CN114924259A - Performance evaluation method for different vehicle-mounted laser radars - Google Patents
Performance evaluation method for different vehicle-mounted laser radars Download PDFInfo
- Publication number
- CN114924259A CN114924259A CN202210545114.9A CN202210545114A CN114924259A CN 114924259 A CN114924259 A CN 114924259A CN 202210545114 A CN202210545114 A CN 202210545114A CN 114924259 A CN114924259 A CN 114924259A
- Authority
- CN
- China
- Prior art keywords
- grid
- block
- height
- obstacle
- laser radar
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 claims abstract description 24
- 238000001514 detection method Methods 0.000 claims abstract description 22
- 230000004888 barrier function Effects 0.000 claims abstract description 19
- 238000003384 imaging method Methods 0.000 claims abstract description 14
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000004519 manufacturing process Methods 0.000 claims abstract description 5
- 238000010586 diagram Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 2
- 101100083446 Danio rerio plekhh1 gene Proteins 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- 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
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
- G01S7/497—Means for monitoring or calibrating
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The invention discloses a performance evaluation method of different vehicle-mounted laser radars, which comprises the steps of firstly compressing three-dimensional point cloud data of different laser radars into a two-dimensional grid map, and identifying and judging obstacles by taking height information of the point cloud as a judgment index; then, marking the barrier grids on the two-dimensional grid map by adopting a target clustering algorithm according to the types of the barriers, and forming a growth imaging graph with a barrier track; and finally, the detection range of the laser radar can be clearly and intuitively judged by taking the outermost obstacle track end point as an evaluation index, so that the performance of the laser radar is evaluated. Therefore, the method has the advantages of simple algorithm, strong universality, high processing efficiency and the like, can realize the transverse performance evaluation among a plurality of laser radars with the same line number but different production batches or manufacturers, and has important significance on the automatic driving technology.
Description
Technical Field
The invention relates to the technical field of laser radars, in particular to a performance evaluation method for different vehicle-mounted laser radars.
Background
The automatic driving automobile can automatically acquire the environmental information around the automobile and make decisions and path planning by means of technologies such as artificial intelligence, computer vision, radar, global positioning system and high-precision maps, and therefore automatic driving completely independent of human operation is achieved. The laser radar positioning is one of the mainstream positioning schemes of the current automatic driving system, and has important significance in evaluating the positioning effect of the laser radar positioning in various environments and scenes.
Currently, the evaluation process of the laser radar positioning effect is generally as follows: the method comprises the steps of firstly collecting point cloud data required by positioning, then inputting the point cloud data into a positioning algorithm, obtaining a positioning result through calculation of the positioning algorithm, and finally obtaining an evaluation result of the positioning effect through comparison of the positioning result and a positioning true value of a vehicle. However, the method can only realize the positioning effect evaluation of a single laser radar product, and cannot realize the transverse performance comparison of different laser radar products under the same line number. Therefore, the performance evaluation method for different vehicle-mounted laser radars is provided.
Disclosure of Invention
The invention provides a performance evaluation method for different vehicle-mounted laser radars, and mainly aims to solve the problems in the prior art.
The invention adopts the following technical scheme:
a performance evaluation method for different vehicle-mounted laser radars comprises the following steps:
s1, placing at least two laser radars which have the same line number and different production batches or manufacturers in the same simulation environment for detection test, thereby obtaining three-dimensional point cloud data of each laser radar;
s2, establishing a two-dimensional grid map, rasterizing X, Y and Z coordinates under a laser radar coordinate system, projecting point clouds of different frame numbers of each laser radar into the two-dimensional grid map according to the X and Y coordinates, and recording Z-axis information of each point cloud as the height of a grid where the point cloud is located;
s3, taking the height difference between two adjacent grids as a judgment index, defining the grid meeting the height difference requirement as a non-obstacle grid, and defining the grid not meeting the height difference requirement as an obstacle grid;
s4, processing the two-dimensional grid map by adopting a target clustering algorithm on the basis of the domain marking method, thereby marking the obstacle grids on the two-dimensional grid map according to the types of the obstacles;
s5, superposing two-dimensional grid maps of the same laser radar and different frames to obtain a complete growth imaging map;
and S6, comparing the growth imaging graph of each laser radar, and judging the detection range of the laser radar by taking the outermost obstacle track endpoint as an evaluation index, thereby evaluating the performance of the laser radar.
