# US 11,994,593 B2 · Detection method and device based on laser radar, and computer readable storage medium

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,994,593 B2
- Document ID: US11994593B2
- Status: Granted
- Kind: B2
- Application number: 17/639,905
- Filed: 2020-12-22
- Published: 2022-10-27
- Granted: 2024-05-28
- Earliest claimed: 2019-12-27
- Assignees: Zte Corporation
- Applicants: Zte Corporation
- Inventors: Zeqi Chen; Lei Xue; Xianlin Mei; Pengcheng Lai; Xinke Chen; Yuehui Luo
- Web page: https://patentgrep.com/patent/US11994593B2

## CPC

- B60W 40/02: related to ambient conditions
- G01S 17/931: of land vehicles
- G01S 7/4808: {Evaluating distance, position or velocity data}
- G01S 7/4817: {relating to scanning}
- G01S 17/89: for mapping or imaging
- G06V 10/757: {Matching configurations of points or features}
- B60W 2050/005: {Sampling}
- B60W 2050/0052: {Filtering, filters}
- B60W 2420/408: Radar; Laser, e.g. lidar
- B60W 2554/20: Static objects
- B60W 2554/40: Dynamic objects, e.g. animals, windblown objects
- G01S 17/08: for measuring distance only (indirect measurement G01S17/46; active triangulation systems G01S17/48)
- B60W: CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A detection method and a device based on a laser radar, and a computer readable storage medium are disclosed. The detection method includes: obtaining scanning data of the laser radar; performing an algorithm splitting on a feature algorithm for detection based on the scanning data to obtain at least one sub-algorithm capable of parallel processing in the feature algorithm; and performing a heterogeneous acceleration for the at least one sub-algorithm to process the scanning data and to obtain a processing result; and obtaining a detected position of an obstacle and a detected drivable area based on the processing result.

## Claims

### Claim 1

A detection method based on a laser radar, comprising:
obtaining scanning data of the laser radar;
performing algorithm splitting on a feature algorithm used for detection based on the scanning data to obtain at least one sub-algorithm capable of parallel processing in the feature algorithm;
performing heterogeneous acceleration on the at least one sub-algorithm to process the scanning data and to obtain a processing result; and
obtaining a detected position of an obstacle and a detected drivable area based on the processing result.

### Claim 2 (depends on 1)

The detection method based on the laser radar according to claim 1, wherein the performing the algorithm splitting on the feature algorithm used for detection based on the scanning data to obtain the at least one sub-algorithm capable of parallel processing in the feature algorithm comprises:
performing the algorithm splitting on a first feature algorithm used to detect a position of the obstacle based on the scanning data and on a second feature algorithm used to detect a drivable area based on the scanning data, to obtain a plurality of first sub-algorithms capable of parallel processing in the first feature algorithm and a plurality of second sub-algorithms capable of parallel processing in the second feature algorithm; and
the performing the heterogeneous acceleration on the at least one sub-algorithm to process the scanning data specifically comprises:
performing the heterogeneous acceleration on at least one of the plurality of first sub-algorithms and/or at least one of the plurality of second sub-algorithms to process the scanning data.

### Claim 3 (depends on 2)

The detection method based on the laser radar according to claim 2, wherein the first feature algorithm comprises:
a fitting algorithm for fitting a ground based on the scanning data;
a segmentation clustering algorithm for segmenting and clustering the obstacle based on a result of fitting the ground;
an edge extraction algorithm for extracting an edge frame of the obstacle based on a result of the segmentation and clustering; and
a position recognition algorithm for recognizing the position of the obstacle based on the edge frame; and
the performing the algorithm splitting for the first feature algorithm used to detect the position of the obstacle based on the scanning data comprises:
performing the algorithm splitting on at least one of the fitting algorithm, the segmentation clustering algorithm, the edge extraction algorithm, and the position recognition algorithm.

### Claim 4 (depends on 3)

The detection method based on the laser radar according to claim 3, wherein when performing the algorithm splitting on the edge extraction algorithm, the edge extraction algorithm is split into following operations:
extracting coordinates of M two-dimensional mapping outer contour points of the obstacle based on the result of the segmentation and clustering, wherein M is an integer greater than 1;
obtaining M fitting straight lines based on the coordinates of the M two-dimensional mapping outer contour points, and calculating M inclination angles of the M fitting straight lines, wherein one of the two-dimensional mapping outer contour points corresponds to one inclination angle;
rotating corresponding two-dimensional mapping outer contour points based on the inclination angles to obtain a cube frame of the obstacle.

