Sensor static miss detection by monitoring occupancy values

1 . A method, comprising:

receiving scene data associated with an environment of an autonomous vehicle based on signals detected by a sensor;

generating an occupancy array based on the scene data, wherein the occupancy array comprises:

a plurality of points, wherein each point represents a location within the environment, and

an occupancy value for each point, wherein the occupancy value for a particular point of the plurality of points indicates whether an object is detected at a particular location within the environment that corresponds to the particular point defined at least partly relative to a position of the sensor;

modifying the occupancy value of at least one first point of the plurality of points that initially indicates non-occupancy to indicate occupancy, wherein the at least one first point is selected from the plurality of points based at least on the at least one first point indicating non-occupancy at a current time and a corresponding point from a previously generated occupancy array indicating occupancy during a future period of time;

identifying at least one static object miss based on a group of occupancy values of a group of points of the plurality of points within a particular region of the occupancy array, wherein the group of occupancy values includes the occupancy value of that at least one first point; and

navigating the autonomous vehicle based on modifying the occupancy value of the at least one first point of the plurality of points that initially indicates non-occupancy to indicate occupancy and based on identifying the at least one static object miss.

2 . The method of claim 1 , wherein each column of the occupancy array represents a different azimuth around the autonomous vehicle and each row of the occupancy array represents a different angular elevation.

3 . The method of claim 1 , wherein identifying at least one static object miss based on the group of occupancy values of the group of points within the particular region comprises determining that a threshold quantity of the group of occupancy values indicate no object is detected.

4 . The method of claim 1 , wherein the group of points comprises points that correspond to a same azimuth angle.

5 . The method of claim 1 , wherein modifying the occupancy value of at least one first point of the plurality of points further comprises modifying the occupancy value of at least one first point of the plurality of points based on a determined location of the autonomous vehicle.

6 . The method of claim 5 , wherein modifying the occupancy value of at least one first point of the plurality of points based on the determined location of the autonomous vehicle comprises:

identifying a body of water within the environment of the autonomous vehicle based on a comparison of the determined location of the autonomous vehicle with a map that includes the determined location;

determining that the at least one first point corresponds to the body of water; and

modifying the at least one first point to indicate a presence of an object.

7 . The method of claim 1 , wherein the occupancy value of the at least one first point is further modified based on the occupancy value of at least one second point, and wherein modifying the occupancy value of the at least one first point based on the occupancy value of at least one second point comprises determining each occupancy value corresponding of a set of points proximate the at least one first point indicate an object is detected; and modifying the occupancy value of the at least one first point to indicate an object is detected.

8 . A system comprising:

a data store storing computer-executable instructions; and

a processor configured to execute the computer-executable instructions, wherein execution of the computer-executable instructions causes the system to:

receive scene data associated with an environment of an autonomous vehicle based on signals detected by a sensor;

generate an occupancy array based on the scene data, wherein the occupancy array comprises:

a plurality of points, wherein each point represents a location within the environment defined at least partly relative to a position of the sensor, and

an occupancy value for each point, wherein the occupancy value for a particular point of the plurality of points indicates whether an object is detected at a particular location within the environment that corresponds to the particular point;

modify the occupancy value of at least one first point of the plurality of points that initially indicates non-occupancy to indicate occupancy, wherein the at least one first point is selected from the plurality of points based at least on the at least one first point indicating non-occupancy at a current time and a corresponding point from a previously generated occupancy array indicating occupancy during a future period of time;

identify at least one static object miss based on a group of occupancy values of a group of points of the plurality of points within a particular region of the occupancy array, wherein the group of occupancy values includes the occupancy value of that at least one first point; and

navigate the autonomous vehicle based on modifying the occupancy value of the at least one first point of the plurality of points that initially indicates non-occupancy to indicate occupancy and based on identifying the at least one static object miss.

9 . The system of claim 8 , wherein to modify the occupancy value of at least one first point of the plurality of points, the execution of the computer-executable instructions cause the system to modify the occupancy value of at least one first point of the plurality of points based on a determined location of the autonomous vehicle.

10 . The system of claim 9 , wherein to modify the occupancy value of at least one first point of the plurality of points based on the determined location of the autonomous vehicle, the execution of the computer-executable instructions cause the system to

identify a body of water within the environment of the autonomous vehicle based on a comparison of the determined location of the autonomous vehicle with a map that includes the determined location;

determine that the at least one first point corresponds to the body of water; and

modify the at least one first point to indicate a presence of an object.

11 . The system of claim 8 , wherein the occupancy value of the at least one first point indicates no object is detected,

wherein the processor is configured to modify the occupancy value of the at least one first point based on the occupancy value of at least one second point, and

wherein to modify the occupancy value of the at least one first point based on the occupancy value of at least one second point, the execution of the computer-executable instructions cause the processor to determine each occupancy value corresponding to a set of points proximate the at least one first point indicate an object is detected; and modify the occupancy value of the at least one first point to indicate an object is detected.

12 . One or more non-transitory computer-readable media comprising computer-executable instructions that, when executed by a computing system, cause the computing system to:

receive scene data associated with an environment of an autonomous vehicle based on signals detected by a sensor;

generate an occupancy array based on the scene data, wherein the occupancy array comprises:

a plurality of points, wherein each point represents a location within the environment defined at least partly relative to a position of the sensor, and

an occupancy value for each point, wherein the occupancy value for a particular point of the plurality of points indicates whether an object is detected at a particular location within the environment that corresponds to the particular point;

modify the occupancy value of at least one first point of the plurality of points that initially indicates non-occupancy to indicate occupancy, wherein the at least one first point is selected from the plurality of points based at least on the at least one first point indicating non-occupancy at a current time and a corresponding point from a previously generated occupancy array indicating occupancy during a future period of time;

identify at least one static object miss based on a group of occupancy values of a group of points of the plurality of points within a particular region of the occupancy array, wherein the group of occupancy values includes the occupancy value of that at least one first point; and

navigate the autonomous vehicle based on modifying the occupancy value of the at least one first point of the plurality of points that initially indicates non-occupancy to indicate occupancy and based on identifying the at least one static object miss.

13 . The one or more non-transitory computer-readable media of claim 12 , wherein to identify at least one static object miss based on the group of occupancy values of the group of points within the particular region, the execution of the computer-executable instructions cause the system to determine that a threshold quantity of the group of occupancy values indicate no object is detected.

14 . The one or more non-transitory computer-readable media of claim 12 , wherein to modify the occupancy value of at least one first point of the plurality of points, the execution of the computer-executable instructions cause the system to modify the occupancy value of at least one first point of the plurality of points based on a determined location of the autonomous vehicle.

15 . The one or more non-transitory computer-readable media of claim 12 , wherein the occupancy value of the at least one first point indicates no object is detected,

wherein the execution of the computer-executable instructions cause the system to modify the occupancy value of the at least one first point based on the occupancy value of at least one second point, and

wherein to modify the occupancy value of the at least one first point based on the occupancy value of at least one second point, the execution of the computer-executable instructions cause the system to determine each occupancy value corresponding to a set of points proximate the at least one first point indicate an object is detected and modify the occupancy value of the at least one first point to indicate an object is detected.