Further, in step S2, if two or more point clouds with different heights exist in the same grid, the highest height h of the grid is recorded max And a minimum height h min 。
Further, the two-dimensional grid map is divided into a plurality of 3-by-3 grid blocks, each grid block comprises 9 grids, the point cloud data distribution characteristics of the laser radar are integrated, and the highest height h of the 9 grids contained in each grid block is used as the highest height h max And a minimum height h min Based on the height H of each grid block max And a minimum height H min 。
Further, any grid block is marked as A n Then grid block A n Highest height H of maxn And an initial minimum height H minn ' are respectively:
H maxn ＝h maxn5
H minn ’＝min{h minn1 ，h minn2 ，......h minni ，......，h minn9 }
in the formula: h is minni Is a grid block A n The lowest height of the ith grid of the 9 grids contained.
Further, the grid block A is divided into n Comparing with the initial minimum height of four grid blocks at the upper, lower, left and right sides, taking the lowest value of the initial minimum height of the 5 grid blocks as the grid block A n Minimum height H of minn 。
Further, a grid block A is calculated n And if any height difference value is smaller than a set value, the grid block A is divided into four grid blocks A n Defining as barrier grid block, if all height difference values are less than set value, defining grid block A n Defined as non-obstacle blocks.
Further, the step S4 includes the following sub-steps:
s41, traversing the two-dimensional grid map, if a certain grid block is an obstacle grid block and is not marked, marking the obstacle grid block, and recording the position of the obstacle grid block;
s42, with the recorded positions of the obstacle grid blocks as scanning midpoints, expanding 2 grid blocks to the periphery of the scanning midpoints to form scanning areas with the areas of 5 x 5, recording position information of all the obstacle grid blocks which are not marked in the scanning areas, marking all the obstacle grid blocks in the scanning areas in the same way, and simultaneously recording coordinate values minx, miny, maxx and maxy of a two-dimensional grid map corresponding to each scanning area;
s43, repeating the steps S41 and S42 until all obstacle grid blocks in the two-dimensional grid map are marked;
and S44, drawing all barrier grid blocks belonging to the same mark by adopting a minimum circumscribed rectangle method to form a barrier box block diagram.
Further, in step S6, learning the obstacle grid growth image map of each lidar by using a rectogram training set supervised learning method to compare the detection range of the lidar.
Compared with the prior art, the invention has the beneficial effects that:
the invention provides a performance evaluation method of different vehicle-mounted laser radars, which comprises the steps of firstly compressing three-dimensional point cloud data of different laser radars into a two-dimensional grid map, and identifying and judging obstacles by taking height information of the point cloud as a judgment index; then, processing the two-dimensional grid map by adopting a target clustering algorithm on the basis of a domain marking method, marking the barrier grids on the two-dimensional grid map according to the types of the barriers, and forming a growth imaging graph with a barrier track; and finally, the detection range of the laser radar can be clearly and visually judged by taking the outermost obstacle track endpoint as an evaluation index, so that the performance of the laser radar can be evaluated. Therefore, the method has the advantages of simple algorithm, strong universality, high processing efficiency and the like, can realize the transverse performance evaluation among a plurality of laser radars with the same line number but different production batches or manufacturers, and has important significance on the automatic driving technology.
Drawings
FIG. 1 is a schematic overall flow diagram of the present invention.
FIG. 2 is a diagram showing a relationship between a two-dimensional grid map and a lidar in the present invention.
Fig. 3 is a flow chart of creating a two-dimensional grid map according to the present invention.
FIG. 4 is a block diagram of a grid according to the present invention.
FIG. 5 is a flow chart of the objective clustering algorithm of the present invention.
FIG. 6 is a template diagram of connected domain of the target class-based algorithm of the present invention.
FIG. 7 is a growth imaging diagram of the first type of lidar.
FIG. 8 is a growth image of the second laser radar.
Detailed Description
The following describes embodiments of the present invention with reference to the drawings. Numerous details are set forth below in order to provide a thorough understanding of the present invention, but it will be apparent to those skilled in the art that the present invention may be practiced without these details.
Referring to fig. 1 to 8, a performance evaluation method for different vehicle-mounted laser radars includes the following steps:
and S1, placing at least two laser radars with the same line number but different production batches or manufacturers in the same simulation environment for detection test, thereby obtaining the three-dimensional point cloud data of each laser radar. In order to ensure accurate subsequent performance evaluation, a control variable method is adopted in a detection test to ensure that the detection environment, the detection position and the detection object of each laser radar are the same.
