights, or construction warning lights.
4. The method of claim 1 , wherein the determined stationary state is further is further based on whether the first vehicle is proximate to a traffic control location.
5. The method of claim 4 , wherein the traffic control location is one of a traffic light, a stop sign, a yield sign, a caution sign, a crosswalk, or a railroad crossing.
6. The method of claim 1 , further comprising:
determining that the first vehicle has remained stationary for a predetermined period of time; and
updating the determined stationary state based on the determination that the first vehicle has remained stationary for the predetermined period of time.
7. The method of claim 6 , wherein the first vehicle is proximate a first type of traffic control location of a plurality of traffic control location types, and wherein the predetermined period of time is based on the traffic control location type.
8. The method of claim 1 , further comprising identifying one or more traffic control factors, and wherein determining the determined stationary state is further based on the identified one or more traffic control factors.
9. The method of claim 1 , wherein determining the stationary state includes determining that the first vehicle is temporarily obstructed from moving by one or more objects of the sensor data.
10. The method of claim 1 , wherein determining the stationary state is based on a behavior database regarding how often previously observed vehicles have been in various stationary states.
11. The method of claim 10 , wherein determining the stationary state is based on a behavior database regarding how often previously observed vehicles have been in various stationary states given a particular set of visible indicia.
12. The method of claim 10 , wherein determining the stationary state is based on pre-stored data regarding how often previously observed vehicles have been in various stationary states given one or more traffic control factors.
13. The method of claim 1 , wherein determining the stationary state is further based on whether, prior to the sensor data was generated, the first vehicle was observed in motion.
14. The method of claim 1 , wherein determining the stationary state is further based on whether a door of the first vehicle is detected to be open based on the sensor data.
15. The method of claim 1 , wherein determining the stationary state is further based on a plurality of weighted factors.
16. The method of claim 15 , wherein at least one of the plurality of weighted factors is determined from the sensor data.
17. The method of claim 1 , wherein the control strategy includes changing lanes.
18. The method of claim 1 , further comprising, prior to controlling the second vehicle based on the control strategy, controlling the second vehicle, by the one or more computing devices, based on a second control strategy to follow a first route to a destination, wherein the control strategy is different from the control strategy, and wherein the control strategy includes determining a second route to the destination.
19. The method of claim 1 , further comprising, identifying a third vehicle from the sensor data, and wherein determining the stationary state is further based on a location of the third vehicle relative to the first vehicle.
20. The method of claim 19 , further comprising determining from the sensor data whether the third vehicle is blocking the first vehicle, and wherein determining the stationary state is further based on the determination of whether the third vehicle is blocking the first vehicle.