CN-121999138-A - Three-dimensional reconstruction system of high-dust closed storage environment based on edge-cloud cooperative architecture

CN121999138ACN 121999138 ACN121999138 ACN 121999138ACN-121999138-A

Abstract

High dust seals storage environment three-dimensional reconstruction system based on edge-cloud cooperation framework relates to intelligent storage technical field. The application aims to solve the problems that the existing silo scanning system is difficult to ensure the real-time performance and the integrity of data transmission and the system resources are wasted. The three-dimensional model comprises a plurality of intelligent perception terminals and a cloud management platform, wherein the plurality of intelligent perception terminals can communicate with the cloud management platform, the intelligent perception terminals acquire distance information between a target and an observation point through radar information and fuse the distance information and a radar attitude angle into a data frame and then send the data frame to the cloud management platform, the cloud management platform analyzes the received data frame to obtain a point cloud data stream, and a point cloud processing algorithm is adopted to calculate target coordinates according to the point cloud data stream so as to construct the three-dimensional model.

Inventors

CHENG LIWEI
WU WENXIANG
WEI JIN
LIU ZITAO
Zhou Lunjie

Assignees

哈尔滨工业大学

Dates

Publication Date: 20260508
Application Date: 20260209

Claims (10)

1. The high-dust closed storage environment three-dimensional reconstruction system based on the edge-cloud collaborative architecture comprises a plurality of intelligent perception terminals and a cloud management platform, wherein the intelligent perception terminals are respectively arranged in different silos in the storage environment and can communicate with the cloud management platform, and the system is characterized in that; The intelligent perception terminal acquires distance information of a target and an observation point through radar information, fuses the distance information and a radar attitude angle into a data frame, and sends the data frame to the cloud management platform; And the cloud management platform analyzes the received data frames to obtain point cloud data streams, calculates target coordinates according to the point cloud data streams by adopting a point cloud processing algorithm, and further constructs a three-dimensional model.
2. The three-dimensional reconstruction system of the high-dust closed storage environment based on the edge-cloud cooperative architecture, which is disclosed in claim 1, is characterized in that the intelligent perception terminal comprises a millimeter wave radar module, a double-degree-of-freedom scanning holder, an edge calculation module and a redundant communication module; the millimeter wave radar module is used for transmitting detection waves to a target and receiving echo signals; The double-degree-of-freedom scanning holder is used for adjusting the attitude angle of the millimeter wave radar module; the edge calculation module is used for analyzing the echo signals into distance information and fusing the distance information and the attitude angle into a data frame; And the redundant communication module is used for sending the data frame to the cloud management platform.
3. The three-dimensional reconstruction system for a high dust closed warehouse environment based on an edge-cloud co-architecture according to claim 2, wherein the analyzing the echo signal into distance information comprises: Mixing the echo signal with a transmitting signal to generate a beat signal; performing analog-to-digital conversion on the beat signal to form original digital echo information; Performing one-dimensional fast Fourier transform on the original digital echo information to obtain a distance dimension spectrum; Searching a frequency index of peak amplitude in the distance dimension spectrum as a beat signal frequency of a target; and calculating the distance information between the target and the observation point by using the beat signal frequency.
4. The three-dimensional reconstruction system of a high dust closed warehouse environment based on an edge-cloud collaborative architecture according to claim 3, wherein the performing one-dimensional fast fourier transform on the original digital echo information to obtain a distance-dimensional spectrum comprises: Obtaining a distance dimension spectrum according to : , Wherein, the As the original digital echo information is obtained, For the total number of sample points, For the time-domain index, For the frequency domain index, In imaginary units.
5. The three-dimensional reconstruction system for a high dust closed storage environment based on an edge-cloud cooperative architecture according to claim 3, wherein the calculating the distance information between the target and the observation point by using the beat signal frequency comprises: calculating distance information between the target and the observation point by using the following method : , Wherein, the For the period of the radar sweep, For the swept bandwidth of the radar transmit signal, Is the speed of sound, Is the beat signal frequency.
6. The three-dimensional reconstruction system of the high-dust closed storage environment based on the edge-cloud cooperative architecture according to claim 2, 3, 4 or 5, wherein the intelligent perception terminal further comprises a storage module; The storage module is used for storing the data frames.
7. The three-dimensional reconstruction system of the high-dust closed storage environment based on the edge-cloud collaborative architecture of claim 1, wherein the cloud management platform comprises a data receiving and analyzing module and a three-dimensional reconstruction and calculating module; The data receiving and analyzing module is used for receiving the data frame, analyzing distance information and attitude angles from the data frame and converting the distance information and the attitude angles into a real-time point cloud data stream; the three-dimensional reconstruction and calculation module is used for calculating coordinates of a target according to the real-time point cloud data stream by adopting a point cloud processing algorithm and constructing a material three-dimensional model according to the coordinates.
8. The three-dimensional reconstruction system for a high dust closed warehouse environment based on an edge-cloud co-architecture as set forth in claim 7, wherein said calculating coordinates of an object from said real-time point cloud data stream comprises: The coordinates of the object are calculated according to the following formula: , Wherein, the As the distance information, there is provided a distance information, Is at a horizontal angle, and is at a vertical angle, In order to be a pitch angle, Is the coordinates of the target.
9. The three-dimensional reconstruction system for a high dust closed warehouse environment based on an edge-cloud co-architecture as set forth in claim 7, wherein the constructing a three-dimensional model of materials based on coordinates comprises: removing coordinate noise by adopting a statistical outlier removing algorithm; And reconstructing the three-dimensional curved surface by using a poisson reconstruction algorithm to generate a three-dimensional model.
10. The three-dimensional reconstruction system of the high-dust closed warehouse environment based on the edge-cloud collaborative architecture according to claim 7,8 or 9, wherein the cloud management platform further comprises a data management and visualization module for displaying a three-dimensional model, a historical data curve and an analysis report through a graphical interface.

