An airspace risk modeling method for autonomous operation scenarios

Abstract

The application discloses a kind of airspace risk modeling methods for autonomous operation scene, belong to aviation transport safety field.For meet the requirement of airspace risk assessment refinement and immediacy under future autonomous operation mode, first, establish aircraft ellipsoid risk influence zone to determine risk influence range, to improve model calculation efficiency, airspace is gridded;Second, based on artificial potential field theory, the real-time risk influence of factors such as distance, speed, heading to airspace each grid is considered to construct airspace risk assessment model;Next, set simulation scene, calculate the correlation between conflict index and airspace risk value, explain the rationality of model;Finally, propose airspace hotspot grid marking method, combine multiple grid scales to visualize airspace hotspot distribution.The application is based on artificial potential field theory, proposes a kind of aircraft autonomous operation model under airspace any position point risk fast calculation method, also combines hotspot grid marking method and multiple scale grid size, realizes from spatial geographic angle to airspace safety situation Multi-particle perception, provides reference for future autonomous operation mode under air traffic management and airborne auxiliary system to carry out path adjustment.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of civil aviation transportation safety, and particularly relates to an airspace risk modeling method for autonomous operation scenarios. BACKGROUND

[0002] In order to meet the requirements of accuracy and immediacy of airspace risk assessment under the future autonomous operation mode, and to adapt to the complex aviation environment of high-density and multi-type operation of future flights, an airspace risk assessment method under the autonomous operation mode of aircraft is urgently needed to improve the airspace safety situation awareness capability under the future autonomous operation mode.

[0003] At present, in the research on airspace safety situation awareness, many experts and scholars take the conflict between aircraft as the starting point to measure the airspace safety situation or give the air traffic physical structure based on complex network to realize the global evaluation of airspace safety situation. However, the existing research on airspace operation risk assessment focuses on the traditional air traffic management mode, and few researches focus on the future autonomous operation mode. Under the autonomous operation mode, the aircraft relies on its own perception, decision and control system to complete the flight task, and the real-time risk measurement is the key to ensure the safety and reliability of autonomous operation. The existing research cannot meet the requirements of real-time measurement of airspace risk at any position.

[0004] Therefore, it is necessary to provide an airspace risk modeling method for autonomous operation scenarios to solve the above problems. SUMMARY

[0005] The purpose of the present application is to provide an airspace risk modeling method for autonomous operation scenarios, which comprehensively considers distance, speed, heading and other factors to realize real-time risk assessment of each position in the airspace, and realizes multi-particle perception of airspace safety situation from the spatial geographic angle by combining the airspace hotspot grid marking method.

[0006] To achieve the above purpose, the present application provides an airspace risk modeling method for autonomous operation scenarios, comprising the following steps:

[0007] Step 1: Establishing an aircraft ellipsoid risk influence area to determine the risk influence range, and gridding the airspace;

[0008] Step 2: Based on the artificial potential field theory, comprehensively considering the real-time risk influence of distance, speed, heading and other factors on each grid of the airspace, constructing an airspace risk assessment model;

[0009] Step 3: Setting a simulation scenario, calculating the correlation between the conflict index and the airspace risk value, and verifying the rationality of the model;

[0010] Step 4: Proposing an airspace hotspot grid marking method, and visualizing the distribution of airspace hotspots in combination with multiple grid scales.

[0011] Step 1 specifically comprises:

[0012] Step 1.1: Establishing the aircraft ellipsoid risk influence area;

[0013] Step 1.2: Airspace gridding.

[0014] Step 2 specifically comprises:

[0015] Step 2.1: Based on the artificial potential field theory, a general expression of the real-time risk influence of distance, speed, heading and other factors on each position point in the airspace is constructed;

[0016] Step 2.2: Correcting the distance risk function;

[0017] Step 2.3: Combining distance, speed, heading angle and other risk influencing factors, a risk calculation formula for specific positions in the airspace is proposed.

[0018] Step 3 specifically comprises:

[0019] Step 3.1: Setting up an autonomous running simulation scene;

[0020] Step 3.2: Calculating the correlation between the conflict index and the airspace risk value.

[0021] Step 4 specifically comprises:

[0022] Step 4.1: Airspace hotspot grid marking;

[0023] Step 4.2: Combining multiple grid scales to visualize the distribution of airspace hotspots.

[0024] The beneficial effects of the present application are that the present application proposes a rapid risk calculation method for any position point in the airspace under the autonomous running model of the aircraft based on the artificial potential field theory, and combines the hotspot grid marking method and multiple scale grid sizes to realize multi-particle perception of the airspace safety situation from the spatial geographic angle, and provides a reference for future air traffic management and airborne auxiliary system flight path adjustment under the autonomous running mode. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 : A flowchart of an airspace risk modeling method for an autonomous running scenario.

