Object detection based on three-dimensional distance measurement sensor point cloud data
AssigneeGM Cruise Holdings LLC
InventorsSandeep Gangundi, Sarthak Sahu, Nathan Harada, Phil Ferriere
Life of this patent
Estimating the term…
Abstract
Distance measurements are received from one or more distance measurement sensors, which may be coupled to a vehicle. A three-dimensional (3D) point cloud are generated based on the distance measurements. In some cases, 3D point clouds corresponding to distance measurements from different distance measurement sensors may be combined into one 3D point cloud. A voxelized model is generated based on the 3D point cloud. An object may be detected within the voxelized model, and in some cases may be classified by object type. If the distance measurement sensors are coupled to a vehicle, the vehicle may avoid the detected object.
Claims19
A method of object detection, the method comprising:
- receiving a plurality of distance measurements from one or more distance measurement sensors;
- generating a three-dimensional point cloud including a plurality of points in a volume of space within a radius around the one or more distance measurement sensors, wherein each of the plurality of points in the three-dimensional point cloud corresponds to one of the plurality of distance measurements;
- generating a voxelized model of the volume of space based on the three-dimensional point cloud, the voxelized model including a plurality of voxels positioned based on the plurality of points in the three-dimensional point cloud;
- generating a plurality of images corresponding to a plurality of elevations based on the voxelized model, wherein each image of the plurality of images represents a two-dimensional planar portion of the voxelized model at one elevation of the plurality of elevations;
- and
- detecting an object other than the one or more distance measurement sensors within the voxelized model based on at least a subset of the plurality of images.
A system for object detection, the system comprising:
- one or more distance measurement sensors that measure a plurality of distance measurements;
- a memory storing instructions;
- and
- a processor that executes the instructions, wherein execution of the instructions by the processor causes the processor to:
- receive the plurality of distance measurements from the one or more distance measurement sensors,
- generate a three-dimensional point cloud including a plurality of points in a volume of space within a radius around the one or more distance measurement sensors, wherein each of the plurality of points in the three-dimensional point cloud corresponds to one of the plurality of distance measurements,
- generate a voxelized model of the volume of space based on the three-dimensional point cloud, the voxelized model including a plurality of voxels positioned based on the plurality of points in the three-dimensional point cloud,
- generate a plurality of images corresponding to a plurality of elevations based on the voxelized model, wherein each image of the plurality of images represents a two-dimensional planar portion of the voxelized model at one elevation of the plurality of elevations, and
- detect an object other than the one or more distance measurement sensors within the voxelized model based on at least a subset of the plurality of images.
A non-transitory computer readable storage medium having embodied thereon a program, wherein the program is executable by a processor to perform a method of object detection, the method comprising:
- receiving a plurality of distance measurements from one or more distance measurement sensors;
- generating a three-dimensional point cloud including a plurality of points in a volume of space within a radius around the one or more distance measurement sensors, wherein each of the plurality of points in the three-dimensional point cloud corresponds to one of the plurality of distance measurements;
- generating a voxelized model of the volume of space based on the three-dimensional point cloud, the voxelized model including a plurality of voxels positioned based on the plurality of points in the three-dimensional point cloud;
- generating a plurality of images corresponding to a plurality of elevations based on the voxelized model, wherein each image of the plurality of images represents a two-dimensional planar portion of the voxelized model at one elevation of the plurality of elevations;
- and
- detecting an object other than the one or more distance measurement sensors within the voxelized model based on at least a subset of the plurality of images.
Dependencies are inferred from the claim text.