Autonomous vehicle maneuvering based upon risk associated with occluded regions
An autonomous vehicle (AV) is described herein. The AV is configured to identify an occluded region where a portion of a field of view of a sensor is occluded by an object. The AV is further configured to hypothesize that an object exists in the occluded region and is moving in the occluded region. The AV is still further configured to perform a driving maneuver based upon the hypothesized object existing in the occluded region.
1. An autonomous vehicle (AV) comprising:
a computing system;
a sensor system that is in communication with the computing system; and
a mechanical system that is operably coupled to the computing system, where the computing system is programmed to perform acts comprising:
controlling the mechanical system of the AV in connection with the AV navigating a route from a beginning location to a destination location, where the route includes a turn at an intersection from a first roadway to a second roadway, where the turn is across a lane of oncoming traffic in the first roadway;
as the AV approaches the intersection, identifying an occluded region based upon output of the sensor system, where the occluded region is a spatial region where the AV lacks visibility, and further where the occluded region includes a portion of the lane of oncoming traffic in the first roadway;
hypothesizing that a vehicle is travelling in the lane of oncoming traffic in the occluded region towards the AV;
inferring that the vehicle will decelerate upon the AV becoming visible to a driver of the vehicle; and
based upon the inferring that the vehicle will decelerate upon the AV becoming visible to the driver of the vehicle, controlling the mechanical system of the AV in connection with completing the turn from the first roadway to the second roadway despite the occluded region including the portion of the lane of oncoming traffic.
2. The AV of claim 1 , the acts further comprising:
based upon inferring that the vehicle will decelerate upon the AV becoming visible to the driver of the vehicle, computing a time when the vehicle is expected to reach the intersection due to the vehicle decelerating, wherein the turn is completed based upon the computed time.
3. The AV of claim 2 , where computing the time when the vehicle is expected to reach the intersection comprises:
computing a position of the vehicle in the occluded region, where the AV is estimated to be observable to the vehicle at the position.
4. The AV of claim 3 , where computing the time when the vehicle is expected to reach the intersection further comprises:
assigning an initial velocity to the vehicle, where the initial velocity is a speed that the vehicle is travelling when the vehicle is located at the position.
5. The AV of claim 4 , where computing the time when the vehicle is expected to reach the intersection further comprises:
assigning a final velocity to the vehicle, the final velocity being a speed to which the hypothetical vehicle is expected to slow as the vehicle reaches the intersection.
6. The AV of claim 5 , where computing the time when the vehicle is expected to reach the intersection further comprises:
assigning a deceleration rate to the vehicle, where the vehicle is inferred to decelerate at the deceleration rate until the vehicle reaches the final velocity.
7. The AV of claim 1 , where the turn from the first roadway to the second roadway is a left turn across the lane of oncoming traffic.
8. The AV of claim 1 , where the turn from the first roadway to the second roadway is a right turn across the lane of oncoming traffic.
9. The AV of claim 1 , the acts further comprising:
prior to controlling the mechanical system in connection with completing the turn from the first roadway to the second roadway, computing an amount of time required for the AV to complete the turn, where the amount of time is computed based upon a profile for the AV, and further where the mechanical system is controlled in connection with completing the turn based upon the computed amount of time required for the AV to complete the turn.
10. A method performed by an autonomous vehicle (AV), the method comprising: navigating a route from a beginning location to a destination location, where the route includes a turn at an intersection from a first roadway to a second roadway, where the turn is across a lane of oncoming traffic in the first roadway;
obtaining a signal generated by a sensor system of the AV as the AV approaches the intersection;
identifying an occluded region based upon the signal generated by the sensor system, where the occluded region is a spatial region where the AV lacks visibility, and further where the occluded region includes a portion of the lane of oncoming traffic in the first roadway;
inferring that a vehicle is travelling in the lane of oncoming traffic in the occluded region towards the AV;
inferring that the vehicle will decelerate upon the AV becoming visible to a driver of the vehicle; and
based upon the inferring that the vehicle will decelerate upon the AV becoming visible to the driver of the vehicle, completing the turn from the first roadway to the second roadway despite the occluded region including the portion of the lane of oncoming traffic.
11. The method of claim 10 , further comprising:
based upon inferring that the vehicle will decelerate upon the AV becoming visible to the driver of the vehicle, computing a time when the vehicle is expected to reach the intersection due to the vehicle decelerating, wherein the turn is completed based upon the computed time.
12. The method of claim 11 , where computing the time when the vehicle is expected to reach the intersection comprises:
computing a position of the vehicle in the occluded region, where the AV is estimated to be observable to the vehicle at the position.
13. The method of claim 12 , where computing the time when the vehicle is expected to reach the intersection further comprises:
assigning an initial velocity to the vehicle, where the initial velocity is a speed that the vehicle is travelling when the vehicle is located at the position.
14. The method of claim 13 , where computing the time when the vehicle is expected to reach the intersection further comprises:
assigning a final velocity to the vehicle, the final velocity being a speed to which the hypothetical vehicle is expected to slow as the vehicle reaches the intersection.
15. The method of claim 14 , where computing the time when the vehicle is expected to reach the intersection further comprises:
assigning a deceleration rate to the vehicle, where the vehicle is inferred to decelerate at the deceleration rate until the vehicle reaches the final velocity.
16. The method of claim 10 , where the turn from the first roadway to the second roadway is a left turn across the lane of oncoming traffic.
17. The method of claim 10 , where the turn from the first roadway to the second roadway is a right turn across the lane of oncoming traffic.
18. The method of claim 10 , further comprising:
prior to completing the turn from the first roadway to the second roadway, computing an amount of time required for the AV to complete the turn, where the amount of time is computed based upon a profile for the AV, and further where the turn is completed based upon the computed amount of time required for the AV to complete the turn.
19. An autonomous vehicle (AV) comprising a computing system that includes a computer-readable storage medium, where the computer-readable storage medium includes instructions that, when executed by a processor, cause the processor to perform acts comprising: controlling a mechanical system of the AV to cause the AV to navigate a route from a beginning location to a destination location, where the route includes a turn at an intersection from a first roadway to a second roadway, where the turn is across a lane of oncoming traffic in the first roadway;
obtaining a signal generated by a sensor system of the AV as the AV approaches the intersection;
identifying an occluded region based upon the signal generated by the sensor system, where the occluded region is a spatial region where the AV lacks visibility, and further where the occluded region includes a portion of the lane of oncoming traffic in the first roadway;
inferring that a vehicle is travelling in the lane of oncoming traffic in the occluded region towards the AV;
inferring that the vehicle will decelerate upon the AV becoming visible to a driver of the vehicle; and
based upon the inferring that the vehicle will decelerate upon the AV becoming visible to the driver of the vehicle, controlling the mechanical system of the AV to cause the AV to complete the turn from the first roadway to the second roadway despite the occluded region including the portion of the lane of oncoming traffic.
20. The computer-readable storage medium of claim 19 , the acts further comprising:
based upon inferring that the vehicle will decelerate upon the AV becoming visible to the driver of the vehicle, computing a time when the vehicle is expected to reach the intersection due to the vehicle decelerating, wherein the mechanical system is controlled to cause the AV to complete the turn based upon the computed time.