[0026] Figure 2 : An aircraft ellipsoid risk area schematic diagram of an airspace risk modeling method for an autonomous running scenario.

[0027] Figure 3 : An airspace grid schematic diagram of an airspace risk modeling method for an autonomous running scenario.

[0028] Figure 4 : A schematic diagram of the risk influence of an aircraft on the surrounding airspace in a space risk modeling method for autonomous operation scenarios, wherein the upper left is the XY cross-section risk influence, the upper right is the XZ cross-section risk influence, and the lower middle is the three-dimensional risk influence.

[0029] Figure 5 : A schematic diagram of airspace hotspot distribution in a space risk modeling method for autonomous operation scenarios, wherein the upper left is the 10km grid airspace hotspot distribution, the upper right is the 5km grid airspace hotspot distribution, and the lower middle is the 1km grid airspace hotspot distribution. DETAILED DESCRIPTION

[0030] In order to make the objects, technical solutions and advantages of the present application clearer, further descriptions will be given below with reference to the drawings.

[0031] The present application discloses a space risk modeling method for autonomous operation scenarios, and the specific implementation process is as shown in Figure 1 , which comprises the following steps:

[0032] Step 1: Establishing an aircraft ellipsoid risk influence area to determine the risk influence range, and gridding the airspace.

[0033] Step 1.1: Establishing an aircraft ellipsoid risk influence area, as shown in Figure 2 . The ellipsoid has a long semi-axis a, a short semi-axis b, and a vertical semi-axis c, and the mathematical expression is: O is the aircraft position, P is any point in the airspace, and θ is the included angle between and the aircraft heading.

[0034] Step 1.2: Airspace gridding. The coordinates describing the grid layout position are defined as grid coordinates, while the coordinates describing the specific position point are defined as actual coordinates. The grid coordinates are represented by G(X, Y, Z), which represents that the grid coordinates are (X, Y, Z), wherein X, Y and Z represent the arrangement order of the grid on the three axes, and the actual coordinates are represented by P(X, Y, Z). The airspace is divided by a cuboid grid, as shown in Figure 3 , which has a length g len , a width g wid , and a height g hei , and the origin coordinates of the coordinate system are (x0, y0). The actual coordinates of the 9 important position points of G(X, Y, Z) are represented as follows:

[0035]

[0036] In the formula, P1 to P8 correspond to vertex 1 to vertex 8 of the cuboid grid, and P C represents the center point of the cuboid grid.

[0037] The airspace is divided into cuboid grids with length and width of 10 km and height of 300 m. In order to minimize the error of grid risk calculation, the risk values at nine important position points of the grid are measured, and the average value is processed to obtain the risk value of the entire grid.

[0038] Step 2: Based on the artificial potential field theory, the real-time risk influence of distance, speed, heading and other factors on each grid of the airspace is considered to construct an airspace risk assessment model.

[0039] Step 2.1: Based on the artificial potential field theory, the general expression of the real-time risk influence of distance, speed, heading and other factors on each position point of the airspace is constructed. The traditional potential field intensity model is:

[0040]

[0041] In the formula: η is the field strength coefficient; d is the distance from the center of the field source to any position; σ is the potential field intensity change rate; E is the potential field intensity.

[0042] The aircraft is the main source of generating airspace safety potential field, and the potential field intensity generated by itself will be affected by its inherent properties and kinematic state parameters. Therefore, on the basis of the traditional potential field intensity model involving only distance, the key influencing factor of the angle between the aircraft speed, heading and the direction of the connecting line of the specific position point is considered, and the real-time risk assessment model of each position point of the airspace is proposed, and the calculation formula is as follows:

[0043]

[0044] In the formula: r is the risk value; is the distance risk function; is the speed risk function; is the heading risk function; η d is the distance risk coefficient; σ d is the distance risk change rate.

[0045] Step 2.2: Correct the distance risk function. For the distance risk function, d is the distance from the center of the field source to any position. Assuming that the coordinates of a point in space are (x, y, z), and the aircraft is regarded as a particle with coordinates (x0, y0, z0), the spatial distance calculation formula is as follows:

[0046]

[0047] Because the measurement standards of horizontal distance and vertical distance are different during the operation of the aircraft, the concept of pseudo-distance is proposed, that is, to judge the proximity of the position point to be evaluated to the aircraft position in the ellipsoid risk area center, and to correct the distance risk function.

