Aircraft accident investigation method, system and controller

Abstract

The present invention relates to the field of accident investigation technology, and discloses an aircraft accident investigation method, system and controller, wherein the aircraft accident investigation method comprises investigating the accident site, and using ORB‑SLAM2 technology to model the accident site in real time, generating a real-life three-dimensional model of the accident site; importing weather data and geographic environment data into the real-life three-dimensional model of the accident site; constructing an aircraft accident process evolution model; superimposing the aircraft accident process evolution model on the real-life three-dimensional model to generate a virtual-real accident analysis scene; obtaining a shared analysis result record of the collaborative deduction and analysis of the accident analysis scene by the investigation experts and the on-site investigators; and generating an investigation report based on the shared analysis result record. The present invention relies on mixed reality glasses to perform real-life three-dimensional modeling and on-site data collection, combines the accident process evolution model, and superimposes a virtual analysis model on the real-life three-dimensional model for collaborative investigation, thereby improving the efficiency and accuracy of aircraft accident investigations.

Description

Technical Field

[0001] The present invention relates to the technical field of accident investigation, and in particular to an aircraft accident investigation method and a system controller. Background Art

[0002] With the increase in aircraft accidents, the importance of accident investigation has become increasingly prominent. In traditional aircraft accident investigations, on-site surveys and the recording and preservation of aircraft wreckage information mainly rely on manual sketching, taking photos, and video recording. This method is not only inefficient, but also difficult to provide a comprehensive and real environment for rear experts to analyze. In the prior art, there are some improved accident investigation methods and systems, but these methods are based on on-site images for investigation, which makes it difficult to provide remote personnel with an immersive experience; there are also some methods that use three-dimensional scanners or photogrammetry technology to investigate the accident site, but they are too dependent on hardware and require a large amount of professional equipment to be carried and deployed, which is not suitable for complex environments such as aircraft accident sites. In summary, the existing technical solutions are still insufficient in terms of efficiency and authenticity, and a more efficient and real accident investigation method and system are needed to improve the efficiency and accuracy of aircraft accident investigations. Summary of the invention

[0003] The purpose of the present invention is to overcome the technical problems of low efficiency and accuracy of aircraft accident investigation in the prior art, and to provide an aircraft accident investigation method, system and controller.

[0004] The purpose of the present invention is achieved through the following technical solutions:

[0005] An aircraft accident investigation method, including:

[0006] Investigate the accident scene, use ORB-SLAM2 technology to model the accident scene in real time, and generate a real-life 3D model of the accident scene;

[0007] Use mixed reality glasses to collect preliminary evidence and aircraft wreckage distribution data at the accident scene, and mark the distribution location and status of the aircraft wreckage into the scene of the real-life 3D model of the accident scene;

[0008] Obtain weather data and geographic environment data at the time of the accident, and import the weather data and geographic environment data into the real-life 3D model of the accident scene;

[0009] Conduct a preliminary analysis of the aircraft wreckage distribution data, determine the aircraft's key data, obtain the aircraft's flight data and cockpit voice recording data during normal flight, and build an aircraft accident process evolution model based on the aircraft's key data and aircraft wreckage distribution data;

[0010] Overlay the aircraft accident process evolution model on the real-life 3D model to generate a virtual-real accident analysis scenario;

[0011] Obtain records of shared analysis results from accident analysis scenario rehearsals and collaborative analysis by investigation experts and on-site investigators;

[0012] Generate investigative reports based on shared analysis results records.

[0013] An aircraft accident investigation system based on mixed reality glasses, comprising:

[0014] The generation module is used to investigate the accident scene and use ORB-SLAM2 technology to model the accident scene in real time and generate a real-life 3D model of the accident scene;

[0015] The identification module is used to collect preliminary evidence and aircraft wreckage distribution data at the accident scene using mixed reality glasses, and identify the distribution location and status of the aircraft wreckage into the scene of the real-life 3D model of the accident scene;

[0016] An import module is used to obtain weather data and geographic environment data when the accident occurs, and import the weather data and geographic environment data into a real-life 3D model of the accident scene;

[0017] A construction module is used to conduct a preliminary analysis of the aircraft wreckage distribution data, determine the aircraft key data, obtain the flight data and cockpit voice recording data of the aircraft during normal flight, and construct an aircraft accident process evolution model by combining the aircraft key data and the aircraft wreckage distribution data;

[0018] Overlay module: used to overlay the aircraft accident process evolution model on the real-scene 3D model to generate a virtual-real accident analysis scene;

[0019] The collaborative analysis module is used to obtain the shared analysis results records of the collaborative analysis of accident analysis scenarios by investigation experts and on-site investigators;

[0020] The investigation report generation module is used to generate an investigation report based on the shared analysis result records.