S2, establishing a two-dimensional grid map, rasterizing X, Y and Z coordinates under a laser radar coordinate system, projecting point clouds of different frame numbers of each laser radar into the two-dimensional grid map according to the X and Y coordinates, and recording Z-axis information of each point cloud as the height of the grid where the point cloud is located. Specifically, step S2 includes the following sub-steps:
s21, combining the detection range and relative requirements of the actual vehicle-mounted laser radar, the invention sets the detection range of the laser radar as 6m in front, 3m in back and 3m on two sides respectively, so the size of the two-dimensional grid map is 9m by 6m, and accordingly, a 300 by 200 two-dimensional grid map is built, wherein each grid is 3 x 3cm, and 6w grids are provided. The positional relationship with the lidar of the two-dimensional grid map is shown in fig. 1.
S22, after the two-dimensional grid map is built, the X, Y and Z coordinates under the laser radar coordinate system need to be rasterized, and the point clouds of different frame numbers of each laser radar are projected into the two-dimensional grid map according to the X and Y coordinates. If the three-dimensional point cloud data is only converted into two-dimensional coordinates, the height information (Z-axis information) of the three-dimensional data is lost, but the height information is important characteristic information of the point cloud data, so that the height information of each point cloud needs to be synchronously stored as the height of a grid during rasterization. Since two or more point clouds may exist in each grid at the same time, only the maximum height h of the grid needs to be recorded during projection max And a minimum height h min And (4) finishing. As can be seen with reference to fig. 3, this step comprises the following sub-steps:
s221, judging whether each point cloud is in the two-dimensional grid map or not according to the X and Y coordinates, if not, ignoring the point cloud, and if so, projecting the point cloud into the two-dimensional grid map, and executing the next step;
s222, judging whether the height information of the point cloud is higher than the highest height h of the grid or not max Or below the minimum height h of the grid min If yes, the maximum height h of the grid is adjusted max Or a minimum height h min Updating, and if not, ignoring the height information of the point cloud;
and S223, checking whether the point clouds are processed completely, outputting a two-dimensional grid map if the point clouds are processed completely, and otherwise, repeatedly executing the steps.
And S3, taking the height difference between two adjacent grids as a judgment index, defining the grid meeting the height difference requirement as a non-obstacle grid, and defining the grid not meeting the height difference requirement as an obstacle grid. Specifically, the step includes the following substeps:
s31, since the two-dimensional grid map includes a large amount of point cloud data information, it is necessary to perform appropriate optimization processing on the two-dimensional grid map before performing obstacle grid recognition in order to reduce the amount of computation and improve the processing efficiency. As shown in fig. 4, the two-dimensional grid map is divided into 3 × 3 grid blocks, each of which contains 9 grids, and in fact, the two-dimensional grid map is reduced by three times, so that 9 grids are arranged in 3 × 3 and combined into one grid block.
S32, synthesizing the point cloud data distribution characteristics of the laser radar, and calculating the highest height h of 9 grids contained in each grid block max And a minimum height h min Based on the height H of each grid block max And a minimum height H min 。
First, a certain grid block is marked as A n The 9 grids contained therein are marked with a in turn n1 ，a n1 ，......，a ni ，......，a n9 The highest height of the 9 grids is recorded as h maxni The minimum height is recorded as h minni Where n represents the number of the grid block and i represents the number of the grid. Then, grid block A n Highest height H of maxn And an initial minimum height H minn ' are respectively:
H maxn ＝h maxn5
H minn ’＝min{h minn1 ，h minn2 ，......h minni ，......，h minn9 }。
by adopting the method, each grid block A can be obtained n Highest height H of maxn And an initial minimum height H minn ’。
Next, the grid block A n Comparing with the initial minimum heights of the four raster blocks at the upper, lower, left and right sides, and taking the lowest value of the initial minimum heights of the 5 raster blocks as the raster block A n Minimum height H of minn . As shown in FIG. 4, assume that a certain grid block is labeled A 1 The grid blocks adjacent to the grid blocks at the upper, lower, left and right sides are respectively marked as A 2 ，A 2 ，A 3 ，A 4 ，A 5 Then the initial minimum heights of the five grid blocks are respectively H min1 ’，H min2 ’，H min3 ’，H min4 ’，H min5 ', then grid block A 1 Minimum height H of min1 Comprises the following steps:
H min1 ＝min{H min1 ’，H min2 ’，H min3 ’，H min4 ’，H min5 ’}。
by adopting the method, each grid block A can be further calculated n Minimum height H of minn 。
S33, calculating a grid block A n And if any height difference value is smaller than a set value, the grid block A is divided into four grid blocks A n Defining as barrier grid block, if all height difference values are less than set value, defining grid block A n Defined as non-obstacle blocks.
