Determining inputs for perception system
AssigneeZoox, Inc.
InventorsChuang Wang, Yuesong Xie
Life of this patent
Estimating the term…
Abstract
Techniques for clustering sensor data are discussed herein. Sensors of a vehicle may detect data points in an environment. Clustering techniques can be used in a vehicle safety system to determine connection information between the data points. The connection information can be used by a vehicle computing device that employs clustering and/or segmenting techniques to detect objects in an environment and/or to control operation of a vehicle.
Claims20
A system comprising:
- one or more processors;
- and
- one or more non-transitory computer-readable media storing instructions executable by the one or more processors, wherein the instructions, when executed, cause the system to perform operations comprising:
- receiving lidar data points captured by a lidar sensor associated with an autonomous vehicle;
- associating the lidar data points with a two-dimensional data representation comprising input data, the input data including elevation data and azimuth angle data;
- inputting the two-dimensional data representation into a model;
- receiving, from the model, connectivity data representing a connection between two or more features of the two-dimensional data;
- clustering, based at least in part on the connectivity data, the two or more features of the two-dimensional data;
- and
- identifying, based at least in part on the clustering, object data representing an object in the two-dimensional data representation;
- controlling, based at least in part on the object data, the autonomous vehicle.
A method comprising:
- receiving sensor data captured by a sensor associated with a vehicle;
- associating the sensor data with a two-dimensional data representation;
- inputting the two-dimensional data representation into a model;
- receiving, from the model, connectivity data representing a connection between two or more features of the two-dimensional data representation;
- clustering, based at least in part on the connectivity data, features of the two-dimensional data representation;
- and
- identifying, based at least in part on the clustering, object data representing an object in the two-dimensional data representation;
- controlling, based at least in part on the object data, the vehicle.
One or more non-transitory computer-readable media storing instructions executable by one or more processors, wherein the instructions, when executed, cause the one or more processors to perform operations comprising:
- receiving sensor data captured by a sensor associated with a vehicle;
- associating the sensor data with a two-dimensional data representation;
- inputting the two-dimensional data representation into a model;
- receiving, from the model, connectivity data representing a connection between two or more features of the two-dimensional data representation;
- clustering, based at least in part on the connectivity data, features of the two-dimensional data representation;
- identifying, based at least in part on the clustering, object data representing an object in the two-dimensional data representation;
- and
- controlling, based at least in part on the object data, the vehicle.
Dependencies are inferred from the claim text.