Ship collision avoidance navigation system with dynamic obstacle recognition function
By integrating multiple sensors, the ship collision avoidance system solves the limitations of single radar perception and the problem of decreased accuracy of inertial navigation systems, achieving all-weather, high-precision obstacle recognition and collision avoidance decision-making, and improving the safety and efficiency of ship navigation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- NANJING YULI TECH CO LTD
- Filing Date
- 2025-06-17
- Publication Date
- 2026-05-08
AI Technical Summary
Existing ship collision avoidance systems rely on a single radar sensor, which cannot provide high-resolution target images, makes it difficult to identify small targets, and is greatly affected by lighting and weather conditions. The accuracy of the inertial navigation system decreases when the GPS signal is poor, affecting navigation safety.
By integrating solid-state lidar, millimeter-wave radar, vision sensors, AIS receivers, GNSS receivers, inertial measurement units, and central processing units, a multi-sensor fusion system is constructed to leverage the advantages of various sensors and achieve high-precision obstacle recognition in all weather conditions and at all times.
It achieves high-precision, all-weather obstacle recognition capabilities, improves the reliability and adaptability of ship collision avoidance systems, and ensures safe navigation.
Smart Images
Figure CN224216097U_ABST
Abstract
Description
Technical Field
[0001] This utility model relates to the field of navigation technology, and in particular to a ship collision avoidance navigation system with dynamic obstacle recognition function. Background Technology
[0002] Currently, obstacle identification and avoidance in ship navigation primarily rely on single radar devices. While radar performs well in terms of detection range and adverse weather conditions, it has some inherent limitations, as follows:
[0003] 1) Ships rely on a single type of radar (such as X-band or S-band) to monitor their surroundings. However, this single-mode radar cannot provide high-resolution target images, especially in complex close-range environments, and has limited detection capabilities for small targets or low-reflectivity objects. Furthermore, radar struggles to accurately distinguish the specific shape, outline, and motion of targets, posing a challenge to precise obstacle avoidance.
[0004] 2) Using radar as the sole sensing method ignores the advantages of other sensors. These sensors can provide complementary information under different conditions; for example, lidar can provide high-precision distance and shape information, while vision sensors excel at identifying and classifying obstacles. Failing to fully utilize the advantages of multiple sensors limits the overall performance of the system.
[0005] 3) Existing visual recognition systems are easily affected by lighting conditions and weather conditions. In low light or in severe weather (such as fog, rain, snow, etc.), their performance deteriorates significantly and they cannot guarantee stable obstacle recognition capabilities, thus affecting navigation safety.
[0006] 4) Inertial navigation systems perform poorly when GPS signals are weak: When GNSS signals are lost or interfered with (such as under bridges, in canyons, etc.), traditional inertial navigation systems cannot maintain high-precision attitude estimation for a long time, leading to the accumulation of dead reckoning errors and affecting the reliability of collision avoidance systems. Utility Model Content
[0007] This invention provides a ship collision avoidance navigation system with dynamic obstacle recognition function. By integrating a variety of advanced sensors and control modules at the hardware level, it constructs an intelligent collision avoidance system with high precision, all-weather, and all-time perception and decision-making capabilities.
[0008] To achieve the objectives of this utility model, the technical solution adopted is as follows: a ship collision avoidance navigation system with dynamic obstacle recognition function, comprising a solid-state lidar, a millimeter-wave radar, a visual sensor, an AIS receiver, a GNSS receiver, an inertial measurement unit, and a central processing unit. The solid-state lidar transmits the shape, outline, size, and distance information of obstacles detected within several hundred meters to the central processing unit; the millimeter-wave radar transmits the position and distance of obstacles within a range of several kilometers to the central processing unit; the visual sensor provides obstacle image information under good lighting and low lighting conditions and transmits it to the central processing unit; the AIS receiver receives broadcast information from nearby ships equipped with AIS devices and transmits it to the central processing unit; the GNSS receiver provides the ship's absolute geographical location, SOG / COG, and provides a precise timestamp, which is transmitted to the central processing unit; the inertial measurement unit provides the ship's acceleration and angular velocity information, and helps calculate the ship's speed and direction changes when GPS signals are poor.
[0009] As an optimized solution of this utility model, the central processing unit includes an STM32F103 controller.
[0010] As an optimized solution of this utility model, the visual sensor includes a visible light camera and a thermal infrared camera.
[0011] As an optimized solution of this utility model, the inertial measurement unit includes a 6-axis MEMS motion sensor. The 9th pin of the 6-axis MEMS motion sensor is connected to the 34th pin of the STM32F103 controller, the 25th pin of the 6-axis MEMS motion sensor is connected to the 33rd pin of the STM32F103 controller, and the 26th pin of the 6-axis MEMS motion sensor is connected to the 32nd pin of the STM32F103 controller.