[0021] A controller comprises a memory, a processor and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the above-mentioned aircraft accident investigation method when executing the computer-readable instructions.

[0022] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0023] (1) The present invention relies on mixed reality glasses to perform real-scene three-dimensional modeling and on-site data collection, combines the data collected on-site and the data obtained by the ADS-B system to generate a virtual accident process evolution model, and then superimposes the virtual analysis model on the real-scene three-dimensional model for collaborative investigation. This not only improves the efficiency and accuracy of aircraft accident investigations, but also brings new working methods and technical applications to aircraft accident investigations.

[0024] (2) The present invention uses mixed reality technology to superimpose physical models and mathematical analysis models onto the on-site environment for analysis and deduction, helping experts to more intuitively understand the accident process, find the cause of the accident as quickly as possible, and help remote experts to more accurately understand the situation at the accident site, thereby improving the accuracy of accident analysis; remote collaborative operations and investigation guidance based on mixed reality scenarios facilitate communication and collaboration between on-site investigators and experts; investigation reports and original records, evidence, material attribute data, etc. can be directly attached to aircraft wreckage components to directly present various detailed information, facilitating information analysis; the information required for accident investigation can be directly exported from the analysis scene to quickly generate an investigation report; during on-site investigations, only mixed reality glasses are needed to achieve on-site real-life three-dimensional modeling and investigation information collection, without the need for additional expensive and bulky equipment; relying on mixed reality glasses, quick operations can also be performed through a combination of gaze, voice, and gestures, without the need to unfold a computer, which is convenient and fast in complex environments; in addition to mixed reality glasses, there is no need to carry and unfold a large number of professional equipment, simplifying on-site investigation work.

[0025] (3) The present invention overcomes the shortcoming of traditional oblique photography technology that requires collecting full-scene data before starting modeling. It can realize modeling while collecting data, with high modeling efficiency and low requirements on hardware performance. The lens of the mixed reality glasses can meet the data collection and calculation requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A flowchart of an aircraft accident investigation method according to an embodiment of the present invention;

[0027] Figure 2 is a structural diagram of an aircraft accident investigation system according to an embodiment of the present invention;

[0028] Figure 3 is a schematic structural diagram of a controller according to an embodiment of the present invention;

[0029] Figure 4 A three-dimensional real scene diagram of the distribution of aircraft wreckage in a three-dimensional real scene model of an embodiment of the present invention;

[0030] Figure 5 This is a three-dimensional real-life picture of the distribution of aircraft wreckage in a three-dimensional on-site real-life model according to another embodiment of the present invention. DETAILED DESCRIPTION

[0031] The present invention is further described in detail below in conjunction with embodiments:

[0032] In one embodiment, if Figure 1 As shown, an aircraft accident investigation method comprises the following steps:

[0033] S10. Investigate the accident scene, use ORB-SLAM2 technology to model the accident scene in real time, and generate a real-life 3D model of the accident scene. The camera of the mixed reality glasses and ORB-SLAM2 technology can be used to perform 3D modeling of the scene to generate a more realistic and intuitive real-life 3D model of the scene.

[0034] Specifically, ORB-SLAM2 technology is an advanced visual SLAM technology that uses ORB (Oriented FAST and Rotated BRIEF) feature points in image sequences to quickly extract and match features, thereby achieving 3D modeling and positioning of the environment. The steps of using ORB-SLAM2 technology to perform 3D modeling of the scene in combination with the camera of mixed reality glasses usually include:

[0035] Initialize ORB-SLAM2: select camera internal parameters and set feature extraction parameters;

[0036] On-site data collection: Wearing mixed reality glasses, moving around the accident scene, and capturing image data in real time through the camera;

[0037] Feature extraction and matching: ORB-SLAM2 extracts ORB feature points in the captured image and tracks the movement of the camera by matching feature points between consecutive frames;

[0038] Map construction: Using the matched feature points, ORB-SKAM2 calculates the position and posture of the camera to achieve positioning; the system combines the positioning information with the feature points to gradually build a 3D map of the accident scene;

[0039] 3D model generation: As the camera moves, ORB-SLAM2 continuously updates the 3D map and generates a continuous 3D model of the real scene of the accident. The construction of the 3D map can be viewed in real time through the software interface.

[0040] Model export: Import the exported 3D model into a mixed reality environment and combine it with virtual elements to achieve augmented reality effects.

[0041] Furthermore, ORB-SLAM3 can be used to obtain better modeling effects.