[0048] Firstly, the general expression of ellipsoid based on the heading angle, the pitch angle and the roll angle is determined. The attitude of the aircraft in space is determined by the heading angle, the pitch angle and the roll angle. In order to ensure that the direction of the long semi-axis of the aircraft ellipsoid risk influence area is always consistent with the heading of the aircraft, the ellipsoid is rotated according to the heading angle, the pitch angle and the roll angle of the aircraft. In the three-dimensional space, the rotation of the ellipsoid can be defined by the following three basic matrices, which are respectively:

[0049] Rotation matrix R around Z axis (heading angle) z (θ yaw ):

[0050]

[0051] Rotation matrix R around Y axis (pitch angle) y (θ pitch ):

[0052]

[0053] Rotation matrix R around X axis (roll angle) x (θ roll ):

[0054]

[0055] Wherein, θ yaw is the heading angle; θ pitch is the pitch angle; θ roll is the roll angle;

[0056] In the field of aviation, a common combination order for the rotation positioning of the aircraft is to rotate around the Z axis (heading angle) first, then rotate around the Y axis (pitch angle), and finally rotate around the X axis (roll angle), to obtain the rotation matrix R, that is:

[0057] R=R z (θ yaw )×R y (θ pitch )×R x (θ roll )

[0058] Therefore, the coordinates of the rotated point of any point (x, y, z) on the ellipse are (x', y', z'), wherein (x', y', z') = R × (x, y, z).

[0059] Finally, the pseudo distance is calculated. The ellipsoid equation after rotation is: Then the pseudo distance calculation formula is as follows:

[0060]

[0061] When l < 1, it indicates that the position point to be evaluated is located in the range of the ellipsoid risk zone of the aircraft; when l > 1, it indicates that the position point to be evaluated is not affected by the ellipsoid risk zone of the aircraft. The distance risk function is updated as follows:

[0062]

[0063] η(t) = σ(t) - σ(t-1) l is a pseudo-distance risk coefficient; σ(t) is a pseudo-distance risk value at t moment; σ(t-1) is a pseudo-distance risk value at t-1 moment. l is a pseudo-distance risk change rate.

[0064] Step 2.3: Based on the distance, speed, and heading angle risk factors, a specific airspace position risk calculation formula is proposed. The risk value r is a dimensionless absolute risk indicator. Suitable distance risk function, speed risk function, and heading risk function are designed to make the risk value r range in the interval [0, 1]. For the speed risk function, as the aircraft speed increases, its dynamic change in airspace is more rapid, so its potential risk impact on the surrounding space also increases accordingly. For the heading risk function, the relationship between the aircraft and any position point in space includes convergence and dispersion trends. Under the same motion state, the convergence trend between aircrafts has a much greater risk than the dispersion trend.

[0065] Based on the above aircraft safety potential field concept, combined with distance, speed, and direction angle risk factors, an airspace arbitrary position risk calculation formula is proposed as follows:

[0066]

[0067] η(t) = σ(t) - σ(t-1) i r(t) = l(t) * v(t) * v(t) * sin(θ(t)) i l(t) is a pseudo-distance risk value of the aircraft i to the specific position point at t moment; v(t) is the speed of the aircraft i at t moment; v is the reference speed; θ(t) is the angle between the connection line of the aircraft i and the arbitrary position point at t moment. i ref i

[0068] Then, for m aircrafts in the airspace at t moment, the risk value of a position point P in the airspace at the current moment is:

[0069]

[0070] Based on the above airspace risk value calculation formula, the risk impact of the aircraft on the surrounding airspace is as shown in FIG. 2. Figure 4

[0071] Step 3: Set the simulation scene, calculate the correlation between the conflict index and the airspace risk value, and verify the rationality of the model.​​​​

[0072] Step 3.1: Set up the autonomous operation simulation scenario.

[0073] For the simulation basic condition setting, the simulation airspace area is set to 100 km x 100 km, the simulation time is set to 0.5 h, the number of aircrafts is set to 60, the initial speed of the aircrafts is randomly generated, the initial speed allocation range is set to 800-1000 km / h, and the time when the aircrafts enter the airspace is randomly allocated. To meet the safety considerations, the time when the aircrafts of the same height and the same entrance enter the simulation airspace must meet the safety time interval of 5 min.

[0074] For the selection of aircraft entrances and exits, the sector boundary entrances are selected at equal intervals. In order to reflect the height freedom of future airspace route setting, the route design from any one entrance to any other exit is allowed.