[0012] As an optimized solution of this utility model, the GNSS receiver is a Beidou GNSS receiver.
[0013] As an optimized solution of this utility model, the ship collision avoidance navigation system with dynamic obstacle recognition function also includes a multibeam echo sounder, which is used to identify underwater obstacles and semi-submersible objects approaching the water surface.
[0014] This utility model has the following positive effects: 1) This utility model has multiple sensors covering air, water surface, underwater, long distance, short distance, day and night, and various weather conditions. The Beidou GNSS+IMU combined navigation achieves high-precision and high-stability positioning. Various sensors complement each other, and the system has the ability to work in all weather and all time periods. The STM32 main control platform supports rapid data acquisition and processing, ensuring timely early warning and avoidance. The multi-source redundancy design avoids system paralysis caused by the failure of a single sensor.
[0015] 2) This invention, through a hardware-level multi-sensor fusion design, fully leverages the performance advantages of various sensors, overcoming the limitations of traditional single radar systems, and constructs a high-precision, dynamic obstacle recognition, all-weather ship collision avoidance and navigation system. This is not only reflected in its comprehensive ability to recognize dynamic obstacles, but also in its advanced hardware architecture design, which achieves high reliability, strong adaptability, and intelligence, providing a solid guarantee for the safe navigation of ships. Attached Figure Description
[0016] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.
[0017] Figure 1 This is a schematic diagram of the principle of this utility model;
[0018] Figure 2 This is the circuit schematic diagram of the central processing unit of this utility model;
[0019] Figure 3 This is the circuit diagram of the inertial measurement unit of this utility model. Detailed Implementation
[0020] like Figure 1 As shown, this utility model discloses a ship collision avoidance navigation system with dynamic obstacle recognition function, including a solid-state lidar, a millimeter-wave radar, a visual sensor, an AIS receiver, a GNSS receiver, an inertial measurement unit, and a central processing unit. The solid-state lidar transmits the shape, outline, size, and distance information of obstacles within a few hundred meters to the central processing unit; the millimeter-wave radar transmits the position and distance of obstacles within a range of several kilometers to the central processing unit; the visual sensor provides obstacle image information under good lighting and low lighting conditions and transmits it to the central processing unit; the AIS receiver receives broadcast information from nearby ships equipped with AIS devices and transmits it to the central processing unit; the GNSS receiver provides the ship's absolute geographical location, SOG / COG, and provides a precise timestamp, which is transmitted to the central processing unit; the inertial measurement unit provides the ship's acceleration and angular velocity information, and helps calculate the ship's speed and direction changes when GPS signals are poor.
[0021] The system utilizes the Hesai XT32 solid-state LiDAR, providing high-precision, high-resolution short-range (typically within several hundred meters) 3D point cloud data to accurately detect the shape, contour, size, and distance of obstacles. Its hardware-embedded motion distortion correction algorithm enables dynamic recognition. The millimeter-wave radar offers mid-to-long-range (up to several kilometers) obstacle detection capabilities, accurately measuring the relative velocity (radial velocity) and distance of obstacles. It is capable of all-weather operation and outperforms solid-state LiDAR and visual sensors in adverse visibility conditions such as fog, rain, snow, and smoke. The millimeter-wave radar is the Simrad Halo20+. The visual sensors include a visible light camera and a thermal infrared camera. The visible light camera is a BFS-U3-51S5C-C-FLIR camera, and the thermal infrared camera is a FLIR Boson 640. The visible light camera provides rich scene information (color, texture, shape), which is crucial for identifying obstacle types (such as ship type, floating objects, small boats, animals), reading markers (such as buoy numbers, light signals), and understanding scene semantics. The thermal infrared camera performs best in good lighting conditions, detecting heat-generating targets (such as ship engines and aquatic animals) in complete darkness, fog, or smoke, without relying on visible light. The AIS receiver receives broadcast information from nearby ships equipped with AIS devices, detecting distant targets outside the radar / lidar detection range (especially in open waters), buying time for avoidance decisions. The AIS receiver is the FURUNO FA-30 AIS receiver.
[0022] Millimeter-wave radar is responsible for long-range coarse perception and motion status judgment; solid-state lidar is responsible for short-range high-precision modeling and obstacle recognition; the combination of the two improves the resolution and reliability of overall environmental perception, especially significantly enhancing obstacle avoidance capabilities in complex near-shore environments. Lidar supplements radar's shortcomings in target shape recognition, providing structured information; although visual sensors are greatly affected by lighting conditions, they can provide rich semantic information, such as vessel type, color, and navigation light status, during the day and in clear weather; AIS receivers acquire real-time navigation information of other vessels (position, speed, heading, name, etc.), compensating for sensor blind spots, especially at night or in poor visibility conditions, still enabling the monitoring of surrounding vessels.