[0042] S20. Use mixed reality glasses to collect preliminary evidence and aircraft wreckage distribution data at the accident scene, and mark the location and status of the aircraft wreckage distribution in the scene of the real-life 3D model of the accident scene; investigators arrive at the accident scene in time and wear mixed reality glasses to collect preliminary evidence and aircraft wreckage distribution data at the accident scene to ensure the integrity and freshness of the evidence. Preliminary evidence mainly includes eyewitness testimony, on-site photos and on-site videos; aircraft wreckage distribution data mainly includes the location of the aircraft wreckage, the state after the fall, the location of important aircraft components, and the distribution of wreckage.

[0043] Specifically, investigators can collect and verify first-hand information from eyewitness testimony, including the time, sound, smoke, fire, and other information that are very important for reconstructing the accident scene. By wearing mixed reality glasses, they can collect photos and videos of the scene to obtain the original state of the scene, which can provide visual reference for subsequent analysis. Investigators record and mark the specific location and status of the aircraft wreckage at the scene, paying attention to the location and relative distribution of important components, such as the physical damage status of the engine cabin facilities, the functional failure status, the functional failure status, the wreckage distribution, the safety status information, the equipment oil leakage and other data; they can also capture the distribution of the wreckage, the impact point, the degree of damage and other key details through image data. It is also necessary to check the degree of damage to the aircraft engine and cabin and the equipment oil leakage to ensure that all collected evidence is properly kept to prevent the loss or damage of evidence. Based on the preliminary evidence collected, a rapid assessment of the scene is conducted to determine the severity of the accident and the possible scope of impact. Based on the information collected, investigators form preliminary hypotheses about the cause of the accident to provide direction for subsequent in-depth investigations.

[0044] Preferably, based on preliminary evidence and aircraft wreckage distribution data, laser scanners and mixed reality glasses can be used separately or in combination. Laser scanners can accurately capture three-dimensional data of the accident scene. Laser scanners determine the position of the surface of an object by emitting a laser beam and measuring the time it takes for the beam to be reflected back, thereby creating high-precision point cloud data. The point cloud data can reflect every detail of the scene in great detail, including the size, shape, and position of the wreckage. The photography technology of mixed reality glasses can be used to take multi-angle photos of the accident scene to capture the color and texture information of the scene.

[0045] S30. Obtain weather data and geographic environment data at the time of the accident, and import the weather data and geographic environment data into the real-life three-dimensional model of the accident scene; the weather data includes temperature, wind speed, wind direction, precipitation, and visibility, etc., the geographic environment data includes terrain features, vegetation features, and buildings, and the aircraft key data includes the impact point and fall trajectory of the aircraft, etc.

[0046] The ways to obtain weather data include: using a meteorological database; collecting descriptions of weather conditions from eyewitnesses at the accident site, such as wind speed, visibility, rainfall, etc.; and obtaining real-time weather data from a meteorological station.

[0047] The ways to obtain geographic environment data include: using GIS to obtain geographical environment information such as terrain, landforms, vegetation and buildings at the accident site; obtaining a topographic map of the accident site to understand information such as terrain undulations and altitude; using high-resolution satellite images to analyze the geographical features of the accident site; and collecting soil, vegetation, hydrology and other geographical environment data through on-site surveys.

[0048] The collected weather data and geographic environment data are formatted to meet the requirements of importing the real-life 3D model of the accident site. A special data layer is created in the real-life 3D model of the accident site to store and display weather data and geographic environment data, and the weather data and geographic environment data are mapped to the corresponding positions of the 3D model. The weather data and geographic environment data are integrated with the real-life 3D model to create a comprehensive accident analysis scenario. The 3D model that integrates weather data and geographic environment data is used to perform accident analysis and simulation, and to evaluate the potential impact of various factors on the accident. During the accident investigation, the weather data and geographic environment data are continuously updated to ensure the timeliness and accuracy of the 3D model. Integrating weather and geographic environment data into the real-life 3D model of the accident site can provide a more comprehensive and in-depth analysis for the accident investigation.

