US 12,612,082 B1Granted

Identifying relevant objects based on artificial paths

AssigneeZoox, Inc.

InventorsMarin Kobilarov, Linjun Zhang

Life of this patent

Estimating the term…

Loading patent…

Abstract

Techniques for identifying relevant objects in an environment based on artificial paths are described herein. A vehicle may receive and/or analyze sensor data of an environment to identify an object. The vehicle may determine whether the object is located within a region of interest by projecting the object into discretized map data and determining which cell of the discretized map data the object is located within. Based on the object being located within a cell corresponding to a region of interest, the vehicle can generate an artificial path (e.g., a spatial representation of potential movements of the object) for the object which may be used to determine whether the vehicle and the object overlap (e.g., spatially and temporally) at any cell within an occupancy map. In such cases, the vehicle may generate a relevance score for the object which may be used to control the vehicle.

Claims20

Dependent claims

A system comprising:

one or more processors;
and
one or more non-transitory computer-readable media storing computer-executable instructions that, when executed, cause the system to perform operations comprising:
receiving, from a sensor associated with a vehicle, sensor data;
detecting, based at least in part on the sensor data, an object associated with an object location;
determining, based at least in part on inputting the object into a relevance filter, that the object is irrelevant to safe operation of the vehicle;
determining, based at least in part on the object location, that the object is at least partially within a cell of discretized map data;
determining, based at least in part on the object being at least partially within the cell, that the object is associated with a region of interest, wherein determining the object is associated with the region of interest is further based at least in part on an amount of uncertainty associated with an intent of the object meeting or exceeding a threshold;
generating, based at least in part on the object being associated with the region of interest, an artificial path associated with the object, the artificial path being a spatial representation of a movement of the object;
determining, based at least in part on the artificial path of the object and a trajectory of the vehicle, that the object and the vehicle are associated with an occupancy area;
generating, based at least in part on the object and the vehicle being associated with the occupancy area, a relevance score associated with the object;
and
controlling the vehicle based at least in part on the relevance score regardless of the object being determined to be irrelevant.

10 more elements

One or more non transitory computer readable media storing instructions executable by one or more processors, wherein the instructions, when executed, cause a system to perform operations comprising:

receiving sensor data associated with an environment;
detecting, based at least in part on the sensor data, an object;
receiving a trajectory for a vehicle to traverse through the environment;
determining, based at least in part on the trajectory and map data, a region of interest along the trajectory;
generating, based at least in part on the object being associated with a region of interest and having an intent associated with a level of uncertainty that meets or exceeds a threshold, an artificial path associated with the object;
determining, based at least in part on the artificial path of the object and the trajectory of the vehicle, that the object and the vehicle are associated with an occupancy area;
generating, based at least in part on the object and the vehicle being associated with the occupancy area, a relevance score associated with the object;
and
controlling the vehicle based at least in part on the relevance score.

6 more elements

Dependencies are inferred from the claim text.