Speed reasoning in the presence of uncertainty when pulling over
The technology involves pullover maneuvers for vehicles operating in an autonomous driving mode. A vehicle operating in an autonomous driving mode is able to identify parking locations that may be occluded or otherwise not readily visible. A best case estimation for any potential parking locations, including partially or fully occluded areas, is compared against predictions for identified visible parking locations. This can include calculating a cost for each potential parking location and comparing that cost to a baseline cost. Upon determining that an occluded location would provide a more viable option than any visible parking locations, the vehicle is able to initiate a driving adjustment (e.g., slow down) prior to arriving at the occluded location. This enables the vehicle to minimize passenger discomfort while also providing notice to other road users of an intent to pull over.
1 . A method, comprising:
obtaining, by one or more processors of a computing system, a clearance offset to a side of a roadway based on sensor data;
applying, by the one or more processors, a field of view restriction to the clearance offset to obtain a final offset, the field of view restriction corresponding to one or more occluded areas along the side of the roadway with respect to a vehicle;
comparing, by the one or more processors for a set of possible pullover locations, one or more respective factors based on the final offset to one or more baseline factors;
selecting, by the one or more processors based on the comparison, a pullover location along the side of the roadway; and
causing, by the one or more processors, the vehicle to perform a pullover maneuver in an autonomous driving mode according to the selected pullover location.
2 . The method of claim 1 , wherein causing the vehicle to perform the pullover maneuver includes determining a slow region beginning at a first point prior to the selected pullover location and ending at a second point farther from a current position of the vehicle than the first point.
3 . The method of claim 1 , wherein at least one pullover location of the set of possible pullover locations is an occluded location.
4 . The method of claim 1 , further comprising determining, by the one or more processors, the one or more respective factors for the set of possible pullover locations.
5 . The method of claim 4 , wherein determining the one or more respective factors includes determining spatial information and using the spatial information associated with the set of possible pullover locations.
6 . The method of claim 5 , wherein determining the spatial information includes determining a position of the vehicle when parked at a given one of the possible pullover locations.
7 . The method of claim 6 , wherein determining the position of the vehicle includes estimating a parking angle and a distance from the vehicle to a nearest driving lane.
8 . The method of claim 5 , wherein determining the spatial information includes evaluating one or more of a distance from a road edge, a parking angle, or a distance from a requested pullover location.
9 . The method of claim 1 , wherein the final offset is based on the clearance offset, a field of view limited offset, and a predicted offset.
10 . The method of claim 9 , wherein the field of view limited offset encompasses detectable objects from sensor data obtained by a perception system of the vehicle.
11 . The method of claim 10 , wherein the field of view limited offset further encompasses a field of view of the vehicle of the side of the roadway.
12 . The method of claim 9 , wherein the predicted offset is an offset around objects detectable by a perception system of the vehicle.
13 . The method of claim 1 , wherein the one or more baseline factors are one or more thresholds associated with selecting the pullover location.
14 . The method of claim 1 , wherein selecting the pullover location includes determining whether a given factor for a pullover location associated with any of the one or more occluded areas is below a threshold by a selected amount.
15 . A vehicle comprising:
a control system including one or more processors and configured to:
receive sensor data;
obtain, based on the received sensor data, a clearance offset to a side of a roadway;
apply a field of view restriction to the clearance offset to obtain a final offset, the field of view restriction corresponding to one or more occluded areas along a side of the roadway with respect to the vehicle;
compare, for a set of possible pullover locations, one or more respective factors based on the final offset to one or more baseline factors;
select, based on the comparison, a pullover location along the side of the roadway; and
cause the vehicle to perform a pullover maneuver in an autonomous driving mode according to the selected pullover location.
16 . The vehicle of claim 15 , wherein the control system is further configured to determine the one or more respective factors for the set of possible pullover locations.
17 . The vehicle of claim 15 , wherein the control system is further configured to determine a slow region beginning at a first point prior to the selected pullover location and ending at a second point farther from a current position of the vehicle than the first point.
18 . The vehicle of claim 15 , wherein the final offset is based on the clearance offset, a field of view limited offset, and a predicted offset.
19 . The vehicle of claim 15 , wherein the one or more baseline factors are one or more thresholds associated with selecting the pullover location.
20 . A non-transitory, tangible, computer-readable medium on which instructions are stored, the instructions, when executed by one or more processors, cause the one or more processors to implement a method, the method comprising:
obtaining a clearance offset to a side of a roadway based on sensor data;
applying a field of view restriction to the clearance offset to obtain a final offset, the field of view restriction corresponding to one or more occluded areas along the side of the roadway with respect to a vehicle;
comparing, for a set of possible pullover locations, one or more respective factors based on the final offset to one or more baseline factors;
selecting, based on the comparison, a pullover location along the side of the roadway; and
causing the vehicle to perform a pullover maneuver in an autonomous driving mode according to the selected pullover location.