[0049] S40. Preliminary analysis of aircraft wreckage distribution data, and determination of aircraft key data, acquisition of aircraft flight data and cockpit voice recording data during normal flight, and construction of an aircraft accident process evolution model in combination with aircraft key data and aircraft wreckage distribution data;

[0050] Analyze the distribution data of aircraft wreckage (the specific location and status of aircraft wreckage, pay attention to the location and relative distribution of important components), use statistical and computational tools to analyze the wreckage distribution pattern, infer the dynamic situation when the accident occurred, and determine the key data of the aircraft (the impact point and fall trajectory of the aircraft) through analysis. The flight data of the aircraft can be extracted from the black box of the aircraft, and the maintenance and flight logs of the aircraft can be consulted to obtain historical flight data (speed, heading, altitude before loss of contact, speed, attitude, fuel volume and flight trajectory parameters). Use professional flight simulation software, input flight data, simulate the flight path and status of the aircraft, combine with the wreckage distribution data, adjust the simulation parameters, reproduce the process of the accident, and analyze the possibility of accidents under different conditions by changing the parameters in the simulation (such as speed, angle, external environment, etc.). Ask experts in flight operations, aviation engineering and accident investigation to review and verify the simulation results, so as to generate an aircraft accident process evolution model from the normal operation of the aircraft, the fall in the air, the fall location, the ground trajectory, the state after the fall, and the distribution of important aircraft components. The aircraft accident process evolution model is constructed, including but not limited to the above process.

[0051] S50, superimposing the aircraft accident process evolution model on the real-scene three-dimensional model to generate a virtual-real accident analysis scenario.

[0052] S60. Obtain the shared analysis result records of the accident analysis scenario preview and collaborative analysis by the investigation experts and on-site investigators; if remote technical support is required, the on-site investigators can share the accident analysis scenario with the rear investigation experts through the mobile private network. The experts can remotely enter the accident analysis scenario by wearing mixed reality glasses, move freely in the accident analysis scenario, observe the virtual and real accident scene from any angle, view the various on-site record information anchored on the aircraft, and freely place models of aircraft, passengers, injured persons, wreckage, etc. on the real-life 3D model for deduction.

[0053] S70. Generate an investigation report based on the shared analysis result records.

[0054] In one embodiment, the step S40 includes the following steps:

[0055] S401. Use ADS-B technology to obtain flight data of aircraft during normal flight; Automatic Dependent Surveillance-Broadcast (ADS-B) technology can provide real-time position, speed, altitude, heading and other data of aircraft. Obtain flight data before and after the accident through ADS-B ground station or satellite system. Parse ADS-B data into a usable format for analysis and simulation. Verify the acquired ADS-B data to ensure the accuracy and completeness of the data. Integrate ADS-B data with other flight data sources (such as flight recorder data) to form a comprehensive flight data set.

[0056] S402, analyzing the distribution data of the aircraft wreckage, preliminarily determining the key data of the aircraft, and determining the impact point and ground falling trajectory data of the aircraft collected from the accident site data;

[0057] The impact point and fall trajectory of the aircraft are rehearsed. Through on-site investigation and measurement, the location, size, type and distribution of the wreckage are collected, and key information is extracted from the wreckage distribution data, such as the possible location of the impact point, the scattering pattern of the wreckage, the fall trajectory, etc. The collected data and professional knowledge are used to preliminarily analyze the impact point and ground fall trajectory data of the aircraft.

[0058] S403. Analyze flight data, and use accident analysis software in combination with weather data, geographic environment data, and aircraft key data to analyze the accident process, construct a differential model of the fall trajectory and a probability density function model of the wreckage distribution; integrate weather data into geographic environment data, and use professional accident analysis software, such as FlightRecon, to simulate and analyze the accident process.

[0059] Based on the principles of physics and engineering, a differential model of the aircraft's falling trajectory is constructed to simulate the aircraft's movement and falling process in the air. Specifically, the following processes are included:

[0060] Select an appropriate coordinate system and define the aircraft's position, velocity, and acceleration;

[0061] Taking flight data into consideration and combining weather data and geographic environment data, a differential model of the aircraft fall trajectory is established. The differential model of the aircraft fall trajectory is specifically expressed as follows:

[0062]

[0063]

[0064] Among them, F d It represents the resistance of the falling aircraft, ρ represents the air density, v represents the speed of the aircraft, t represents the time, k represents the drag coefficient, S represents the frontal area, M represents the mass of the aircraft, and g represents the acceleration due to gravity.

[0065] Using statistical methods, a probability density function model of the debris distribution is constructed to predict the possible distribution range of the debris, which specifically includes the following processes:

[0066] Analyze the characteristics of wreckage distribution based on the collected aircraft wreckage distribution data and select an appropriate probability distribution model, such as normal distribution, Weibull distribution, etc.;

[0067] Estimate parameters of probability distributions using statistical methods (e.g., maximum likelihood estimation, moment estimation);

[0068] The probability distribution model is fitted to the debris distribution data, a probability density function model of the debris distribution is established and the goodness of fit of the model is evaluated;

[0069] Verify the accuracy of the model using field data and make adjustments if necessary;

[0070] Analyze the sensitivity of the model to key parameters (e.g. wind speed, impact angle);

[0071] Use the probability density function model to predict the possible distribution range of the wreckage.