### Claim 5 (depends on 4)

The detection method based on the laser radar according to claim 4, wherein the performing the heterogeneous acceleration on at least one of the plurality of first sub-algorithms and/or at least one of the plurality of second sub-algorithms specifically is:
performing the heterogeneous acceleration on at least one of edge extraction algorithms; and
the performing the heterogeneous acceleration on at least one of the edge extraction algorithms comprises:
parallel operations of obtaining M fitting straight lines based on the coordinates of the M two-dimensional mapping outer contour points and calculating the M inclination angles of the M fitting straight lines.

### Claim 6 (depends on 2)

The detection method based on the laser radar according to claim 2, wherein the second feature algorithm comprises:
a classification storage algorithm for the scanning data by a 360-degree grid classification and storage;
a ground point determining algorithm for acquiring ground points in the scanning data based on a result of the 360-degree grid classification and storage, and performing a preliminary coloring processing on the ground points; and
a grid coloring algorithm of the drivable area obtained by establishing a grid image based on the ground points and performing a refined coloring processing on the grid image; and
the performing the algorithmic splitting on the second feature algorithm used to detect the drivable area based on the scanning data comprises:
performing the algorithmic splitting on at least one of the classification storage algorithm, the ground point determining algorithm, and the grid coloring algorithm.

### Claim 7 (depends on 6)

The detection method based on the laser radar according to claim 6, wherein when performing the algorithmic splitting on the classification storage algorithm, the classification storage algorithm is split into following operations:
obtaining polar coordinates of original laser radar points based on the scanning data;
dividing a scanning area of the laser radar by a preset radian to obtain N sectors, wherein N is an integer greater than 1;
storing each of the original laser radar points in a two-dimensional array by sector based on a sector where the polar coordinates of each of the original laser radar points are located.

### Claim 8 (depends on 7)

The detection method based on the laser radar according to claim 7, wherein the performing the heterogeneous acceleration on at least one of the plurality of first sub-algorithms and/or at least one of the plurality of second sub-algorithms specifically is:
performing the heterogeneous acceleration on at least one of the classification storage algorithms; and
the performing the heterogeneous acceleration on at least one of the classification storage algorithms comprises:
parallel operations of the sector where the polar coordinates of each of the original laser radar points are located and storing each of the original laser radar points in the two-dimensional array formed by the sector.

### Claim 9 (depends on 6)

The detection method based on the laser radar according to claim 6, wherein performing the algorithm splitting on at least one of the classification storage algorithm, the ground point determining algorithm, and the grid coloring algorithm specifically is:
performing the algorithmic splitting on the grid coloring algorithm; and
the performing the algorithmic splitting on the grid coloring algorithm specifically is:
splitting the grid coloring algorithm into following operations:
determining whether each grid in the grid image is a ground, and performing a coloring processing on the grid determined to be the ground.

### Claim 10 (depends on 9)

The detection method based on the laser radar according to claim 9, wherein the performing the heterogeneous acceleration on at least one of the plurality of first sub-algorithms and/or at least one of the plurality of second sub-algorithms specifically is:
performing the heterogeneous acceleration on at least one of the plurality of second sub-algorithms; and
the performing the heterogeneous acceleration on at least one of the plurality of second sub-algorithms specifically comprises:
parallel operations of determining whether each grid in the grid image is the ground and performing the coloring processing for the grid determined to be the ground.

### Claim 11

A non-transitory computer-readable storage medium storing a computer program, when executed by a processor, allows the processor to perform a detection method based on a laser radar, wherein the detection method comprises:
obtaining scanning data of the laser radar;
performing algorithm splitting on a feature algorithm used for detection based on the scanning data to obtain at least one sub-algorithm capable of parallel processing in the feature algorithm;
performing heterogeneous acceleration on the at least one sub-algorithm to process the scanning data and to obtain a processing result; and
obtaining a detected position of an obstacle and a detected drivable area based on the processing result.

### Claim 12

A detection device based on a laser radar, comprising a memory and a processor, wherein the memory stores a computer program that, when executed by the processor, allows the processor to perform a detection method based on a laser radar;
wherein the detection method comprises:
obtaining scanning data of the laser radar;
performing algorithm splitting on a feature algorithm used for detection based on the scanning data to obtain at least one sub-algorithm capable of parallel processing in the feature algorithm;
performing heterogeneous acceleration on the at least one sub-algorithm to process the scanning data and to obtain a processing result; and
obtaining a detected position of an obstacle and a detected drivable area based on the processing result.