An airborne scene and terrain matching-assisted navigation method and a highly realistic simulation platform

Abstract

This invention relates to the field of aviation navigation and simulation technology, specifically disclosing an airborne scene-terrain matching assisted navigation method and a highly realistic simulation platform. The method involves acquiring measured terrain information via an airborne camera, and collecting SAR image information, altimeter and atmospheric sensor information via radar. A matching algorithm between the camera terrain and radar images is used to match the measured data with pre-stored information, outputting the position correction error for inertial navigation to achieve inertial navigation correction. Even in complex electromagnetic environments without satellites or in electromagnetic silence, combined inertial navigation and matching navigation can still be achieved, ensuring stable flight of the aircraft. Simultaneously, a highly realistic 3D visual simulation platform is constructed, including a digital aircraft, flight control law, inertial navigation, 3D terrain, and various airborne equipment, providing full-process simulation support for the design, implementation, and testing of airborne scene-terrain matching assisted navigation systems. This invention solves the problem of aircraft navigation relying on satellites in complex electromagnetic environments, improves the anti-interference and reliability of navigation, and the simulation platform can effectively reduce the R&D and testing costs of navigation systems and improve R&D efficiency.

Description

Technical Field

[0001] This invention relates to the field of aviation navigation and simulation technology, specifically to an airborne scene and terrain matching-assisted navigation method and a highly realistic simulation three-dimensional visual simulation platform. Background Technology

[0002] Aircraft navigation systems are the core of ensuring flight safety and achieving precise flight. Currently, most mainstream aircraft navigation methods rely on satellite navigation systems. Satellite navigation has advantages such as high positioning accuracy and wide coverage. However, in complex electromagnetic environments, satellite navigation signals are easily interfered with and shielded. In scenarios where electromagnetic silence is required for mission execution, satellite navigation cannot be used normally. In such cases, when the aircraft relies solely on the inertial navigation system, the position error of the inertial navigation system will accumulate over time, leading to a continuous decline in navigation accuracy and seriously threatening the aircraft's flight safety.

[0003] Airborne terrain-matching navigation is an autonomous navigation method that can serve as a supplementary navigation tool when satellite navigation fails. However, its existing technology has several limitations: First, it often relies solely on camera terrain matching or radar SAR image matching, failing to integrate the two methods, resulting in insufficient matching accuracy and anti-interference capabilities. Second, the construction of pre-stored information databases for terrain and SAR images lacks systematicity, and the integration with information from sensors such as altimeters and atmospheric sensors is low, with the matching process not fully considering the impact of environmental factors. Third, the development and testing of airborne terrain-matching assisted navigation systems largely depend on actual flight tests, which not only results in high development costs and long development cycles but also poses flight safety risks. Existing simulation platforms have limited functionality, only capable of simple visual display, and cannot reproduce the real working characteristics of digital aircraft, inertial navigation systems, and various airborne equipment, nor can they complete the simulation testing of the entire matching navigation process.

[0004] Therefore, developing an airborne scene and terrain matching-assisted navigation method with strong anti-interference capabilities and high navigation accuracy, as well as a highly realistic 3D visual simulation platform that can provide full-process simulation support for this method, has become an urgent problem to be solved in the field of aviation navigation. Summary of the Invention

[0005] The technical problem to be solved by this invention is to overcome the shortcomings of the prior art and provide an airborne scene and terrain matching assisted navigation method and a highly realistic simulation three-dimensional scene simulation platform. This method realizes the fusion matching of camera terrain and radar SAR images, and completes inertial navigation correction by combining information from various sensors, thus solving the navigation problem of satellite navigation failure in complex electromagnetic environments. At the same time, it constructs a highly realistic simulation platform to provide simulation support for the design, testing and optimization of navigation systems, reduce R&D costs and improve R&D efficiency.

[0006] To solve the above-mentioned technical problems, the technical solution provided by the present invention is as follows.