[0023] The GNSS receiver is a BeiDou GNSS receiver. It uses the BeiDou satellite navigation system as the primary positioning source to improve positioning accuracy and reliability. This receiver can provide the ship's high-precision three-dimensional geographical location (longitude, latitude, altitude), speed (SOG), heading (COG), and accurate timestamp information (UTC time) in real time, and transmit this information to the central processing unit. The time synchronization signal provided by the GNSS receiver can be used for time unification of the multi-sensor system, ensuring the consistency of the timing of data acquisition and processing from each sensor, enabling all-weather detection, and improving the overall system's collaborative performance. The BeiDou GNSS receiver is a UFirebird UC6226.
[0024] In the event of interference or temporary loss of GNSS signals (such as when crossing bridges, entering canyons, or densely built-up areas), the Inertial Measurement Unit (IMU) will take over the attitude and motion estimation tasks, forming a combined navigation system with GNSS to maintain the continuity and stability of the system.
[0025] Ship collision avoidance navigation systems with dynamic obstacle recognition capabilities also include multibeam echo sounders (MDS). MDS are used to identify underwater obstacles and semi-submersible objects approaching the surface. MDS can generate detailed seabed topographic maps, helping to identify potential hazards such as unmarked shipwrecks, reefs, and other underwater obstacles. By continuously scanning the waters below and around the vessel, MDS can detect any suddenly appearing underwater obstacles in real time and promptly warn the crew to take appropriate evasive action. For objects partially above or only slightly below the waterline (such as containers and whales), which may be difficult to detect with traditional radar systems, especially in poor sea conditions, the presence and location of such objects can be effectively detected through sound wave reflection, thereby improving navigational safety. Sonar technology can operate stably in various weather conditions, including severe weather environments such as fog, rain, and snow. This ensures effective monitoring of the surrounding environment even in extremely low visibility conditions.
[0026] like Figure 2 As shown, the central control module includes an STM32F103 controller. Specifically, it is an STM32F103C8T6 microcontroller. The STM32F103C8T6 microcontroller serves as the control module. It has an ARM Cortex-M3 core, a 32-bit CPU, 64K of memory, a 72MHz system clock, and a 10×12-bit analog-to-digital converter (A / D).
[0027] like Figure 3As shown, the inertial measurement unit (IMU) includes a 6-axis MEMS motion sensor. Pin 9 of the 6-axis MEMS motion sensor is connected to pin 34 of the STM32F103 controller, pin 25 is connected to pin 33 of the STM32F103 controller, and pin 26 is connected to pin 32 of the STM32F103 controller. The 6-axis MEMS motion sensor is a 6-axis motion processing component, specifically an MPU6050, which integrates a 3-axis gyroscope and a 3-axis accelerometer. It measures the ship's angular velocity and linear acceleration. When GNSS signals are briefly lost (e.g., under bridges, in canyons) or interfered with, it provides continuous attitude (roll, pitch, bow) and heading (HDG) information to assist in trajectory calculation and ensure the continuity of positioning and attitude data. Deep fusion with GNSS (integrated navigation) can improve overall positioning accuracy and reliability, and assist in calculating changes in the ship's speed and direction, especially when GPS signals are unstable. During GNSS outages (such as when crossing bridges, traversing canyons, or experiencing interference), inertial measurement unit (IMU) data continuously provides position, velocity, heading, and attitude estimates, unlike pure GNSS systems which completely lose their positioning capabilities. This is crucial for the safe navigation of ships. Furthermore, IMU data assists in velocity / direction calculations; when GPS signal instability causes jumps or unreliable speed and heading readings, the raw acceleration and angular velocity data provided by the IMU become the most reliable basis for calculating instantaneous velocity and direction changes.
[0028] The GNSS receiver is a BeiDou GNSS receiver. When used in conjunction with other sensors on the ship (such as radar, AIS, and lidar), a comprehensive collision avoidance and navigation system can be constructed. This integrated approach leverages the precise position data from the BeiDou system and the advantages of other sensors to achieve comprehensive perception of the surrounding environment and effective identification of dynamic obstacles, significantly improving the safety and efficiency of ship collision avoidance.
[0029] Working principle;
[0030] 1. Data collection;
[0031] Solid-state LiDAR (Hesai XT32): Responsible for acquiring high-precision 3D point cloud data at close range (usually within several hundred meters), providing information on the shape, outline, size, and distance of obstacles.
[0032] Millimeter-wave radar (Simrad Halo20+): Used for detecting the position and relative velocity of obstacles at medium to long distances (up to several kilometers), suitable for detection under all weather conditions.
[0033] Visual sensors:
[0034] Visible light camera (BFS-U3-51 S5C-C-FLIR camera): Provides rich scene information, such as color, texture, and shape, under good lighting conditions, which helps to identify obstacle types.