[0072] The expression of the probability density function model of the debris distribution is:

[0073]

[0074] Where x is the location of the debris, y is the mean of the debris distribution, and σ is the standard deviation.

[0075] like Figure 4 As shown, based on the aircraft wreckage distribution data and using the probability density function model of wreckage distribution, a three-dimensional map of the aircraft wreckage distribution range can be predicted.

[0076] S404. Estimate the aircraft's fall trajectory based on the fall trajectory differential model, estimate the aircraft wreckage distribution range based on the wreckage distribution probability density function model, and generate an aircraft accident process evolution model based on the fall trajectory differential model and the wreckage distribution probability density function model. Use the fall trajectory differential model to estimate the aircraft's detailed trajectory from loss of control to impact with the ground, apply the wreckage distribution probability density function model to estimate the distribution range and density of the wreckage on the ground, and combine the fall trajectory and wreckage distribution model to generate an aircraft accident process evolution model. This model can show the entire process from the occurrence of the accident to the distribution of the wreckage, and verify the accuracy and reliability of the evolution model by comparing it with on-site evidence and other data sources. Through these detailed steps, the accident investigation team can analyze aircraft accidents more comprehensively and deeply, build a scientific and accurate accident process evolution model, and provide important information for determining the cause of the accident and formulating preventive measures.

[0077] In one embodiment, the step S50 includes the following steps:

[0078] S501. Perform coordinate system 1 on the local coordinates of the real-scene 3D model and the GPS coordinates of the aircraft accident process evolution model. Perform spatial correction by collecting the GPS coordinates and local coordinates of at least 4 points, project the real-scene 3D model to the WGS84 coordinate system, and complete the superposition of the aircraft accident process evolution model and the real-scene 3D model. Specifically, the coordinate system of the real-scene 3D model generated by the mixed reality glasses is the local coordinate system, and the aircraft flight data obtained by ADS-B is the WGS84 coordinate system. Therefore, the coordinate systems need to be unified before data superposition can be performed.

[0079] By using the GPS of the mixed reality glasses or a device with GPS positioning, such as a mobile phone, to collect the collision points, ground track points, wreckage distribution points and other points that have been marked on the real-life 3D model, and simultaneously obtaining the GPS coordinates (WGS84 coordinate system) of these points (no less than 4) and their coordinates in the local coordinate system of the mixed reality glasses, spatial correction can be performed and the real-life 3D model can be projected to the WGS84 coordinate system, thereby realizing the superposition of the aircraft accident process evolution model and the real-life 3D model of the accident scene.

[0080] S502. Use a fusion software platform to fuse the aircraft accident process evolution model and the virtual analysis model, and combine the dynamic data in the aircraft accident process evolution model and the static data in the virtual analysis model to generate an accident analysis scenario of the complete process of the accident; select a fusion software platform to import the aircraft key data determined after preliminary analysis of the aircraft wreckage distribution data into the real-life three-dimensional model of the accident site to generate a virtual analysis model; use a fusion software platform to integrate the collected aircraft key data into the real-life three-dimensional model of the accident site to simulate the flight path, speed, attitude changes, etc. of the aircraft. The fusion software platform may be, but is not limited to, one of Blender, Unity3D, Unreal Engine, LIDAR data processing software, and GIS software.

[0081] Import the aircraft wreckage distribution data and the real-life 3D model of the accident scene into the selected fusion software platform, integrate the dynamic data (such as flight path, speed change, etc.) in the aircraft accident process evolution model into the 3D model, combine the static data (such as wreckage location, terrain features, etc.) in the virtual analysis model with the dynamic data, build an accident analysis scene at the accident scene in the software, ensure that the position and attributes of all elements are consistent with the actual situation, create animations to simulate the complete process from flight to crash of the aircraft, including key events and time points, and use real-time rendering technology to enable investigators to observe and analyze the accident process in real time, increase interactivity, and allow users to explore the accident scene through different views and parameters.

[0082] S503, archiving the generated accident analysis scenario. Backing up all generated data, models and scenarios to a safe location to prevent data loss.

[0083] In one embodiment, the step S60 includes the following steps:

[0084] S601. Investigation experts and on-site investigators share the accident analysis scene. Investigation experts and on-site investigators wear mixed reality glasses to remotely enter the shared accident analysis scene. Investigation experts can remotely connect to the shared accident analysis scene via the Internet and load the accident analysis scene into the mixed reality glasses to ensure that all participants can see the same three-dimensional view. Different access rights are set according to the roles of the participants to ensure data security and process control.

