Ship stern wave identification and removal method based on PointNet network
A technology for ships and networks, applied in the field of environmental perception, can solve problems affecting the accurate planning of obstacle avoidance paths for unmanned boats, and achieve the effect of improving detection capabilities, being easy to identify, and ensuring accuracy.
Pending Publication Date: 2021-03-09
航天时代(青岛)海洋装备科技发展有限公司
View PDF7 Cites 11 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The wake waves generated by surface target ships can increase the detection probability and detection distance of targets, but it will also increase the positioning error of surface target ships, thus affecting the accurate planning of obstacle avoidance paths for unmanned vehicles
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0085]S310 is mounted in the height of the unmanned boat mast, and there is no shield in the range of 360 °.
[0086]The S320 marine computer runs laser radar driver connects laser radar equipment, and collects three-dimensional dot cloud data in the range of 360 ° range in real time.
[0087]The S330 performs a denoising treatment based on point cloud density statistics to remove the current point cloud data, remove discrete noise points in the laser radar point cloud data.
[0088]The S340 is divided by the laser radar point cloud to remove the noise point, dividing the point cloud into a number of independent point clouds, as a suspicious objective, and thus the target point cloud, boundary range, target center for suspicious objectives Point, target radius and other information for statistical analysis.
[0089]The S350 marine computer automatically loads the well-trained PointNet depth learning model and weight parameters, and sets to the reasoning mode, in each point cloudset, looking for...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a ship stern wave identification and removal method based on a PointNet network. The method comprises the following steps: S1, acquiring point cloud data of an unmanned ship ina 360-degree range in real time by using a laser radar; s2, carrying out denoising processing on the point cloud data; s3, performing Euclidean distance clustering on the denoised point cloud data toobtain a suspicious target list; s4, constructing a PointNet deep learning model, and loading the trained weight parameters; s5, inputting the clustered point cloud subsets into a PointNet deep learning model for classification and identification; s6, removing the point cloud subset identified as the tail wave target from the suspicious target list; s7, judging whether each tail wave target is atail wave generated by the unmanned ship or tail waves generated by other ships, and positioning the water surface ship target through the position of the tail wave under the condition that the tail wave target is not the tail wave of the unmanned ship; and S8, sending a water surface obstacle detection result to the unmanned ship control platform. According to the method, the PointNet deep learning model is utilized to achieve rapid identification and removal of the stern wave of the boat, and the target boat is positioned based on the stern wave identification result, so that the adaptive capacity of the laser radar in water surface application is improved.
Description
Technical field[0001]The present invention relates to a boat-based energizing and removal method based on a POINTNET network, which belongs to the field of environmental perception.Background technique[0002]Laser radar is one of the most important and indispensable sensors in automatic driving implementation. Its importance is self-evident, such as obstacle detection, road edge detection, map buildings, etc. are inseparable from it. The boat tail wave can be detected by the laser radar by reflecting the laser radar, but the somatic waves do not belong to the water barrier, which does not affect the sailing route of the boat, so it needs to be filtered from the target of laser radar detection. The lave produced by the unknown itself can be determined behind the boat, but its change in the voyage speed of the boat is also real-time. By filtering the dot cloud within a certain range of boats, the elongated waves produced by the unmanned boat can be removed, but the filtering range need...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62
CPCG06F18/23213G06F18/214
Inventor 李清洲刘新新杨长坤胡常青刘柳
Owner 航天时代(青岛)海洋装备科技发展有限公司