CN-121978730-A - Multi-source data fusion-based bulk cargo wharf operation area personnel vehicle position monitoring system

CN121978730ACN 121978730 ACN121978730 ACN 121978730ACN-121978730-A

Abstract

The invention discloses a multi-source data fusion-based system for monitoring a person and vehicle in a bulk cargo wharf operation area, which comprises a heterogeneous sensing layer, a reliable transmission layer, an intelligent fusion and decision-making layer and a panoramic application layer, wherein the heterogeneous sensing layer is used for collecting multi-element heterogeneous data of a person and vehicle target through a UWB positioning unit, a GNSS/INS combined positioning unit, an AI visual unit and a regional sensing unit, the reliable transmission layer is used for transmitting the multi-element heterogeneous data to a processing center in a highly reliable and low-time-delay manner through a heterogeneous fusion network, the intelligent fusion and decision-making layer comprises a space-time unified gateway, a self-adaptive confidence evaluation and fusion engine, a dynamic digital twin model and a predictive early warning engine, the predictive early warning engine is used for performing risk early warning, and the panoramic application layer is used for realizing visualization, interaction and decision support of a monitored scene. The invention can realize a full-scene, high-precision, high-reliability and predictable personnel-vehicle collaborative safety monitoring system, and essentially improves the active safety prevention and control capability and the operation efficiency of the wharf.

Inventors

XU WENPENG
YANG JINKAI
HE DAKAI
LIU JIE
XU JIE
YANG YANG

Assignees

中交机电工程局有限公司

Dates

Publication Date: 20260505
Application Date: 20251230

Claims (10)

1. The system is characterized by comprising a heterogeneous sensing layer, a reliable transmission layer, an intelligent fusion and decision layer and a panoramic application layer; The heterogeneous sensing layer acquires multi-component heterogeneous data of a human-vehicle target through a UWB positioning unit, a GNSS/INS combined positioning unit, an AI visual unit and a region sensing unit; the reliable transmission layer is used for transmitting the multi-element heterogeneous data to the processing center in a highly reliable and low-time delay manner through the heterogeneous fusion network; The intelligent fusion and decision layer comprises a space-time unified gateway, a self-adaptive confidence assessment and fusion engine, a dynamic digital twin model and a predictive early warning engine, wherein the predictive early warning engine is used for assessing the confidence of a data source, carrying out multi-layer data fusion, driving twin update and carrying out risk early warning based on track prediction; and the panoramic application layer is used for realizing the visualization, interaction and decision support of the monitoring scene.
2. The multi-source data fusion-based bulk terminal workplace personnel location monitoring system of claim 1, wherein the UWB positioning unit comprises a UWB positioning base station deployed at a critical risk focus, a UWB personnel positioning tag worn on a person, and a UWB vehicle-mounted positioning unit mounted on a mobile vehicle.
3. The multi-source data fusion-based bulk cargo terminal workplace personnel car position monitoring system according to claim 2, wherein the GNSS/INS combined positioning unit comprises a multi-frequency GNSS receiving module and a six-axis inertial measurement unit mounted on a moving vehicle.
4. A bulk terminal operation area personnel parking spot monitoring system based on multi-source data fusion according to claim 3, characterized in that the AI vision unit comprises an intelligent network camera and/or a thermal imaging camera with AI algorithm cards built in.
5. The multi-source data fusion-based bulk cargo terminal operation area personnel car position monitoring system according to claim 4, wherein the area sensing unit comprises an RFID deployed at the entrance and exit of the enclosed space and an electronic fence sensing cable or a laser radar correlation sensor deployed at a dangerous area.
6. The system for monitoring the personnel and vehicle positions in the bulk cargo wharf operation area based on multi-source data fusion according to claim 1, wherein the reliable transmission layer adopts a redundant architecture combining an optical fiber ring network, a 5G private network slice and an industrial Wi-Fi 6 network, and is provided with an intelligent network manager to realize seamless roaming of the mobile terminal among heterogeneous networks.
7. The system for monitoring the human car position in the bulk cargo wharf operation area based on multi-source data fusion according to claim 1, wherein the self-adaptive confidence assessment and fusion engine is configured to calculate a dynamic confidence score for each data source in real time, wherein the confidence score is comprehensively determined based on the signal quality parameters of the data source, the environmental interference factors and the consistency degree of the data source and other independent data sources, and an iterative estimation algorithm based on confidence weighting is adopted to output the optimal state estimation of the target.
8. The multi-source data fusion-based system for monitoring the human car position in the bulk cargo wharf operation area according to claim 7, wherein the dynamic digital twin body model not only contains static geographic information, but also creates dynamic twin bodies with position, speed, state and confidence attribute for each monitoring target, and supports real-time spatial relation calculation and visualization among twin bodies.
9. The system for monitoring the personnel car position in the bulk cargo wharf operation area based on multi-source data fusion according to claim 8 is characterized in that the predictive early warning engine is configured to establish a short-term motion track predictive model for each moving target, calculate the nearest meeting distance CPA and the nearest point reaching time TCPA between any two target predictive tracks, dynamically set a safety threshold according to the target type and the operation scene, and trigger hierarchical early warning based on comparison of the CPA and the TCPA with the safety threshold.
10. The multi-source data fusion-based bulk cargo terminal operation area personnel parking space monitoring system according to claim 9, wherein the hierarchical early warning comprises sending warning information to a vehicle-mounted or personnel terminal of a dangerous target, sending warning information to a monitoring center, and sending a deceleration suggestion instruction to a power control system or an auxiliary braking system of the target vehicle through a predefined interface when the emergency risk is determined.

