Zoox, Inc.

Arvind Srivastav

Jifei Qian

Yifan Zuo

Techniques for detecting radar data inaccuracies using a multistage filtering technique are discussed herein. A vehicle may capture radar data including a set of radar points. The vehicle can input the radar data into a machine-learning model configured to detect objects and/or generate bounding boxes associated with such detected objects. In some instances, the vehicle may cluster the radar points such that a cluster includes the radar points located proximate to and/or within a bounding box. To identify double bounce object detections, the vehicle may determine whether a detected object is a false-positive double bounce object detection based on evaluating the azimuth, range, and relative velocity double bounce criteria. Based on the detected object satisfying the double bounce criteria for azimuth, range, and relative velocity, the vehicle may modify the radar data.

Radar double bounce detection

Geo-localization is the process by which an entity, such as an autonomous vehicle, can determine its precise location. In an example, a vehicle may include sensors, such as 4D radar sensors, that can capture a point cloud representing an environment within which the vehicle is positioned. An efficient representation of the point cloud is generated, which may be preceded by filtering the point cloud by removing points associated with transient elements. The efficient representation may be based on applying dimensionality reduction techniques to the point cloud or the filtered point cloud. The efficient representation may be compared with previously-generated representations which include geo-location, and were generated by the same dimensionality reduction techniques. Based on determining a match, the geo-location associated with the matched previously-generated representation may be assigned as a current geo-location of the vehicle.

Geo-localization using 3D sensor data

Techniques for detecting and determining a density for a non-impeding object based on correlation between radar and lidar returns are described herein. The techniques provide for receiving lidar data and radar data from a vehicle system operating in an environment. Portions of the lidar data and radar data are determined based on being associated with moving objects in the environment. The portions are then correlated to determine a similarity between the radar and lidar data. The similarity may then be used to determine an indication and density of the non-impeding object in the environment and cause the vehicle to operate within the environment accordingly.

Growing community of inventors

San Francisco, CA, United States of America

Arvind Srivastav