[0007] On the one hand, an airborne scene-terrain matching-assisted navigation method is provided, including the following steps:

[0008] 1. Multi-source measured data acquisition: Real-time measured terrain information during the flight of the aircraft is acquired through an airborne high-definition industrial-grade aerial camera, including terrain outline, ground feature, and texture information; SAR image information, including terrain elevation and landform scattering characteristics, is acquired through an airborne radar; and information from airborne sensors such as altimeters and atmospheric instruments is acquired through a data acquisition interface, including environmental and flight status information such as flight altitude, atmospheric pressure, temperature, wind speed and direction. All acquired measured data is transmitted to the data processing module via wireless or wired communication.

[0009] 2. Construction of Pre-stored Information Database: Based on the flight area planned for the spacecraft mission, a pre-stored information database of terrain and SAR images is constructed in advance. Data sources include satellite remote sensing mapping data, aerial mapping data, and field survey data. The database stores three-dimensional terrain data, standard SAR image data, and corresponding environmental parameter benchmark data within the area. The pre-stored information database supports real-time updates and supplementation of data to adapt to different flight missions and terrain environments.

[0010] 3. Data Fusion and Image Matching: An algorithm for fusing and matching camera terrain and radar images is designed. First, feature extraction is performed on the measured terrain information and SAR image information. Edge features, corner features, and shape features are extracted from the terrain information, and grayscale features, texture features, and elevation features are extracted from the SAR image information. Then, a coarse matching method based on feature points combined with a fine matching method using least squares is adopted to match the measured feature data with the corresponding data in the pre-stored information database. At the same time, environmental factor correction is performed on the matching process by combining sensor information from altimeters and atmospheric instruments, thus completing the fusion of multi-source data.

[0011] 4. Real-time Inertial Navigation Correction and Integrated Navigation: Based on the deviation results of image matching, combined with parameters such as the aircraft's flight speed and attitude, the inertial navigation position correction error is calculated and output, including longitude correction, latitude correction, and elevation correction. This position correction error is fed back to the aircraft's inertial navigation system in real time to complete the real-time position correction of the inertial navigation, realizing integrated navigation of inertial navigation and scene / terrain matching, ensuring the navigation accuracy of the aircraft in complex electromagnetic environments without satellites or in electromagnetic silence.

[0012] On the other hand, a highly realistic 3D visual simulation platform is provided for the design, implementation, and testing of the aforementioned airborne terrain matching assisted navigation method. This platform includes a data simulation module, a digital twin module, a 3D visual module, a matching navigation simulation module, and a test analysis module. These modules work collaboratively to achieve full-process simulation testing of the matching navigation system. Specifically:

[0013] 1. Data Simulation Module: Simulates the working characteristics and data output patterns of various airborne equipment such as airborne cameras, radars, altimeters, and atmospheric instruments, generating simulated measured terrain information, SAR image information, and sensor information. At the same time, it simulates the storage, retrieval, and updating of pre-stored information databases, providing multi-source data support for simulation testing.

[0014] 2. Digital Twin Module: Constructs a 1:1 digital aircraft model that replicates the actual aircraft's shape, structure, and performance parameters, and incorporates actual flight control laws, including core control logic such as attitude control, trajectory control, and power control, accurately reproducing the aircraft's dynamic and motion characteristics; integrates a high-precision inertial navigation simulation model to simulate the real characteristics of inertial navigation such as position drift, error accumulation, and attitude calculation, providing a foundation for inertial navigation correction simulation.

[0015] 3. 3D Visualization Module: Constructs a high-precision 3D terrain model that is 1:1 scale with the actual flight area, restoring details such as terrain features, landforms, and vegetation distribution. Combined with environmental simulation algorithms such as lighting, weather, and visibility, it achieves highly realistic 3D visual display. It supports free switching between multiple perspectives, including airborne first-person view, ground top-down view, and 3D panoramic view, and supports custom adjustment of environmental parameters to meet the needs of different test scenarios.

