US 11,535,265 B2Granted

Point cloud data processing method, apparatus, device, vehicle and storage medium

AssigneeApollo Intelligent Driving Technology (Beijing) Co., Ltd.

InventorsFan Yang, Yunlong Li, Chen Chen, Rongxun Jing

Life of this patent

Estimating the term…

Loading patent…

Abstract

The present application provides a point cloud data processing method, an apparatus, a device, a vehicle, and a storage medium, the method includes: acquiring, according to a preset frequency, raw data collected by sensors on a vehicle; and performing, according to the raw data of the sensors, data fusion processing to obtain a fusion result. By acquiring, according to the preset frequency, the raw data collected by the sensors on the vehicle in a latest period, and performing the data fusion processing to obtain the fusion result, a synchronous clock source may be removed, a weak clock synchronization may be realized, and the cost may be effectively reduced. The preset frequency may be flexibly set, which, when set with a larger value, can reduce a time difference between the raw data of the sensors and improve data accuracy.

Claims12

Dependent claims

A point cloud data processing method, comprising:

acquiring, by point cloud data processing apparatus, according to a preset frequency, raw data collected by sensors on a vehicle, wherein a period corresponding to the preset frequency is shorter than a time required for the sensors to scan one turn;
and performing, by the point cloud data processing apparatus, when a number of acquisition times reaches a preset number of times, data fusion processing on the raw data of the sensors for the preset number of times is determined according to the time required for the sensors to scan one turn and the preset frequency;
wherein the raw data comprises coordinate information;
the performing, by the point cloud data processing apparatus, when a number of acquisition times reaches a preset number of times, the data fusion processing on the raw data of the sensors for the preset number of times to obtain the fusion result for controlling the vehicle, comprises:
converting, by the point cloud data processing apparatus, the raw data of the sensors to a same coordinate to acquire point cloud data of the sensors in the same coordinate;
and performing, by the point cloud data processing apparatus, the data fusion processing on the point cloud data of the sensors to obtain the fusion result;
wherein the performing, by the point cloud data processing apparatus, the data fusion processing on the point cloud data of the sensors to obtain the fusion result, comprises:
dividing, by the point cloud data processing apparatus, the point cloud data of the sensors into an overlapping point data set and a non-overlapping point data set;
wherein the non-overlapping point data set comprises non-overlapping point data of each of the sensor that does not overlap with that of any other of the sensors, and the overlapping point data set comprises overlapping point data that overlaps between the sensors;
performing, by the point cloud data processing apparatus, fitting processing on the overlapping point data using a least square algorithm to form a least square curved surface, and fusing data of points with a Euclidean distance less than a threshold together based on the Euclidean distance between the points to acquire fused overlapping point data;
and combining, by the point cloud data processing apparatus, the non-overlapping point data and the fused overlapping point data to form fused point cloud data.

8 more elements

A point cloud data processing apparatus, comprising:

at least one processor and a memory;
the memory stores a computer program;
the at least one processor executes the computer program stored in the memory to:
acquire, according to a preset frequency, raw data collected by sensors on a vehicle, wherein a period corresponding to the preset frequency is shorter than a time required for the sensors to scan one turn;
perform, when a number of acquisition times reaches a preset number of times, data fusion processing on the raw data of the sensors for the preset number of times to obtain a fusion result for controlling the vehicle, wherein the preset number of times is determined according to the time required for the sensors to scan one turn and the preset frequency;
wherein the raw data comprises coordinate information;
the at least one processor executes the computer program stored in the memory to:
convert the raw data of the sensors to a same coordinate to acquire point cloud data of the sensors in the same coordinate;
and perform the data fusion processing on the point cloud data of the sensors to obtain the fusion result;
wherein the at least one processor executes the computer program stored in the memory to:
divide the point cloud data of the sensors into an overlapping point data set and a non-overlapping point data set;
wherein the non-overlapping point data set comprises non-overlapping point data of each of the sensor that does not overlap with that of any other of the sensors, and the overlapping point data set comprises overlapping point data that overlaps between the sensors;
perform fitting processing on the overlapping point data using a least square algorithm to form a least square curved surface, and fuse data of points with a Euclidean distance less than a threshold together based on the Euclidean distance between the points to acquire fused overlapping point data;
and combine the non-overlapping point data and the fused overlapping point data to form fused point cloud data.

11 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/708,560Filed 10 Dec 2019
Priority: CN 201811502444.X10 Dec 2018
Published: US 2020/0180652 A111 Jun 2020
Granted: 27 Dec 2022
Term ends: Not estimatedThe source records don’t support an estimate.

Classification12 codes

G06V 20/56
Context or environment of the image: exterior to a vehicle by using sensors mounted on the vehicle
B60W 30/18
Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of…:…
B60W 50/0205
Ensuring safety in case of control system failures, e.g. by diagnosing, circumventing or fixing failures: diagnosing or detecting failures;…
B60W 50/038
Ensuring safety in case of control system failures, e.g. by diagnosing, circumventing or fixing failures: limiting the input power, torque or speed