Taking FIG. 4 as an example, the grid block A is calculated 1 Highest height H of max1 And a grid block A 2 ，A 2 ，A 3 ，A 4 ，A 5 Minimum height H of min1 ，H min2 ，H min3 ，H min4 And recording each height difference as:
ΔH＝{|ΔH A12 |，|ΔH A13 |，|ΔH A14 |，|ΔH A15 |}
if any height difference value in the delta H is larger than 0.1, the grid block A is divided into 1 Defining as barrier grid block, if all height difference values in delta H are less than 0.1, then grid block A 1 Defined as a non-obstacle grid block.
During laser radar detection, a large amount of three-dimensional point cloud data is generated, but due to instrument jitter, current interference, high temperature and the like, the laser radar detection system also comprises a large amount of data irrelevant to target detection besides the point cloud data of obstacles. In order to improve the accuracy of target detection and the efficiency of data processing, the invention can also filter out irrelevant point clouds in the point cloud data.
And S4, processing the two-dimensional grid map by adopting a target clustering algorithm on the basis of the through-domain marking method, and marking the obstacle grid on the two-dimensional grid map according to the type of the obstacle. As shown in fig. 5, this step can be divided into the following sub-steps:
s41, traversing the two-dimensional grid map, if a certain grid block is an obstacle grid block and is not marked, marking the obstacle grid block, and recording the position information of the obstacle grid block.
S42, using the recorded position of the obstacle grid block as a scan midpoint, expanding 2 grid blocks to the periphery of the scan midpoint to form scan regions (as shown in fig. 6) with an area of 5 × 5, recording position information of all the obstacle grid blocks that are not marked in the scan region, marking all the obstacle grid blocks in the scan region identically, and recording coordinate values minx, miny, maxx, and maxy of the two-dimensional grid map corresponding to each scan region.
And S43, repeating the steps S41 and S42 until all obstacle grid blocks in the two-dimensional grid map are marked. Specifically, all the barrier grid blocks with intersections in the scanning area are marked identically, so that the discrete grid blocks are divided into a set of different types of barriers.
And S44, drawing all barrier grid blocks belonging to the same mark by adopting a minimum external rectangle method by combining coordinate values minx, miny, maxx and maxy of the two-dimensional grid map corresponding to each scanning area to form a barrier box block diagram.
S5, superposing two-dimensional grid maps of the same laser radar and different frames to obtain a complete growth imaging map;
and S6, comparing the growth imaging graph of each laser radar, and judging the detection range of the laser radar by taking the outermost obstacle track endpoint as an evaluation index, thereby evaluating the performance of the laser radar.
Fig. 7 and 8 are growth imaging graphs of two tested lidar types, wherein fig. 7 is a growth imaging graph of the itanium-16 line-lidar, and fig. 8 is a growth imaging graph of the leishen-16 line-lidar. Comparing the two images shows that as the detection distance increases, the point cloud becomes more sparse, which results in the decrease of accuracy. In order to accurately measure the visible range of the two types of laser radars, gridwide 0.03m is changed into gridwide 0.135m, namely, the grid map in front of the laser radars is 27 m. It can be seen that the footprint of the obstacle cannot be detected by the Itanium-16 line laser radar at about 25m, and the footprint of the obstacle cannot be detected by the Reynes-16 line laser radar at about 24m, so that the detection range of the Itanium-16 line laser radar is clearly and intuitively judged to be farther than that of the Reynes-16 line laser radar, and the product performance is better.
Besides manual visual judgment, the two growth imaging graphs can be further compared through supervised learning of a Rectangle training set, and therefore the performance of the two vehicle-mounted laser radar products can be evaluated intelligently.
The above description is only an embodiment of the present invention, but the design concept of the present invention is not limited thereto, and any insubstantial modifications made by using the design concept should fall within the scope of infringing the present invention.