[0016] 4. Matching Navigation Simulation Module: This module incorporates the aforementioned camera terrain and radar image fusion matching algorithm to simulate the feature extraction, matching calculation, and data fusion correction process between measured data and the pre-stored information database. It calculates the simulated inertial navigation position correction error based on the matching deviation and feeds the error back to the inertial navigation simulation model. This completes real-time inertial navigation correction in the simulation environment, realizing the simulation and restoration of the entire matching navigation process.

[0017] 5. Test and Analysis Module: This module collects, stores, and analyzes all data during the simulation process in real time, including measured simulation data, pre-stored data, matching results, inertial navigation correction errors, and navigation and positioning accuracy. It automatically generates reports on core test indicators such as navigation accuracy, matching success rate, system response speed, and inertial navigation error suppression effect. It supports online adjustment of matching algorithm parameters, optimizes the algorithm through multiple simulation tests, and verifies and improves the hardware selection and software design of the navigation system.

[0018] Furthermore, the simulation platform supports hardware-in-the-loop simulation testing and reserves hardware interfaces to connect to actual airborne cameras, radars, inertial navigation systems, and other hardware devices, enabling joint testing between the simulation environment and actual hardware, and making the simulation test results closer to real-world application scenarios.

[0019] Furthermore, the matching calculation and inertial navigation correction of the airborne landscape and terrain matching assisted navigation method are both real-time closed-loop feedback, and the correction frequency is synchronized with the data acquisition frequency to ensure the timeliness and accuracy of navigation correction. This method is applicable to various fixed-wing aircraft, helicopters, drones and other aircraft, and can achieve stable navigation in complex electromagnetic environments in scenarios such as military reconnaissance, civilian surveying and mapping, and emergency rescue. Beneficial effects

[0020] 1. Improved navigation anti-interference and reliability: The system adopts a fusion matching method that combines camera terrain and radar SAR images with data fusion correction from sensors such as altimeters and atmospheric instruments. Compared with a single matching method, the matching accuracy and anti-interference are significantly improved in complex weather conditions. In complex electromagnetic environments or electromagnetic silence without satellites, the system achieves autonomous combined navigation with inertial navigation and matching, solving the problem of aircraft navigation relying on satellites, effectively suppressing the accumulation of inertial navigation errors, and ensuring navigation accuracy.

[0021] 2. Multi-source data fusion and precise matching: A systematic pre-stored information database of terrain and SAR images is constructed to support real-time data updates. During the matching process, the influence of factors such as flight altitude and atmospheric environment is fully considered. Through the design of coarse matching plus fine matching algorithms, the precise matching of measured data and pre-stored information is achieved, providing a reliable basis for inertial navigation correction.

[0022] 3. Highly Realistic and Full-Process Simulation Platform: The constructed highly realistic 3D visual simulation platform realizes digital twins of digital aircraft, flight control laws, inertial navigation, 3D terrain, and various airborne equipment, restoring the real flight and navigation working characteristics; it supports simulation testing matching the entire navigation process, accurately simulating everything from data acquisition and image matching to inertial navigation correction, and also supports semi-physical simulation, making the test results closer to reality.

[0023] 4. Reduced R&D costs and improved efficiency: The simulation platform provides full-process simulation support for the design, testing and optimization of airborne scene and terrain matching assisted navigation systems, which greatly reduces the number of actual flight tests and effectively reduces R&D costs and flight safety risks; at the same time, it supports online adjustment of algorithm parameters and multi-scenario simulation testing, which greatly improves the R&D and optimization efficiency of navigation systems.

[0024] 5. Strong versatility and scalability: The navigation method of this invention is applicable to various fixed-wing aircraft, helicopters, drones and other aircraft, and can be adapted to different flight missions and terrain environments; the modules of the simulation platform adopt a modular design, which supports the flexible expansion and updating of airborne equipment models, matching algorithms and flight scenarios, and can meet the simulation testing needs of different navigation systems. Attached Figure Description

[0025] Figure 1 This is a flowchart of an airborne scene and terrain matching assisted navigation method according to the present invention:

[0026] 1. Airborne cameras collect measured terrain information, radar collects SAR image information, and information from sensors such as altimeters and atmospheric instruments is collected.