[0085] S602. Investigation experts and on-site investigators conduct a preliminary understanding of the accident site, simulate the real-life 3D model of the accident site using weather data and geographic environment data, rehearse the impact point, fall trajectory, wreckage distribution, and casualties of the aircraft wreckage, use the accident process evolution model for inversion, and collaboratively analyze the process of the accident; integrate weather data and geographic environment data into the 3D model to provide more comprehensive background information, use the accident process evolution model to simulate the impact point and fall trajectory of the aircraft, simulate the distribution of aircraft wreckage, analyze the possible impact force and direction, analyze the situation of the casualties in the accident, evaluate the effectiveness of the rescue response, and investigate the possible causes of the accident by reverse engineering the accident process. Investigation experts and on-site investigators jointly analyze and propose hypotheses and insights.

[0086] S603. Investigators and field investigators communicate and discuss in real time through voice and text chat tools, share concerns and gather discussion results using the shared perspective function of mixed reality glasses, and summarize key opinions, issues to be investigated and consensus in the discussion results into a shared analysis result record. Use voice and text chat tools to achieve real-time communication and discussion. Using the shared perspective function of mixed reality glasses, all participants can focus on the same area or data point, record key opinions and assumptions raised in the discussion, record issues that require further investigation and unresolved questions, summarize the consensus reached in the discussion as the basis for further investigation, and plan subsequent investigation actions and task assignments based on the shared analysis result record. All discussions and analysis results are properly documented and archived.

[0087] The combination of the above steps not only promotes effective communication and collaboration between investigation experts and on-site investigators, but also improves the accuracy and efficiency of accident analysis. This collaborative deduction and analysis method can make full use of the expertise of all parties and provide a comprehensive perspective and in-depth insights for accident investigation.

[0088] In one embodiment, the step S602 further includes the following steps:

[0089] S6021. Obtain the aircraft flight recorder and surveillance video data around the accident site. After the investigation experts and on-site investigators have a preliminary understanding of the accident site, obtain the track angle and flight speed of the aircraft before it crashed;

[0090] S6022: When the aircraft flight recorder and the surveillance video data around the accident site are missing, analyze the evolution of the accident process, and build an inversion model based on the distribution pattern of the aircraft wreckage and the accident process evolution model to determine the track angle and flight speed of the aircraft before it hit the ground. The expression of the inversion model is:

[0091] M(x, y, z) = M θ (x,y,z)·S(v)·A(θ)·T(h)·F(k,e)·W(t,p,r)

[0092] Among them, M θ (x, y, z) is a function related to the spatial distribution of the wreckage, x, y, z are the coordinates of the wreckage in the three-dimensional space, S(v) represents the influence of the aircraft's speed v on the accident, A(θ) represents the influence of the aircraft's track angle θ when it touches down on the accident, T(h) represents the influence of the terrain h on the aircraft impact and wreckage distribution, F(k, e) represents the influence of the aircraft's structural parameters k and e on the accident, and W(t, p, r) represents the influencing factor of meteorological conditions, where t represents temperature, p represents air pressure, and r represents relative humidity.

[0093] In the above-mentioned embodiments of the present invention, the physical model and the mathematical analysis model are superimposed on the on-site environment through mixed reality technology for analysis and deduction, so as to help experts understand the accident process more intuitively, find the cause of the accident as quickly as possible, and help remote experts understand the situation at the accident site more accurately, thereby improving the accuracy of accident analysis; remote collaborative operation and investigation guidance based on mixed reality scenes facilitate communication and collaboration between on-site investigators and experts; investigation reports and original records, evidence, material attribute data, etc. can be directly attached to the aircraft wreckage parts, intuitively presenting various detailed information, and facilitating information analysis; the information required for the accident investigation can be directly exported from the analysis scene to quickly generate an investigation report; during the on-site investigation, only mixed reality glasses are needed to realize the real-life three-dimensional modeling and investigation information collection of the scene, without the need for additional expensive and bulky equipment; relying on mixed reality glasses, quick operations can also be performed through a combination of gaze, voice, and gestures, without the need to unfold a computer, which is convenient and fast in complex environments; in addition to mixed reality glasses, there is no need to carry and unfold a large number of professional equipment, which simplifies the on-site investigation work.