Description

Multi-source data fusion-based bulk cargo wharf operation area personnel vehicle position monitoring system Technical Field The invention relates to the technical field of industrial Internet of things and intelligent safety monitoring, in particular to a system for monitoring a personal vehicle position in a bulk cargo wharf operation area based on multi-source data fusion. Background Bulk cargo wharf is a key node of a logistics chain, and an operation area of the bulk cargo wharf has four characteristics of large area (comprising a storage yard, a road, a front edge, a silo and the like), element impurities (flowing machines, transport vehicles, operators and fixed facilities are staggered), strong dynamic state (continuous change of loading and unloading processes and real-time updating of the form of a cargo stack), and multiple blind areas (a visual blind area, a signal shielding area and a bad weather influence area). Traditional security management modes face serious challenges: Passive response mainly depends on fixed video monitoring manual staring screen to communicate with the interphone, and the accident can be traced back, so that the prevention capability is weak. And the vehicle dispatching system, the access control system and the video system are mutually independent to form an information chimney, and global evaluation on the human-vehicle-environment interaction risk cannot be carried out. The positioning technology is bottleneck in that GNSS generates multipath effect due to pile shielding in a storage yard, accuracy is suddenly reduced to more than 5 meters, and the GNSS is completely invalid in a room (such as a transfer station). Although the UWB has high precision, the UWB is easy to be subjected to multipath interference in an open metal environment (such as a flat storage yard), and the full coverage deployment cost is too high. Visual recognition is greatly affected by light (glare, night), weather (rain, fog, dust), occlusion, and cannot provide accurate absolute coordinates. RFID/ZigBee is only region existence detection, and cannot meet the dynamic tracking requirement. Therefore, an intelligent monitoring solution is needed that can adapt to environmental changes, comprehensively utilize the advantages of multi-source information, and realize cost and performance balance. Disclosure of Invention The invention aims to solve the defects of the prior art and provides a bulk cargo wharf operation area personnel vehicle position monitoring system based on multi-source data fusion. The invention adopts the following technical scheme to realize the aim: the system comprises a heterogeneous sensing layer, a reliable transmission layer, an intelligent fusion and decision layer and a panoramic application layer; The heterogeneous sensing layer acquires multi-component heterogeneous data of a human-vehicle target through a UWB positioning unit, a GNSS/INS combined positioning unit, an AI visual unit and a region sensing unit; the reliable transmission layer is used for transmitting the multi-element heterogeneous data to the processing center in a highly reliable and low-time delay manner through the heterogeneous fusion network; The intelligent fusion and decision layer comprises a space-time unified gateway, a self-adaptive confidence assessment and fusion engine, a dynamic digital twin model and a predictive early warning engine, wherein the predictive early warning engine is used for assessing the confidence of a data source, carrying out multi-layer data fusion, driving twin update and carrying out risk early warning based on track prediction; and the panoramic application layer is used for realizing the visualization, interaction and decision support of the monitoring scene. The UWB positioning unit comprises UWB positioning base stations deployed at key risk focuses, UWB personnel positioning labels worn on personnel and UWB vehicle-mounted positioning units mounted on mobile vehicles; the GNSS/INS combined positioning unit comprises a multi-frequency GNSS receiving module and a six-axis inertial measurement unit which are installed on a moving vehicle. The AI visual unit includes an intelligent network camera and/or a thermal imaging camera with an AI computing card built in. The area sensing unit comprises RFID (radio frequency identification devices) deployed at the entrance and the exit of the closed space and an electronic fence sensing cable or a laser radar correlation sensor deployed in a dangerous area. The reliable transmission layer adopts a redundant architecture combining an optical fiber ring network, a 5G private network slice and an industrial Wi-Fi 6 network, and is provided with an intelligent network manager to realize seamless roaming of the mobile terminal among heterogeneous networks. The self-adaptive confidence assessment and fusion engine is configured to calculate a dynamic confidence score for each data source in real time, the confidence score is compr