[0027] 2. Construct a database of pre-stored information on terrain and SAR images.

[0028] 3. Camera terrain and radar image matching algorithm, feature extraction + data fusion correction + matching operation.

[0029] 4. Calculate and output the inertial navigation position correction error.

[0030] 5. Feedback is sent to the inertial navigation system to complete inertial navigation correction and achieve inertial navigation + matching combined navigation.

[0031] Figure 2 This is a system block diagram of a highly realistic 3D visual simulation platform according to the present invention:

[0032] Data simulation module: simulates airborne equipment data and pre-stored information database data.

[0033] Digital twin module: digital aircraft model, flight control law, and inertial navigation simulation model.

[0034] 3D visual module: high-precision 3D terrain model, environment simulation, multi-view display.

[0035] Matching navigation simulation module: image matching algorithm, inertial navigation correction simulation, integrated navigation simulation.

[0036] Test and analysis module: data acquisition and storage, indicator analysis, report generation, and algorithm optimization. Detailed Implementation

[0037] The present invention will be further described in detail below with reference to specific embodiments. This embodiment takes the navigation application and simulation test of a fixed-wing aircraft in a complex mountainous environment at an altitude of 3,000 meters and in an electromagnetically silent state as an example to describe in detail the airborne scene and terrain matching assisted navigation method and the highly realistic simulation three-dimensional visual simulation platform of the present invention.

[0038] Example 1: Practical application of airborne scene and terrain matching assisted navigation method.

[0039] 1. Multi-source measured data acquisition: A high-definition industrial-grade aerial camera (4K resolution, 30 frames / second acquisition frequency) was installed at the nose of the fixed-wing aircraft to collect real-time measured terrain information such as the outline of the mountainous terrain, ridges, valleys, and ground features; an airborne synthetic aperture radar was used to collect SAR image information of the mountainous area to obtain terrain elevation and geomorphic scattering characteristics; at the same time, sensor information from the fixed-wing aircraft's altimeter and atmospheric engine was collected. The average data for 5 consecutive minutes during this flight were: flight altitude 3100 meters, atmospheric pressure 68 kPa, ambient temperature 10℃, wind speed 3 m / s, and wind direction northwest; all measured data were transmitted to the fixed-wing aircraft's data processing module via the airborne CAN bus.

[0040] 2. Construction of Pre-stored Information Database: Based on the satellite remote sensing and aerial mapping data of the mountainous flight area, a pre-stored information database of terrain and SAR images is constructed to store the 1:5000 three-dimensional terrain data, standard SAR image data, and atmospheric parameter benchmark data at the altitude of the area. The pre-stored information database is imported into the fixed-wing aircraft data processing module in advance, and the on-site data is supplemented and updated.

[0041] 3. Data Fusion and Image Matching: Corner and edge features are extracted from measured terrain information using matching algorithms, and elevation and grayscale features are extracted from SAR image information. First, the SIFT feature point algorithm is used to complete coarse matching, and then the least squares method is used to complete fine matching. Combined with the flight altitude from the altimeter and the wind speed and direction information from the atmospheric turbine, environmental correction is performed on the matching results to eliminate the influence of terrain occlusion and atmospheric refraction on the matching, and finally the matching deviation between the measured data and the pre-stored information is obtained.

[0042] 4. Real-time Inertial Navigation Correction and Integrated Navigation: Based on the matching deviation and parameters such as the fixed-wing aircraft's flight speed (150 km / h) and attitude angles, the inertial navigation position correction error is calculated: longitude correction +0.0002°, latitude correction -0.00015°, and elevation correction +5 meters. This error is fed back to the fixed-wing aircraft's inertial navigation system in real time to complete the real-time correction of the inertial navigation position. During this 2-hour flight, the accumulated inertial navigation position error reached 80 meters before correction. After matching navigation correction, the navigation and positioning error was controlled within 10 meters, achieving precise navigation in an electromagnetically silent state.