[0094] In one embodiment, if Figure 2As shown, an aircraft accident investigation system is provided, which corresponds one-to-one to the aircraft accident investigation method in the above embodiment, and the aircraft accident investigation system includes:

[0095] The generation module is used to investigate the accident scene and use ORB-SLAM2 technology to model the accident scene in real time and generate a real-life 3D model of the accident scene;

[0096] The identification module is used to collect preliminary evidence and aircraft wreckage distribution data at the accident scene using mixed reality glasses, and identify the distribution location and status of the aircraft wreckage into the scene of the real-life 3D model of the accident scene;

[0097] An import module is used to obtain weather data and geographic environment data when the accident occurs, and import the weather data and geographic environment data into a real-life 3D model of the accident scene;

[0098] A construction module is used to conduct a preliminary analysis of the aircraft wreckage distribution data, determine the aircraft key data, obtain the flight data and cockpit voice recording data of the aircraft during normal flight, and construct an aircraft accident process evolution model by combining the aircraft key data and the aircraft wreckage distribution data;

[0099] Overlay module: used to overlay the aircraft accident process evolution model on the real-scene 3D model to generate a virtual-real accident analysis scene;

[0100] The collaborative analysis module is used to obtain the shared analysis results records of the collaborative analysis of accident analysis scenarios by investigation experts and on-site investigators;

[0101] The investigation report generation module is used to generate an investigation report based on the shared analysis result records.

[0102] In one embodiment, a controller is provided, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the above-mentioned aircraft accident investigation method when executing the computer-readable instructions. Figure 3 As shown, the controller includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the controller is used to provide computing and control capabilities. The memory includes a readable storage medium and an internal memory. The readable storage medium stores an operating system, computer-readable instructions and a database. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. When the computer-readable instructions are executed by the processor, an aircraft accident investigation method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0103] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An aircraft accident investigation method, characterized in that: include: S10. Investigate the accident scene, use ORB-SLAM2 technology to model the accident scene in real time, and generate a real-life 3D model of the accident scene; S20, using mixed reality glasses to collect preliminary evidence and aircraft wreckage distribution data at the accident scene, and marking the aircraft wreckage distribution location and status into the scene of the real-life three-dimensional model of the accident scene; S30, obtaining weather data and geographic environment data when the accident occurred, and importing the weather data and geographic environment data into the real-scene three-dimensional model of the accident scene; S40 preliminarily analyzes the aircraft wreckage distribution data, determines the aircraft key data, obtains the flight data and cockpit voice recording data of the aircraft during normal flight, and constructs an aircraft accident process evolution model in combination with the aircraft key data and the aircraft wreckage distribution data; step S40 includes the following steps: S401. Use ADS-B technology to obtain flight data of aircraft during normal flight; S402, analyzing the distribution data of the aircraft wreckage, preliminarily determining the key data of the aircraft, and determining the impact point and ground falling trajectory data of the aircraft collected from the accident site data; S403, analyzing the flight data, and using the accident analysis software to analyze the accident process in combination with the weather data, geographic environment data and aircraft key data, and constructing a differential model of the crash trajectory and a probability density function model of the wreckage distribution; S404, estimating the aircraft's fall trajectory according to the fall trajectory differential model, estimating the aircraft's wreckage distribution range according to the wreckage distribution probability density function model, and generating an aircraft accident process evolution model according to the fall trajectory differential model and the wreckage distribution probability density function model; S50, superimposing the aircraft accident process evolution model on the real-scene three-dimensional model to generate a virtual-real accident analysis scene; S60. Obtain the shared analysis results records of the accident analysis scenario rehearsal and collaborative analysis by the investigation experts and on-site investigators; S70. Generate an investigation report based on the shared analysis result records.

2. The aircraft accident investigation method according to claim 1, characterized in that: The preliminary evidence includes eyewitness testimony, on-site photos and on-site videos; the aircraft wreckage distribution data includes the crash location of the aircraft wreckage, its state after the crash, the location of important aircraft components and the distribution of the wreckage.

3. The aircraft accident investigation method according to claim 1, characterized in that: The weather data includes temperature, wind speed, wind direction, precipitation and visibility; the geographical environment data includes terrain features, vegetation features and buildings; and the aircraft key data includes the impact point and fall trajectory of the aircraft.

4. The aircraft accident investigation method according to claim 1, characterized in that: The step S50 comprises the following steps: S501, coordinate system 1 is performed on the local coordinates of the real-scene 3D model and the GPS coordinates of the aircraft accident process evolution model, spatial correction is performed by collecting the GPS coordinates and local coordinates of at least 4 points, and the real-scene 3D model is projected into the WGS84 coordinate system to complete the superposition of the aircraft accident process evolution model and the real-scene 3D model; S502. Use the fusion software platform to fuse the aircraft accident process evolution model and the virtual analysis model, and combine the dynamic data in the aircraft accident process evolution model and the static data in the virtual analysis model to generate an accident analysis scenario of the complete process of the accident; S503: Archive the generated accident analysis scenario.