[0035] Thermal infrared camera (FLIR Boson 640): Effectively detects heat-generating targets in low light or adverse weather conditions (such as fog, rain, and snow), without relying on visible light.
[0036] AIS Receiver (FURUNO FA-30): Receives broadcast information from nearby ships equipped with AIS devices, supplementing blind spots of other sensors and enabling the detection of distant targets in open sea areas.
[0037] Beidou GNSS receiver (UFirebird UC6226): Provides the ship's precise geographical location, speed (SOG), heading (COG), and UTC timestamp.
[0038] Inertial Measurement Unit (MPU6050): Provides acceleration and angular velocity information when GPS signal is poor, assisting in the calculation of changes in the ship's speed and direction.
[0039] Multibeam echo sounder: Specifically designed to identify underwater obstacles and semi-submersible objects approaching the surface, ensuring navigational safety.
[0040] Central processing unit: Data from all sensors converges here and is fused and analyzed using advanced algorithms.
[0041] Time synchronization: The precise timestamps provided by the BeiDou GNSS receiver are used as a reference to ensure time consistency among the sensors.
[0042] 2. Dynamic obstacle recognition and collision avoidance decision-making;
[0043] Real-time monitoring and early warning: The system monitors the surrounding environment in real time based on the fused data, identifies potential collision risks, and reacts quickly. For example, when an obstacle is detected ahead, the system assesses its position, speed, and possible collision path, and then suggests or automatically executes appropriate avoidance measures.
[0044] Integrated navigation: When GNSS signals are interfered with or lost, the IMU takes over the attitude and motion estimation tasks, forming an integrated navigation system with GNSS to maintain the continuity and stability of the system.
[0045] 3. User interface and controls;
[0046] Human-computer interaction interface: Presents the analysis results to the user, allowing the operator to view the current navigation status, surrounding environment, and any warning messages that require immediate attention.
[0047] Millimeter-wave radar enables dynamic target velocity and position perception, solid-state lidar performs high-precision dynamic obstacle identification, visual sensors identify dynamic target types, and AIS receivers extend the dynamic perception range.
[0048] In summary, this ship collision avoidance navigation system integrates multiple advanced sensing technologies and BeiDou GNSS positioning services to construct an efficient and reliable dynamic obstacle recognition and collision avoidance mechanism, which greatly improves the safety and efficiency of maritime navigation.
[0049] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this utility model. It should be understood that the above descriptions are merely specific embodiments of this utility model and are not intended to limit this utility model. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this utility model should be included within the protection scope of this utility model.
Claims
1. A ship collision avoidance navigation system with dynamic obstacle recognition function, characterized in that: The system includes a solid-state lidar, millimeter-wave radar, a visual sensor, an AIS receiver, a GNSS receiver, an inertial measurement unit, and a central processing unit. The solid-state lidar transmits the shape, outline, size, and distance information of obstacles within a few hundred meters to the central processing unit. The millimeter-wave radar transmits the position and distance of obstacles within a few kilometers to the central processing unit. The visual sensor provides obstacle image information under good and low-light conditions and transmits it to the central processing unit. The AIS receiver receives broadcast information from nearby ships equipped with AIS devices and transmits it to the central processing unit. The GNSS receiver provides the ship's absolute geographical location, SOG / COG, and provides a precise timestamp, which is transmitted to the central processing unit. The inertial measurement unit provides the ship's acceleration and angular velocity information, helping to calculate the ship's speed and direction changes when GPS signals are poor.
2. A ship collision avoidance navigation system with dynamic obstacle recognition function according to claim 1, characterized in that: The central processing unit includes an STM32F103 controller.
3. A ship collision avoidance navigation system with dynamic obstacle recognition function according to claim 2, characterized in that: The visual sensors include a visible light camera and a thermal infrared camera.
4. A ship collision avoidance navigation system with dynamic obstacle recognition function according to claim 3, characterized in that: The inertial measurement unit includes a 6-axis MEMS motion sensor. Pin 9 of the 6-axis MEMS motion sensor is connected to pin 34 of the STM32F103 controller. Pin 25 of the 6-axis MEMS motion sensor is connected to pin 33 of the STM32F103 controller. Pin 26 of the 6-axis MEMS motion sensor is connected to pin 32 of the STM32F103 controller.
5. A ship collision avoidance navigation system with dynamic obstacle recognition function according to claim 4, characterized in that: The GNSS receiver is a BeiDou GNSS receiver.
6. A ship collision avoidance navigation system with dynamic obstacle recognition function according to claim 5, characterized in that: The ship collision avoidance navigation system with dynamic obstacle recognition function also includes a multibeam echo sounder, which is used to identify underwater obstacles and semi-submersible objects approaching the water surface.