[0043] Example 2: Simulation test of a highly realistic 3D visual simulation platform.

[0044] Taking the fixed-wing aircraft mountain navigation in Example 1 as a simulation scenario, the simulation platform of this invention is used to test and optimize the matching navigation system. The steps are as follows:

[0045] 1. Simulation Scene and Data Configuration: In the data simulation module, the working state of the airborne camera, radar, altimeter, and atmospheric machine is simulated to generate simulated measured terrain information, SAR image information, and sensor information consistent with Example 1; in the three-dimensional visual scene module, a 1:1 three-dimensional terrain model of the mountainous area is constructed, and the environmental parameters are adjusted to: altitude 3000 meters, visibility 10km, wind speed 3m / s, and the airborne first-person perspective is selected for visual display.

[0046] 2. Digital Twin Model Construction: In the digital twin module, a digital twin model of this type of fixed-wing aircraft is constructed, the actual flight control law of the fixed-wing aircraft is implanted, and an inertial navigation simulation model is integrated. The initial drift error of the inertial navigation is set to 0.01° / h to simulate the error accumulation characteristics of the inertial navigation.

[0047] 3. Matching Navigation Simulation: The fusion matching algorithm of this invention is embedded in the matching navigation simulation module to simulate the feature extraction, matching operation and environmental correction of measured simulation data and pre-stored information database. The inertial navigation position correction error is calculated in accordance with that in Example 1 and fed back to the inertial navigation simulation model to complete the simulation of real-time inertial navigation correction.

[0048] 4. Test Analysis and Algorithm Optimization: The test analysis module collects simulation data in real time and generates a test report: The image matching success rate in this simulation is 98.5%, the navigation positioning accuracy is ±9 meters, and the system response speed is 200ms. To address the issue of slightly lower matching accuracy for partially occluded terrain during the matching process, the parameters of the feature extraction algorithm were adjusted online, and the weight of texture features was increased. After another simulation test, the matching accuracy for occluded terrain improved by 15%, and the overall matching success rate reached 99.2%, thus achieving optimization of the matching algorithm.

[0049] 5. Hardware-in-the-loop simulation test: The actual fixed-wing aircraft inertial navigation hardware was connected to the simulation platform. The simulation platform and the inertial navigation hardware were linked through the hardware interface. The simulation test was conducted again. The test results showed that the correction response of the inertial navigation hardware was consistent with the simulation model, the navigation accuracy was ±11 meters, and the deviation from the actual flight test results was less than 2 meters, which verified the actual application effect of the navigation system.

[0050] This embodiment demonstrates that the airborne scene and terrain matching assisted navigation method of the present invention can achieve high-precision autonomous navigation in complex electromagnetic environments / electromagnetic silence. The simulation platform can accurately reproduce the entire navigation process and effectively complete algorithm optimization and hardware testing, providing reliable support for the research and development and application of navigation systems.

Claims

1. An airborne scene-terrain matching assisted navigation method, characterized in that, Includes the following steps: S1. Real-time acquisition of measured terrain information during the flight of the aircraft via airborne camera, acquisition of SAR image information via airborne radar, and acquisition of environmental and flight status information from airborne sensors such as altimeter and atmospheric sensor, and transmission of all acquired measured data to the data processing module. S2. Construct a pre-stored information database of terrain and SAR images to store three-dimensional terrain data, standard SAR image data and corresponding environmental parameter benchmark data of the aircraft's flight area; S3. Design a camera terrain and radar image matching algorithm to perform feature extraction and matching operations on the measured terrain information, SAR image information and corresponding data in the pre-stored information database, and complete data fusion correction by combining the sensor information of altimeter and atmospheric machine. S4. Based on the deviation results of image matching, calculate and output the position correction error of inertial navigation, feed this error back to the aircraft's inertial navigation system, complete the real-time position correction of inertial navigation, and realize the combined navigation of inertial navigation and matching.