5. The aircraft accident investigation method according to claim 1, characterized in that: The step S60 comprises the following steps: S601, the investigation expert and the on-site investigator share the accident analysis scene, and the investigation expert and the on-site investigator wear mixed reality glasses to remotely enter the shared accident analysis scene; S602. Investigation experts and on-site investigators conduct a preliminary understanding of the accident site, simulate the weather data and geographic environment data in the real-life 3D model of the accident site, rehearse the impact point, fall trajectory, wreckage distribution and casualties of the aircraft wreckage, use the accident process evolution model for inversion, and collaboratively analyze the process of the accident; S603. Investigation experts and on-site investigators communicate and discuss in real time through voice and text chat tools, use the shared perspective function of mixed reality glasses to share focus, compile discussion results, and summarize key opinions, issues to be investigated, and consensus in the discussion results into a shared analysis result record.

6. The aircraft accident investigation method according to claim 5, characterized in that: The simulation in step S602 includes: S6021. Obtain the aircraft flight recorder and surveillance video data around the accident site. After the investigation experts and on-site investigators have a preliminary understanding of the accident site, obtain the track angle and flight speed of the aircraft before it crashed; S6022: When the aircraft flight recorder and the surveillance video data around the accident site are missing, analyze the evolution of the accident process, and build an inversion model based on the distribution pattern of the aircraft wreckage and the accident process evolution model to determine the track angle and flight speed of the aircraft before it hit the ground. The expression of the inversion model is: M(x,y,z)=M0(x,y,z)·S(v)·A(θ)·T(h)·F(k,e)·W(t,p,r) Among them, M0(x, y, z) is a function related to the spatial distribution of the wreckage, x, y, z are the coordinates of the wreckage in the three-dimensional space, S(v) represents the influence of the aircraft's speed v on the accident, A(θ) represents the influence of the aircraft's track angle θ when it touches down on the accident, T(h) represents the influence of the terrain h on the aircraft impact and wreckage distribution, F(k, e) represents the influence of the aircraft's structural parameters k and e on the accident, and W(t, p, r) represents the influencing factor of meteorological conditions, where t represents temperature, p represents air pressure, and r represents relative humidity.

7. The aircraft accident investigation method according to claim 1, characterized in that: The step S70 specifically includes drawing the shared analysis result records obtained from the collaborative deduction analysis in a three-dimensional scene, and generating an investigation report in combination with the accident scene data.

8. An aircraft accident investigation system, characterized in that: include: The generation module is used to investigate the accident scene and use ORB-SLAM2 technology to model the accident scene in real time and generate a real-life 3D model of the accident scene; The identification module is used to collect preliminary evidence and aircraft wreckage distribution data at the accident scene using mixed reality glasses, and identify the distribution location and status of the aircraft wreckage into the scene of the real-life 3D model of the accident scene; An import module is used to obtain weather data and geographic environment data when the accident occurs, and import the weather data and geographic environment data into a real-life 3D model of the accident scene; A construction module is used to conduct a preliminary analysis of the aircraft wreckage distribution data, determine the aircraft key data, obtain the flight data and cockpit voice recording data of the aircraft during normal flight, and combine the aircraft key data and the aircraft wreckage distribution data to construct an aircraft accident process evolution model; the method includes: S401. Use ADS-B technology to obtain flight data of aircraft during normal flight; S402, analyzing the distribution data of the aircraft wreckage, preliminarily determining the key data of the aircraft, and determining the impact point and ground falling trajectory data of the aircraft collected from the accident site data; S403, analyzing the flight data, and using the accident analysis software to analyze the accident process in combination with the weather data, geographic environment data and aircraft key data, and constructing a differential model of the crash trajectory and a probability density function model of the wreckage distribution; S404, estimating the aircraft's fall trajectory according to the fall trajectory differential model, estimating the aircraft's wreckage distribution range according to the wreckage distribution probability density function model, and generating an aircraft accident process evolution model according to the fall trajectory differential model and the wreckage distribution probability density function model; Overlay module: used to overlay the aircraft accident process evolution model on the real-scene 3D model to generate a virtual-real accident analysis scene; The collaborative analysis module is used to obtain the shared analysis results records of the collaborative analysis of accident analysis scenarios by investigation experts and on-site investigators; The investigation report generation module is used to generate an investigation report based on the shared analysis result records.

9. A controller, characterized in that: The invention comprises a memory, a processor and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the aircraft accident investigation method according to any one of claims 1 to 7 when executing the computer-readable instructions.

Images