[0075] For the embodiment of autonomous operation characteristics in the actual operation process, the connection between the exit and the entrance is regarded as the preset route of the aircraft flight, the preset route direction is calculated, and different height layers are allocated for eastbound and westbound routes. The evolution interval of the aircraft along the preset route is set to 1 min, the heading and speed change in each evolution process with a probability of 30%, the heading change range is-30°-30°, and the speed change range is ±0.1V current , where V current is the current aircraft speed. In order to ensure that the aircraft leaves the simulation airspace along the designated exit, a specific function is set to regulate the behavior of the aircraft, that is, when the distance between the aircraft and the exit is less than a certain threshold, the aircraft is allowed to have a large angle adjustment in each evolution of the heading, so that the heading is quickly adjusted to the direction of the exit, and then flies straight to the exit.

[0076] Step 3.2: Calculate the correlation between the conflict index and the airspace risk value. Select the conflict number and conflict rate indicators, where the conflict rate = conflict number / total number of aircrafts, adjust the number of generated flights, perform multiple independent simulation experiments, and calculate the correlation coefficient between the airspace risk value and each indicator under different flight quantities.

[0077] Step 4: Propose a method for marking airspace hotspots, and visualize the distribution of airspace hotspots in combination with multiple grid scales.

[0078] Step 4.1: Airspace hotspot grid marking. Based on the existing simulation environment, calculate all grid risk values corresponding to different simulation runs, repeat the experiment multiple times independently, record the 0.8 quantile of all grid risk values in each experiment, take the average value as the risk threshold corresponding to the corresponding simulation run, fit the function relationship between the simulation run and the risk threshold, and mark the grid with a risk value greater than the risk threshold as a hotspot grid.

[0079] Step 4.2: Visualize the distribution of hotspots in the airspace by combining multiple grid scales. Based on the function fitting relationship between the simulation sorties and the airspace risk values, the risk threshold corresponding to the simulation sorties of the corresponding airspace is calculated, and the hotspots are visualized by combining the grid sizes of 10 km, 5 km, and 1 km. The schematic diagram is shown in FIG. 10. Figure 5 FIG. 10 shows a schematic diagram of visualizing the distribution of hotspots in the airspace by combining multiple grid scales.

Claims

1. A method for airspace risk modeling for autonomous operation scenarios, characterized in that: Includes the following steps: Step 1: Establish the aircraft ellipsoid risk impact zone to determine the scope of risk impact and rasterize the airspace; Step 2: Based on the artificial potential field theory, construct an airspace risk assessment model by comprehensively considering the real-time risk impact of distance, speed, and heading factors on each grid in the airspace; Step 3: Set up a simulation scenario, calculate the correlation between conflict indicators and airspace risk values, and verify the rationality of the model; Step 4: Propose a spatial hotspot grid labeling method and visualize the distribution of spatial hotspots by combining multiple grid scales; In step 2.1, based on the artificial potential field theory, the general expression for calculating the real-time risk value of each location point in the spatial domain is constructed as follows: The traditional potential field strength model is as follows: ； In the formula: is the field strength coefficient; d is the distance from the center of the field source to any location; The potential field strength is the rate of change of the potential field intensity; E is the potential field intensity. Aircraft are the primary source of airspace safety potential fields, and the strength of these fields is influenced by their inherent properties and kinematic parameters. Therefore, based on traditional potential field strength models that only consider distance, this paper proposes a real-time risk assessment model for various airspace locations, taking into account key influencing factors such as aircraft speed, heading, and the angle between the line connecting the aircraft to a specific location. The calculation formula is as follows: ； ； In the formula: r is the risk value; For distance risk function; For speed risk function; For the heading risk function; Distance risk coefficient; The rate of change of distance risk; In step 2.2, the process of correcting the distance risk function is as follows: For the distance risk function, d is the distance from the source center to any location, and the coordinates of a point in space are (x, y, z). Considering the aircraft as a point mass, its coordinates are (x, y, z). A , y A , z A The formula for calculating spatial distance is as follows: ； Since the measurement standards for horizontal and vertical distances during aircraft operation are different, a pseudo-distance concept is proposed here, which is to judge the proximity of the location to be evaluated to the aircraft's location at the center of the ellipsoidal risk zone and correct the distance risk function. First, the general expression for the ellipsoid based on the heading angle, pitch angle, and roll angle is determined. The aircraft's attitude in space is determined by these three angles. To ensure that the semi-major axis of the risk-affected zone of the aircraft ellipsoid always remains consistent with the aircraft's heading, the ellipsoid is rotated according to the aircraft's heading angle, pitch angle, and roll angle. In three-dimensional space, the rotation of the ellipsoid is defined by the following three fundamental matrices: Rotation matrix of heading angle : ； Rotation matrix of pitch angle : ； Rotation matrix of roll angle : ； We obtain the rotation matrix R, that is: ； in, For heading angle, The pitch angle, This refers to the roll angle; Therefore, the coordinates of any point (x, y, z) on the ellipse after rotation are: ,in ; Finally, the pseudo-distance is calculated; the equation of the ellipsoid after rotation is: The pseudo-distance calculation formula is as follows: ； The major semi-axis of the ellipsoid is a, the minor semi-axis is b, and the vertical semi-axis is c; This indicates that the location to be evaluated is within the risk zone of the ellipsoid; This indicates that the location point to be evaluated is not affected by the risk zone of the aircraft ellipsoid; The distance risk function is updated as follows: ； In the formula: The pseudo-distance risk coefficient; The rate of change of pseudo-distance risk; In step 2.3, considering the risk factors of distance, speed, and heading angle, the formula for calculating the risk at a specific location in the airspace is as follows: Based on the concept of aircraft safety potential field, and considering the risk factors of distance, speed, and directional angle, a formula for calculating the risk at any location in the airspace is proposed, as follows: ； In the formula: Let be the magnitude of the risk value exerted by aircraft i on a specific location point at time t; This is the pseudo-distance from aircraft i to a specific location point; Let t be the velocity of aircraft i at time t; For reference speed; Let be the angle between the direction of the line connecting aircraft i and any point at time t; Given that there are a total of m aircraft in the airspace at time t, the risk value of a certain point P in the airspace at the current time is: 。 2. The airspace risk modeling method for autonomous operation scenarios according to claim 1, characterized in that, Step 1 specifically includes: Step 1.1: Establish the risk impact zone of the aircraft ellipsoid; Step 1.2: Spatial rasterization.