Claims (8)
1. A performance evaluation method for different vehicle-mounted laser radars is characterized by comprising the following steps: the method comprises the following steps:
s1, placing at least two laser radars which have the same line number and different production batches or manufacturers in the same simulation environment for detection test, thereby obtaining three-dimensional point cloud data of each laser radar;
s2, establishing a two-dimensional grid map, rasterizing X, Y and Z coordinates under a laser radar coordinate system, projecting point clouds of different frame numbers of each laser radar into the two-dimensional grid map according to the X and Y coordinates, and recording Z-axis information of each point cloud as the height of a grid where the point cloud is located;
s3, taking the height difference between two adjacent grids as a judgment index, defining the grid meeting the height difference requirement as a non-obstacle grid, and defining the grid not meeting the height difference requirement as an obstacle grid;
s4, processing the two-dimensional grid map by adopting a target clustering algorithm on the basis of the through-domain marking method, and marking the obstacle grids on the two-dimensional grid map according to the types of the obstacles;
s5, superposing two-dimensional grid maps of the same laser radar and different frames to obtain a complete growth imaging map;
and S6, comparing the growth imaging graph of each laser radar, and judging the detection range of the laser radar by taking the outermost obstacle track endpoint as an evaluation index, thereby evaluating the performance of the laser radar.
2. The method of claim 1, wherein the performance evaluation method of different vehicle-mounted lidar comprises: in step S2, if two or more point clouds having different heights exist in the same grid, the highest height h of the grid is recorded max And a minimum height h min 。
3. The method of claim 1, wherein the performance evaluation method of different vehicle-mounted lidar comprises: dividing the two-dimensional grid map into a plurality of 3-by-3 grid blocks, wherein each grid block comprises 9 grids, and integrating the point cloud data distribution characteristics of the laser radar to obtain the highest height h of the 9 grids contained in each grid block max And a minimum height h min Based on the height of each grid block, the maximum height H of each grid block is set max And a minimum height H min 。
4. The method of claim 3, wherein the performance evaluation method of different vehicle-mounted lidar comprises: mark any grid block as A n Then, the grid block A n Highest height H of maxn And an initial minimum height H minn ' are respectively:
H maxn ＝h maxn5
H minn ’＝min{h minn1 ，h minn2 ，……h minni ，……，h minn9 }
in the formula: h is minni Is a grid block A n The lowest height of the ith grid of the 9 grids contained.
5. The method of claim 4, wherein the performance evaluation method of different vehicle-mounted lidar comprises: block A of the grid n Comparing with the initial minimum height of four grid blocks at the upper, lower, left and right sides, taking the lowest value of the initial minimum height of the 5 grid blocks as the grid block A n Minimum height H of minn 。
6. A method of performance assessment of different vehicle-mounted lidar according to claim 3, 4 or 5, wherein: calculating a certain grid block A n And if any height difference value is smaller than a set value, the grid block A is divided into four grid blocks A n Defining as barrier grid block, if all height difference values are less than set value, defining grid block A n Defined as a non-obstacle block.
7. The method of claim 1, wherein the performance evaluation method of different vehicle-mounted lidar comprises: the step S4 includes the following sub-steps:
s41, traversing the two-dimensional grid map, if a certain grid block is an obstacle grid block and is not marked, marking the obstacle grid block, and recording the position of the obstacle grid block;
s42, with the recorded positions of the obstacle grid blocks as scanning midpoints, expanding 2 grid blocks to the periphery of the scanning midpoints to form scanning areas with the areas of 5 x 5, recording position information of all the obstacle grid blocks which are not marked in the scanning areas, marking all the obstacle grid blocks in the scanning areas in the same way, and simultaneously recording coordinate values minx, miny, maxx and maxy of a two-dimensional grid map corresponding to each scanning area;
s43, repeating the steps S41 and S42 until all barrier grid blocks in the two-dimensional grid map are marked;
and S44, drawing all barrier grid blocks belonging to the same mark by adopting a minimum external rectangle method to form a barrier box block diagram.