2. The airborne scene-terrain matching assisted navigation method according to claim 1, characterized in that: In step S1, the airborne camera is a high-definition industrial-grade aerial camera, and the measured terrain information collected includes terrain outline, ground feature features, and texture information; the SAR image information includes terrain elevation and landform scattering features. The altimeter and atmospheric sensor information include flight altitude, atmospheric pressure, temperature, wind speed, and wind direction.

3. The airborne scene-terrain matching assisted navigation method according to claim 1, characterized in that: In step S2, the pre-stored information database is constructed in advance according to the flight area planned for the spacecraft mission, and supports real-time updates and supplementation of data. The data sources of the pre-stored information database include satellite remote sensing mapping data, aerial surveying data, and field survey data.

4. The airborne scene-terrain matching assisted navigation method according to claim 1, characterized in that: In step S3, the feature extraction stage of the image matching algorithm extracts edge features, corner features, and shape features from the measured terrain information, and extracts grayscale features, texture features, and elevation features from the SAR image information; the matching operation stage adopts a coarse matching based on feature points combined with a fine matching method using the least squares method to improve the matching accuracy.

5. The airborne scene-terrain matching assisted navigation method according to claim 1, characterized in that: In step S4, the position correction error of the inertial navigation includes longitude correction value, latitude correction value, and elevation correction value. The correction process is a real-time closed-loop feedback, and the correction frequency is synchronized with the data acquisition frequency.

6. An airborne scene-terrain matching assisted navigation method according to any one of claims 1-5, characterized in that: The method is applicable to scenarios where satellite navigation fails, such as complex electromagnetic environments without satellites or electromagnetic silence, enabling autonomous navigation of aircraft.

7. A highly realistic 3D visual simulation platform, used for designing, implementing, and testing the airborne scene-terrain matching assisted navigation method according to any one of claims 1-6, characterized in that, include: Data simulation module: Simulates the working status of various airborne equipment such as airborne cameras, radar, altimeters, and atmospheric instruments, generates simulated measured terrain information, SAR image information and sensor information, and simulates the retrieval of data from a pre-stored information database; Digital twin module: Constructs a digital aircraft model, implants actual flight control laws, and restores the aircraft's dynamic and motion characteristics; integrates a high-precision inertial navigation simulation model to simulate the inertial navigation system's position drift, error accumulation, and other characteristics. 3D Visualization Module: Constructs a 1:1 3D terrain model of the actual flight area, restoring the terrain features and landforms, and combining it with environmental simulations such as lighting and weather to achieve highly realistic 3D visualization. Matching navigation simulation module: The camera terrain and radar image matching algorithm described in claim 1 is embedded to simulate the matching process between measured data and pre-stored information, output the simulated inertial navigation position correction error, and complete the inertial navigation correction in the simulation environment; Test and Analysis Module: Collects, stores, and analyzes all data during the simulation process in real time, generates test indicator reports such as navigation accuracy, matching success rate, and system response speed, and supports optimization and adjustment of the matching algorithm and navigation system.

8. The highly realistic 3D visual simulation platform according to claim 7, characterized in that: The digital aircraft model in the digital twin module is consistent with the shape, structure, and performance parameters of the actual aircraft, and the flight control logic includes core control logic such as attitude control, trajectory control, and power control.

9. A highly realistic simulation 3D visual simulation platform according to claim 7, characterized in that: The three-dimensional visual module supports multiple perspective switching, including airborne first-person view, ground top-down view, and three-dimensional panoramic view, and supports custom adjustment of environmental parameters, including light intensity, visibility, precipitation, wind speed, etc.

10. A highly realistic simulation 3D visual simulation platform according to claim 7, characterized in that: The simulation platform supports hardware-in-the-loop simulation testing and can be connected to actual airborne cameras, radars, inertial navigation systems, and other hardware devices to achieve joint testing between the simulation environment and the actual hardware.

Images