Tracking vanished objects for autonomous vehicles
Aspects of the disclosure relate to methods for controlling a vehicle having an autonomous driving mode. For instance, sensor data may be received from one or more sensors of the perception system of the vehicle, the sensor data identifying characteristics of an object perceived by the perception system. When it is determined that the object is no longer being perceived by the one or more sensors of the perception system, predicted characteristics for the object may be generated based on one or more of the identified characteristics. The predicted characteristics of the object may be used to control the vehicle in the autonomous driving mode such that the vehicle is able to respond to the object when it is determined that the object is no longer being perceived by the one or more sensors of the perception system.
1 . A method of controlling a vehicle having an autonomous driving mode, the method comprising:
identifying, by one or more processors, characteristics of an object perceived within a field of view of a perception system of the vehicle;
when the object is no longer perceived within the field of view of the perception system, generating, by the one or more processors, based on the identified characteristics, a set of predicted characteristics for the object;
when the object is no longer being perceived within the field of view of the perception system, generating, by the one or more processors and based on the set of predicted characteristics, a trajectory that would cause the vehicle respond to the object according to the set of predicted characteristics when the object is no longer being perceived within the field of view of the perception system; and
controlling, by the one or more processors, the vehicle in the autonomous driving mode based on the trajectory such that the vehicle responds to the object when the object is no longer being perceived within the field of view of the perception system.
2 . The method of claim 1 , wherein the set of predicted characteristics includes a predicted location, and wherein the method further includes generating, by the one or more processors, the predicted location of the object based on a type of the object.
3 . The method of claim 2 , wherein the type of the object is a pedestrian, a bicyclist, or another vehicle.
4 . The method of claim 1 , further comprising receiving, by the one or more processors, sensor data from the perception system at a first point in time, wherein the characteristics are identified based on the sensor data.
5 . The method of claim 4 , further comprising determining, by the one or more processors, a last observed location of the object before a second point in time when the object is no longer being perceived within the field of view of the perception system, wherein the set of predicted characteristics are generated further based on the last observed location of the object.
6 . The method of claim 5 , wherein the set of predicted characteristics for the object are generated at or after the second point in time.
7 . The method of claim 1 , further comprising determining, by the one or more processors, whether a last observed location of the object is within a predetermined distance of an edge of the field of view of the perception system, wherein the generation of the set of predicted characteristics is further based on the determination of whether the last observed location of the object was within the predetermined distance of the edge of the field of view of the perception system.
8 . The method of claim 1 , further comprising storing, by the one or more processors, object data including a last observed location of the object and a behavior prediction for the object, and wherein the object data is used to generate the set of predicted characteristics.
9 . The method of claim 8 , further comprising determining, by the one or more processors, whether the last observed location of the object is within a predetermined distance of an edge of the field of view of the perception system, wherein the storing is in response to the determination of whether the last observed location of the object was within the predetermined distance of the edge of the field of view of the perception system.
10 . A system comprising one or more processors configured to:
identify characteristics of an object perceived within a field of view of a perception system of a vehicle having an autonomous driving mode, the perception system having one or more sensors;
when the object is no longer perceived within the field of view of the perception system, generate, based on the identified characteristics, a set of predicted characteristics for the object;
when the object is no longer being perceived within the field of view of the perception system, generate, based on the set of predicted characteristics, a trajectory that would cause the vehicle to respond to the object according to the set of predicted characteristics when the object is no longer being perceived within the field of view of the perception system; and
control the vehicle in the autonomous driving mode based on the trajectory such that the vehicle responds to the object when the object is no longer being perceived within the field of view of the perception system.
11 . The system of claim 10 , wherein the set of predicted characteristics includes a predicted location, and wherein the one or more processors are further configured to generate the predicted location of the object based on a type of the object.
12 . The system of claim 11 , wherein the type of the object is a pedestrian, a bicyclist, or another vehicle.
13 . The system of claim 11 , wherein the one or more processors are further configured to receive sensor data from the perception system at a first point in time, and to identify the characteristics further based on the sensor data.
14 . The system of claim 13 , wherein the one or more processors are further configured to determine a last observed location of the object before a second point in time when the object is no longer being perceived within the field of view of the perception system, and to generate the set of predicted characteristics further based on the last observed location of the object at or after the second point in time.
15 . The system of claim 11 , wherein the one or more processors are further configured to determine whether a last observed location of the object is within a predetermined distance of an edge of the field of view of the perception system, wherein the generation of the set of predicted characteristics is further based on the determination of whether the last observed location of the object was within the predetermined distance of the edge of the field of view of the perception system.
16 . The system of claim 11 , wherein the one or more processors are further configured to store object data including a last observed location of the object and a behavior prediction for the object, and to use the object data to generate the set of predicted characteristics.
17 . The system of claim 16 , wherein the one or more processors are further configured to determine whether a last observed location of the object is within a predetermined distance of an edge of the field of view of the perception system, and to store the object data in response to the determination of whether the last observed location of the object was within the predetermined distance of the edge of the field of view of the perception system.
18 . The system of claim 10 , further comprising the vehicle.
19 . 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 perform a method of controlling a vehicle having an autonomous driving mode, the method comprising:
identifying characteristics of an object perceived within a field of view of a perception system of the vehicle;
when the object is no longer perceived within the field of view of the perception system, generating, based on the identified characteristics, a set of predicted characteristics for the object;
when the object is no longer being perceived within the field of view of the perception system, generating, based on the set of predicted characteristics, a trajectory that would cause the vehicle to respond to the object according to the set of predicted characteristics when the object is no longer being perceived within the field of view of the perception system; and
controlling the vehicle in the autonomous driving mode based on the trajectory such that the vehicle is able to respond to the object when the object is no longer being perceived within the field of view of the perception system.
20 . The medium of claim 19 , wherein the method further comprises determining, whether a last observed location of the object is within a predetermined distance of an edge of the field of view of the perception system, and wherein the generation of the set of predicted characteristics is further based on the determination of whether the last observed location of the object was within the predetermined distance of the edge of the field of view of the perception system.