Description

Three-dimensional reconstruction system of high-dust closed storage environment based on edge-cloud cooperative architecture Technical Field The application belongs to the technical field of intelligent storage, and particularly relates to storage environment modeling. Background In the industrial fields of mineral products, building materials, grain storage and the like, a silo is used as a core storage facility, and accurate and real-time checking of internal materials is important for guaranteeing production continuity, optimizing inventory management and preventing safety accidents. However, in these typical industrial scenarios, the interior of the silo is often filled with high concentrations of dust, water vapor or corrosive gases, creating a harsh environment with low visibility and strong electromagnetic interference. This makes conventional measurement techniques very challenging. Typical technologies currently used for silo level measurement include laser radar, ultrasonic, weight type level gauge, single-point radar and the like. The laser radar has different application scenes and performance performances, and has the advantages that although the laser radar is high in precision, the laser radar is serious in attenuation in a dust environment, measurement failure is easy to cause, ultrasonic waves are easy to be influenced by temperature, humidity and uneven reflection on the surface of a material, and measurement stability is insufficient. Millimeter wave radar has become a solution with potential because of its strong penetration ability to dust and smoke. However, most of the existing millimeter wave radar application schemes adopt a centralized computing architecture, namely, sensing equipment is only responsible for collecting massive original point cloud data and uploading the data to an upper computer or a server at the rear end for unified processing through a wired or wireless network. The architecture has the defects that when the architecture processes a high-resolution and large-scale three-dimensional scanning task, the defects are increasingly remarkable: On one hand, huge pressure is caused on the bandwidth of a communication network on site by massive original point cloud data, and especially when wireless transmission or multi-device networking is adopted, data congestion, packet loss and high delay are easy to occur, and the real-time performance and the integrity of data transmission are difficult to ensure; On the other hand, all complex analysis, filtering, registration and modeling tasks are completely submitted to the back-end processing, so that the computational power requirement on the server is extremely high, the increasingly-enhanced computational capacity of the front-end sensing equipment cannot be fully utilized, and the waste of system resources is caused. Therefore, how to optimize the data processing architecture of the silo scanning system, the dependence on the communication bandwidth is reduced and the overall operation efficiency of the system is improved while the high-precision three-dimensional reconstruction effect is ensured, and the method is very important for promoting the large-scale application of the intelligent warehousing technology. Disclosure of Invention The application aims to solve the problems that the existing silo scanning system is difficult to ensure the real-time performance and the integrity of data transmission and the system resource is wasted, and provides a high-dust closed storage environment three-dimensional reconstruction system based on an edge-cloud cooperative framework. The application provides a high-dust closed storage environment three-dimensional reconstruction system based on an edge-cloud collaborative architecture, which comprises a plurality of intelligent perception terminals and a cloud management platform, wherein the intelligent perception terminals are respectively arranged in different silos in the storage environment, and can communicate with the cloud management platform; The intelligent perception terminal acquires distance information of a target and an observation point through radar information, fuses the distance information and a radar attitude angle into a data frame, and sends the data frame to the cloud management platform; And the cloud management platform analyzes the received data frames to obtain point cloud data streams, calculates target coordinates according to the point cloud data streams by adopting a point cloud processing algorithm, and further constructs a three-dimensional model. In one possible design, the intelligent perception terminal comprises a millimeter wave radar module, a double-freedom-degree scanning cradle head, an edge calculation module and a redundant communication module; the millimeter wave radar module is used for transmitting detection waves to a target and receiving echo signals; The double-degree-of-freedom scanning holder is used for adjusting the attitude ang