8. The method of claim 1, wherein the performance evaluation method of different vehicle-mounted lidar comprises: in step S6, a rectogram training set supervised learning method is used to learn the obstacle grid growth imaging map of each lidar so as to compare the detection range of the lidar.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210545114.9A CN114924259B (en) | 2022-05-19 | 2022-05-19 | Performance evaluation method for different vehicle-mounted laser radars |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210545114.9A CN114924259B (en) | 2022-05-19 | 2022-05-19 | Performance evaluation method for different vehicle-mounted laser radars |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114924259A true CN114924259A (en) | 2022-08-19 |
| CN114924259B CN114924259B (en) | 2024-06-07 |
Family
ID=82807868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210545114.9A Active CN114924259B (en) | 2022-05-19 | 2022-05-19 | Performance evaluation method for different vehicle-mounted laser radars |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114924259B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019165451A1 (en) * | 2018-02-26 | 2019-08-29 | Nvidia Corporation | Systems and methods for computer-assisted shuttles, buses, robo-taxis, ride-sharing and on-demand vehicles with situational awareness |
| CN111928862A (en) * | 2020-08-10 | 2020-11-13 | 廊坊和易生活网络科技股份有限公司 | Method for constructing semantic map on line by fusing laser radar and visual sensor |
| CN113433937A (en) * | 2021-06-08 | 2021-09-24 | 杭州未名信科科技有限公司 | Heuristic exploration-based layered navigation obstacle avoidance system and layered navigation obstacle avoidance method |
-
2022
- 2022-05-19 CN CN202210545114.9A patent/CN114924259B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019165451A1 (en) * | 2018-02-26 | 2019-08-29 | Nvidia Corporation | Systems and methods for computer-assisted shuttles, buses, robo-taxis, ride-sharing and on-demand vehicles with situational awareness |
| CN111928862A (en) * | 2020-08-10 | 2020-11-13 | 廊坊和易生活网络科技股份有限公司 | Method for constructing semantic map on line by fusing laser radar and visual sensor |
| CN113433937A (en) * | 2021-06-08 | 2021-09-24 | 杭州未名信科科技有限公司 | Heuristic exploration-based layered navigation obstacle avoidance system and layered navigation obstacle avoidance method |
Non-Patent Citations (3)
| Title |
|---|
| 任少杰;杨正才;张英泽;: "基于栅格化四边形区域与改进人工势场的车辆行驶轨迹研究", 湖北汽车工业学院学报, no. 03, 15 September 2020 (2020-09-15) * |
| 侯加林;蒲文洋;李天华;丁小明;: "双激光雷达温室运输机器人导航系统研制", 农业工程学报, no. 14, 23 July 2020 (2020-07-23) * |
| 关泉珍;鲍泓;史志坚;: "基于A*算法的驾驶地图路径规划实现", 北京联合大学学报(自然科学版), no. 02, 20 April 2016 (2016-04-20) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114924259B (en) | 2024-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111563442B (en) | Slam method and system for fusing point cloud and camera image data based on laser radar | |
| CN114782691B (en) | Robot target identification and motion detection method based on deep learning, storage medium and equipment | |
| CN109165549B (en) | Road identification obtaining method based on three-dimensional point cloud data, terminal equipment and device | |
| US9946960B1 (en) | Method for acquiring bounding box corresponding to an object in an image by using convolutional neural network including tracking network and computing device using the same | |
| US9953437B1 (en) | Method and device for constructing a table including information on a pooling type and testing method and testing device using the same | |
| CN113936198B (en) | Low-beam laser radar and camera fusion method, storage medium and device | |
| CN108845569A (en) | Generate semi-automatic cloud method of the horizontal bend lane of three-dimensional high-definition mileage chart | |
| CN112215128A (en) | FCOS-fused R-CNN urban road environment identification method and device | |
| CN111880191B (en) | Map generation method based on multi-agent laser radar and visual information fusion | |
| CN114155265A (en) | Three-dimensional laser radar road point cloud segmentation method based on YOLACT | |
| CN110532963B (en) | Vehicle-mounted laser radar point cloud driven road marking accurate extraction method | |
| EP4174792A1 (en) | Method for scene understanding and semantic analysis of objects | |
| CN114817426A (en) | Map construction device and method | |
| CN115079143B (en) | Multi-radar external parameter quick calibration method and device for double-bridge steering mine card | |
| CN116109601A (en) | Real-time target detection method based on three-dimensional laser radar point cloud | |
| CN113538620A (en) | SLAM mapping result evaluation method oriented to two-dimensional grid map | |
| KR20220013439A (en) | Apparatus and method for generating High Definition Map | |
| Li et al. | Enhancing 3-D LiDAR point clouds with event-based camera | |
| KR20220135186A (en) | Electronic device and control method | |
| CN114924259B (en) | Performance evaluation method for different vehicle-mounted laser radars | |
| CN111353481A (en) | Road obstacle identification method based on laser point cloud and video image | |
| CN116052099A (en) | Small target detection method for unstructured road | |
| CN117029846A (en) | Generalized laser ranging path planning algorithm for mobile robot in complex environment | |
| CN114842660A (en) | Unmanned lane track prediction method and device and electronic equipment | |
| CN114488026A (en) | Underground parking garage passable space detection method based on 4D millimeter wave radar |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |