Shipborne intelligent radar monitoring system

Abstract

The invention relates to the technical field of ship navigation and safety monitoring, in particular to a shipborne intelligent radar monitoring system, which comprises a radar detection module used for transmitting radar signals, receiving target echoes, carrying out 360-degree scanning in a range of a radius of 0-50 sea miles around a ship, and outputting original radar data containing information such as target position, distance and reflection intensity; the radar detection module adopts an X-band or S-band phased array radar and has adaptive scanning capability, the detection precision is high, the phased array radar is combined with a constant false alarm rate algorithm, interference such as sea clutters and rain and snow clutters is effectively inhibited, the accuracy and stability of target detection are improved through AIS information fusion, and the detection precision is improved. An improved YOLOv5 deep learning algorithm model is introduced, automatic identification of multiple target types is achieved, the identification accuracy is larger than or equal to 92%, the identification response time is smaller than or equal to 0.5 second, and the judgment burden of sailors is relieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ship navigation and safety monitoring, and particularly relates to a shipborne intelligent radar monitoring system. BACKGROUND

[0002] In the process of ship navigation, real-time monitoring and accurate judgment of targets in the surrounding sea area are the key to ensuring navigation safety. The traditional shipborne radar system mainly relies on manual operation and judgment, and has many limitations. In complex sea conditions, the traditional radar is easily disturbed by sea clutter, rain and snow, resulting in low target detection accuracy, and it is difficult to accurately distinguish real targets from interference signals, and false alarms or missed alarms may occur. At the same time, the traditional system lacks the ability to automatically identify target types, and the crew needs to judge whether the target is a merchant ship, a fishing boat, a buoy or an island, etc. In a multi-target scene, the workload of the crew is greatly increased, and the judgment accuracy is easily affected by human factors.

[0003] In terms of threat assessment, the traditional method mainly relies on the experience of the crew, and is highly subjective. In the case of multiple targets appearing at the same time, it is difficult to quickly and accurately assess the collision risk, which may lead to untimely avoidance or decision-making errors. In addition, the traditional system has low information fusion degree with other navigation devices such as AIS (Automatic Identification System), and the data utilization rate is not high, which cannot form a unified and comprehensive ship surrounding situation awareness, and is not conducive to the crew to make comprehensive judgment and decision.

[0004] With the development of the shipping industry and the progress of technology, the requirements for the intelligence and automation of the shipborne radar monitoring system are becoming higher and higher, and the existing technology has been difficult to meet the needs of modern ship navigation safety. Therefore, it is of great significance to develop a shipborne intelligent radar monitoring system with high-precision detection, intelligent identification, dynamic assessment and auxiliary decision-making functions. SUMMARY

[0005] The purpose of the present application is to solve the problems existing in the prior art and to provide a shipborne intelligent radar monitoring system.

[0006] In order to achieve the above purpose, the present application adopts the following technical scheme: a shipborne intelligent radar monitoring system, comprising:

[0007] A radar detection module for transmitting radar signals and receiving target echoes, scanning 360 degrees within a radius of 0-50 nautical miles around the ship, and outputting raw radar data containing target position, distance, reflection intensity and other information; the radar detection module adopts an X-band or S-band phased array radar, has adaptive scanning capability, can dynamically adjust the scanning frame rate according to the target threat level, the range is 5-10 Hz, and can dynamically adjust the transmission power;

[0008] An AIS information receiving module receives AIS information broadcasted by surrounding ships through a VHF frequency band, and analyzes to obtain static information of the ships, including ship name, size, type, etc., and dynamic information, such as latitude and longitude, heading, speed, etc.

[0009] A data fusion and tracking module is connected with the radar detection module and the AIS information receiving module, respectively, performs clutter suppression on original radar data, reduces sea clutter, rain and snow clutter and other interference by using a constant false alarm rate (CFAR) algorithm, extracts target point tracks, performs space-time correlation between radar target point tracks and AIS information, realizes data fusion through matching target time stamp error within ≤1 second and position error within ≤10 meters, establishes a target motion model by using a Kalman filtering algorithm, can simultaneously and stably track not less than 200 targets, outputs continuous and smooth target tracking trajectories, and target tracking position error is ≤5 meters.

[0010] An intelligent target identification module is connected with the data fusion and tracking module, extracts motion characteristics of target tracking trajectories, such as speed range, heading stability, acceleration change rate, and radar echo characteristics, including fluctuation law of reflection intensity over time and radar cross section (RCS) value, classifies and identifies target types by using an improved YOLOv5 algorithm model based on deep learning after training, the identified types cover merchant ships, fishing boats, passenger ships, buoys, islands, etc., identification accuracy is ≥92%, and identification response time is ≤0.5 seconds.

[0011] A threat assessment module is connected with the intelligent target identification module, calculates a threat level by using a fuzzy comprehensive evaluation method based on parameters such as relative distance between the target and the ship, minimum encounter distance (DCPA), time to closest point of approach (TCPA), relative heading angle and target type, the threat level is divided into four levels of no threat, low threat, medium threat and high threat; wherein the high threat determination condition is DCPA < 1 nautical mile and TCPA < 10 minutes, or the target type is a dangerous goods ship and the relative speed is > 15 knots.

[0012] A warning and decision module is connected with the threat assessment module, triggers multi-level warnings when the threat level reaches a preset threshold, only marks the target with blue color on the display interface for low threat, marks with yellow color and starts sound and light alarms for medium threat, and the interval between the beeps and the flashing indicator light is 2 seconds, marks with red color, starts high-frequency beeping for high threat, and the interval is 0.5 seconds, accompanied by voice prompts; at the same time, generates 2-3 avoidance schemes based on international collision avoidance rules and target motion trend, each scheme contains a recommended heading adjustment value in the range of 5-30 degrees and a recommended speed adjustment value within ±5 knots, and adjusts the predicted DCPA, TCPA and collision avoidance success rate through motion simulation.

[0013] The control and storage module is connected with each of the above modules, and is used for receiving user operation instructions, adjusting radar range, range being 0.5-50 nautical miles, and scanning frame rate, early warning threshold and other parameters, supporting manual locking of a key target; and storing original radar data, AIS information, target trajectory, 1 record per second, identification result, threat assessment record and operation log, a storage period being not less than 3 months, a storage medium being an encrypted solid-state storage, and supporting data export through an Ethernet interface.

[0014] Preferably, the environmental perception unit is further included, which is connected with the data fusion and tracking module, and is used for collecting wind speed, wind direction and visibility information; in a strong wind and wave weather, automatically enhancing the clutter suppression strength, in a low visibility, i.e. less than 2 nautical miles, reducing the DCPA threshold of a high threat level from 1 nautical mile to 0.8 nautical mile, inputting the environmental parameters into the data fusion and tracking module, and dynamically adjusting the radar related parameters, so as to improve the adaptability of the system in a complex environment.

[0015] Preferably, the control and storage module adopts a 19-inch high-definition touch screen as a human-computer interaction interface, which can simultaneously display a target situation general map, including the position and heading vector of the ship and surrounding targets, radar original scanning data, a threat level heat map marked with red, yellow and blue colors for different threat levels, and a target detailed information panel covering target type, distance, relative motion parameters, threat score and other information.

[0016] Preferably, the phased array radar beam width of the radar detection module is 1.2-1.5 degrees, which can accurately detect targets in a complex sea state and reduce the interference caused by beam diffusion.

[0017] Preferably, the AIS information receiving module conforms to the International Maritime Organization (IMO) AIS standard, the receiving frequency is 161.975 MHz and 162.025 MHz, and the maximum receiving distance is 20 nautical miles, so that the AIS information of surrounding ships can be stably received.

[0018] Preferably, the data fusion and tracking module realizes real-time signal processing based on the FPGA+GPU architecture, the processing delay is less than 100 ms, and the fast response and tracking of the system to the target are ensured.

[0019] Preferably, the intelligent target identification module is deployed on an embedded GPU platform, such as NVIDIA Jetson Xavier, to ensure efficient operation of the deep learning model and realize fast and accurate target identification.

[0020] Preferably, the weights of each parameter in the threat assessment module are determined through statistical analysis of a large amount of ship navigation accident data, so that the threat assessment result is more scientific and reasonable, and is in line with the actual navigation risk.

[0021] Preferably, the avoidance scheme generated by the early warning and decision module is verified through multiple simulation simulations, and the collision avoidance success rate calculation is based on the accurate simulation of the motion parameters of the ship and the target, providing reliable decision basis for the crew.

[0022] Preferably, the encryption solid-state storage of the control and storage module adopts advanced encryption algorithm to prevent data from being illegally stolen or tampered with, ensure the security and integrity of data, and support fast reading and writing of data, meeting the efficient demand of the system for data storage and call.

[0023] Compared with the prior art, the advantages and positive effects of the present application are:

[0024] 1. High detection accuracy: The phased array radar combined with the constant false alarm rate algorithm effectively suppresses sea clutter, rain and snow clutter and other interference, and through fusion with AIS information, the accuracy and stability of target detection are improved, and the target tracking position error is less than or equal to 5 meters.

[0025] 2. Strong intelligent recognition capability: The improved YOLOv5 deep learning algorithm model is introduced to realize automatic recognition of multiple target types, the recognition accuracy is greater than or equal to 92%, the recognition response time is less than or equal to 0.5 seconds, and the judgment burden of the crew is reduced.

[0026] 3. Scientific threat assessment: Based on the weight determined by multiple parameters and a large amount of accident data statistics, the fuzzy comprehensive evaluation method is used for threat level assessment, the result is objective and reasonable, the high threat recognition is accurate and timely.

[0027] 4. Reliable early warning and decision: The multi-level early warning mechanism can remind the crew in time according to the threat level, and the generated avoidance scheme is verified through simulation, providing scientific decision support for the crew and reducing the collision risk.

[0028] 5. High information fusion degree: The radar and AIS information are integrated to form a comprehensive situation awareness, and the control and storage module realizes centralized display, operation and safe storage of information, which is convenient for data backtracking and management.

[0029] 6. Good environmental adaptability: The optional environmental perception unit can dynamically adjust the system parameters according to the environmental changes, improving the performance of the system in complex sea conditions. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical scheme and advantages of the present application more clear and explicit, the present application will be further described in detail below combined with embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, and are not used to limit the present application.

[0031] The application provides a technical solution: a shipborne intelligent radar monitoring system, comprising a radar detection module, an AIS information receiving module, a data fusion and tracking module, an intelligent target identification module, a threat assessment module, an early warning and decision-making module, a control and storage module, and further comprising an environment perception unit. The specific functions of each module are as follows:

[0032] The radar detection module: an X-band or S-band phased array radar is adopted, which can emit radar signals and receive target echoes, and perform 360-degree omnidirectional scanning on a range of 0-50 nautical miles around the ship. The module has adaptive scanning capability, can dynamically adjust the scanning frame rate according to the threat level of the target, and the range is between 5-10 Hz, and can dynamically adjust the transmission power, realize the rational use of energy on the premise of ensuring the detection effect. The raw radar data output by the module contains key information such as target position, distance, and reflection intensity, and the phased array radar beam width is 1.2-1.5 degrees, which can accurately detect targets in complex sea conditions and reduce interference caused by beam diffusion.

[0033] The AIS information receiving module: receives the AIS information broadcast by the surrounding ships through the VHF frequency band. The module conforms to the International Maritime Organization (IMO) AIS standard, the receiving frequency is 161.975MHz and 162.025MHz, the maximum receiving distance is 20 nautical miles, and the information can be stably received. After the received AIS information is parsed, the static information (such as ship name, size, type, etc.) and dynamic information (such as latitude and longitude, heading, speed, etc.) of the ship can be obtained, which is complementary to the radar detection module and improves the comprehensiveness of target information.

[0034] The data fusion and tracking module: connected with the radar detection module and the AIS information receiving module, first processes the raw radar data, uses the constant false alarm rate (CFAR) algorithm to suppress sea clutter, rain and snow clutter and other interference, and extracts target point traces. Then, the radar target point traces are spatio-temporally associated with the AIS information, and data fusion is realized by matching targets with a timestamp error of ≤1 second and a position error of ≤10 meters. The module realizes real-time signal processing based on FPGA+GPU architecture, the processing delay is <100ms, the Kalman filtering algorithm is used to establish the target motion model, and not less than 200 targets can be stably tracked at the same time, the output is a continuous and smooth target tracking trajectory, the target tracking position error is ≤5 meters, and the system ensures fast response and stable tracking of the target.

[0035] Intelligent target recognition module: connected with the data fusion and tracking module, responsible for extracting the feature information of the target, including the motion characteristics of the target tracking trajectory (such as speed range, heading stability, acceleration change rate) and radar echo characteristics (such as the fluctuation law of reflection intensity with time, radar cross section RCS value). This module is deployed on an embedded GPU platform (such as NVIDIA Jetson Xavier), and through the improved YOLOv5 algorithm model based on deep learning after training, the target type is classified and recognized, and the recognition types cover merchant ships, fishing boats, passenger ships, buoys, islands, etc., the recognition accuracy is ≥92%, the recognition response time is ≤0.5 seconds, which guarantees the rapidity and accuracy of target recognition.

[0036] Threat assessment module: connected with the intelligent target recognition module, based on the relative distance between the target and the ship, the minimum encounter distance (DCPA), the time to reach the minimum encounter distance (TCPA), the relative heading angle and the target type, etc. Parameters, use fuzzy comprehensive evaluation method to calculate the threat level. The weight of each parameter is determined through statistical analysis of a large amount of ship navigation accident data, making the threat assessment result more scientific and reasonable. The threat level is divided into four levels: no threat, low threat, medium threat, and high threat. The high threat judgment condition is: DCPA < 1 mile and TCPA < 10 minutes, or the target type is a dangerous goods ship and the relative speed is > 15 knots.

[0037] Early warning and decision-making module: connected with the threat assessment module, when the target threat level reaches the preset threshold, trigger the corresponding multi-level early warning. When the threat is low, only mark the target with blue color on the display interface; when the threat is medium, mark it with yellow color and start the sound and light alarm (2-second interval beeping and flashing indicator light); when the threat is high, mark it with red color, start high-frequency beeping (0.5-second interval) and accompany with voice prompt. At the same time, based on the international collision avoidance rules and the target motion trend, the module generates 2-3 avoidance schemes, each scheme contains recommended heading adjustment value (5-30 degrees) and recommended speed adjustment value (±5 knots), and the scheme is verified through multiple simulation, the collision avoidance success rate is calculated based on the motion parameters of the ship and the target accurate simulation, providing reliable decision-making basis for the crew.

[0038] Control and storage module: connected with each of the above modules, using a 19-inch high-definition touch screen as the human-computer interaction interface, which can display the target situation map (including the position and heading vector of the ship and surrounding targets), radar raw scan data, threat level heat map (red, yellow, blue three color labels) and target detailed information panel (covering target type, distance, relative motion parameters, threat score, etc.). Users can receive operation instructions through this module to adjust radar range (0.5-50 nautical miles), scan frame rate, early warning threshold and other parameters, and support manual locking of key targets. The storage part uses encrypted solid state storage, uses advanced encryption algorithms to prevent data from being illegally stolen or tampered with, and ensures the security and integrity of the data, while supporting fast reading and writing of data. The storage content includes raw radar data, AIS information, target trajectory (1 record per second), identification result, threat assessment record and operation log, with a storage period of not less than 3 months, and supporting data export through the Ethernet interface.

[0039] Environment perception unit: optional, connected with the data fusion and tracking module, used to collect wind speed, wind direction and visibility information. In strong wind and wave weather, automatically enhance the clutter suppression strength, in low visibility (<2 nautical miles), adjust the DCPA threshold of high threat level from 1 nautical mile to 0.8 nautical mile, input the environmental parameters into the data fusion and tracking module, dynamically adjust the radar related parameters, and improve the adaptability of the system in complex environment.

[0040] When the system is working, the radar detection module and the AIS information receiving module collect target information respectively, and obtain target trajectory after processing by the data fusion and tracking module, the intelligent target identification module identifies the target type, the threat assessment module assesses the threat level, the early warning and decision module performs early warning and generates avoidance scheme according to the level, and the control and storage module realizes human-computer interaction and data management, and each module cooperates to ensure the safety of ship navigation. The system can be widely applied to various ships, especially suitable for commercial ships, passenger ships and other ships sailing in complex sea areas, and can effectively improve the navigation safety and automation.

[0041] The above is only a preferred embodiment of the present application, not other forms of the present application, any skilled person in the art can use the above disclosed technical content to make changes or modifications into equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made according to the technical essence of the present application to the above embodiments without departing from the technical solution content of the present application, still belongs to the protection scope of the present application technical solution.

Claims

1. A shipborne intelligent radar monitoring system, characterized in that, include: The radar detection module is used to transmit radar signals and receive target echoes, perform a 360-degree scan of the area around the ship within a radius of 0-50 nautical miles, and output raw radar data containing information such as target position, distance, and reflection intensity. The radar detection module adopts an X-band or S-band phased array radar, has adaptive scanning capability, and can dynamically adjust the scanning frame rate according to the target threat level, with a range of 5-10Hz, while also dynamically adjusting the transmission power. The AIS information receiving module receives AIS information broadcast by surrounding ships via the VHF band and parses it to obtain static information of the ships, including ship name, size, type, etc., as well as dynamic information, such as latitude and longitude, heading, speed, etc. The data fusion and tracking module is connected to the radar detection module and the AIS information receiving module, respectively. It performs clutter suppression on the raw radar data, uses the constant false alarm rate (CFAR) algorithm to reduce interference from sea clutter, rain and snow clutter, etc., and extracts target points. It performs spatiotemporal correlation between radar target points and AIS information, and achieves data fusion by matching targets with timestamp errors ≤1 second and position errors ≤10 meters. It also uses the Kalman filter algorithm to establish a target motion model, which can stably track no less than 200 targets at the same time and output continuous and smooth target tracking trajectories with a target tracking position error ≤5 meters. The intelligent target recognition module, connected to the data fusion and tracking module, extracts the motion characteristics of the target tracking trajectory, such as speed range, heading stability, acceleration rate of change, and radar echo characteristics, including the fluctuation pattern of reflection intensity over time and radar cross-section (RCS) value. It then uses a trained, deep learning-based improved YOLOv5 algorithm model to classify and identify target types, covering merchant ships, fishing vessels, passenger ships, buoys, islands, etc., with an accuracy rate ≥92% and a response time ≤0.5 seconds. The threat assessment module, connected to the intelligent target recognition module, calculates the threat level based on parameters such as the relative distance between the target and the ship, the minimum encounter distance (DCPA), the time to reach the minimum encounter distance (TCPA), the relative heading angle, and the target type, using a fuzzy comprehensive evaluation method. The threat level is divided into four levels: no threat, low threat, medium threat, and high threat. The criteria for determining a high threat are: DCPA < 1 nautical mile and TCPA < 10 minutes, or the target type is a dangerous goods vessel and the relative speed is > 15 knots. The early warning and decision-making module is connected to the threat assessment module. When the threat level reaches a preset threshold, it triggers multi-level early warnings. Low threats are only indicated by a blue target on the display interface; medium threats are indicated by a yellow target and an audible and visual alarm is activated, with a buzzer and flashing indicator light every 2 seconds; high threats are indicated by a red target, a high-frequency buzzer is activated every 0.5 seconds, and a voice prompt is provided. Simultaneously, based on international collision avoidance rules and the target's movement trend, 2-3 avoidance schemes are generated. Each scheme includes a recommended heading adjustment value, ranging from 5 to 30 degrees, and a recommended speed adjustment value, within ±5 knots. The predicted DCPA, TCPA, and collision avoidance success rate are adjusted through motion simulation. The control and storage module is connected to each of the above modules. It is used to receive user operation commands, adjust the radar range (0.5-50 nautical miles), and parameters such as scan frame rate and warning threshold. It supports manual locking of key targets. It also stores raw radar data, AIS information, target trajectory (one record per second), identification results, threat assessment records, and operation logs. The storage period is no less than 3 months. The storage medium is encrypted solid-state storage, and it supports data export via Ethernet interface.

2. The shipborne intelligent radar monitoring system according to claim 1, characterized in that: It also includes an environmental sensing unit, which is connected to the data fusion and tracking module to collect wind speed, wind direction, and visibility information; automatically enhance clutter suppression intensity in windy and wavey weather; and reduce the DCPA threshold for high threat level from 1 nautical mile to 0.8 nautical mile in low visibility (i.e., <2 nautical miles). The environmental parameters are input into the data fusion and tracking module to dynamically adjust the radar-related parameters and improve the system's adaptability in complex environments.

3. The shipborne intelligent radar monitoring system according to claim 1, characterized in that: The control and storage module uses a 19-inch high-definition touch screen as the human-machine interface, which can simultaneously display the overall target situation map, including the position and heading vector of the ship and surrounding targets, raw radar scan data, threat level heat map, different threat levels are marked with red, yellow and blue, and a target detailed information panel, covering target type, distance, relative motion parameters, threat score and other information.

4. The shipborne intelligent radar monitoring system according to claim 1, characterized in that: The phased array radar beamwidth of the radar detection module is 1.2-1.5 degrees, which enables accurate target detection in complex sea conditions and reduces interference caused by beam spread.

5. The shipborne intelligent radar monitoring system according to claim 1, characterized in that: The AIS information receiving module conforms to the International Maritime Organization (IMO) AIS standard, with receiving frequencies of 161.975MHz and 162.025MHz, a maximum receiving distance of 20 nautical miles, and can stably receive AIS information from surrounding ships.

6. The shipborne intelligent radar monitoring system according to claim 1, characterized in that: The data fusion and tracking module is based on an FPGA+GPU architecture to achieve real-time signal processing with a processing latency of <100ms, ensuring the system's rapid response and tracking of the target.

7. The shipborne intelligent radar monitoring system according to claim 1, characterized in that: The intelligent target recognition module is deployed on an embedded GPU platform, such as NVIDIA Jetson Xavier, to ensure the efficient operation of the deep learning model and achieve fast and accurate target recognition.

8. The shipborne intelligent radar monitoring system according to claim 1, characterized in that: The weights of each parameter in the threat assessment module are determined through statistical analysis of a large amount of ship navigation accident data, making the threat assessment results more scientific and reasonable, and more in line with actual navigation risks.

9. The shipborne intelligent radar monitoring system according to claim 1, characterized in that: The collision avoidance scheme generated by the early warning and decision-making module has been verified through multiple simulations. The collision avoidance success rate is calculated based on the accurate simulation of the motion parameters of the ship and the target, providing the crew with a reliable basis for decision-making.

10. The shipborne intelligent radar monitoring system according to claim 1, characterized in that: The encrypted solid-state storage of the control and storage module adopts advanced encryption algorithms to prevent data from being illegally stolen or tampered with, ensuring data security and integrity. At the same time, it supports fast data reading and writing, meeting the system's high-efficiency requirements for data storage and retrieval.