3. The airspace risk modeling method for autonomous operation scenarios according to claim 2, characterized in that, In step 1.1, the mathematical expression for the risk impact zone of the aircraft ellipsoid is: ； The major semi-axis of the ellipsoid is a, the minor semi-axis is b, and the vertical semi-axis is c; In step 1.2, the spatial grid is divided as follows: the coordinates describing the grid layout position are defined as grid coordinates, while the coordinates describing specific location points are defined as actual coordinates; grid coordinates are... This indicates that the raster coordinates are (X, Y, Z), where X, Y, and Z represent the arrangement order of the raster on the three axes, and the actual coordinates are represented by... The spatial domain is represented by a cuboid grid, with dimensions of length, width, and height of [missing information]. , and If the actual coordinates of the origin of the coordinate system are (x0, y0, z0), then for The actual coordinates of the nine key locations are shown below: ； In the formula: P1 to P8 correspond to vertices 1 to 8 of the cuboid grid, respectively. C This represents the center point of the cuboid grid.

4. The airspace risk modeling method for autonomous operation scenarios according to claim 1, characterized in that, Step 3 specifically includes: Step 3.1: Set up the autonomous simulation scenario; Step 3.2: Calculate the correlation between conflict indicators and airspace risk values.

5. The airspace risk modeling method for autonomous operation scenarios according to claim 4, characterized in that, Step 3.1, setting up the autonomous simulation scenario specifically includes: Regarding the basic simulation conditions, the simulation airspace area, simulation time, number of aircraft, initial airspeed of aircraft are randomly generated, and aircraft entry times are randomly assigned. To meet safety requirements, aircraft entering the simulation airspace at the same altitude and entrance must meet the safety time interval requirements. Regarding the selection of aircraft entrances and exits, sector boundary entrances and exits are selected at intervals; in order to reflect the high degree of freedom in the future airspace route designation, route design from any entrance to any other exit is allowed; To reflect the autonomous operation characteristics during actual operation, the line connecting the exit and the entrance is taken as the preset flight path of the aircraft. The direction of the preset flight path is calculated, and different altitude layers are assigned to the eastward and westward flight paths. The evolution interval, the evolution probability of heading and speed and the range of change are set when the aircraft flies along the preset flight path. In order to ensure that the aircraft leaves the simulated airspace along the predetermined exit, specific functions are set to regulate the aircraft behavior. In step 3.2, calculating the correlation between conflict indicators and airspace risk values ​​specifically includes: selecting conflict frequency and conflict rate indicators, where conflict rate = conflict frequency / total number of aircraft; adjusting flight generation frequency; conducting multiple independent simulation experiments; and calculating the correlation coefficient between airspace risk values ​​and each indicator under different flight volumes.

6. The airspace risk modeling method for autonomous operation scenarios according to claim 1, characterized in that, Step 4 specifically includes: Step 4.1: Spatial hotspot grid marking; Step 4.2: Visualize the distribution of spatial hotspots by combining multiple grid scales.

7. The airspace risk modeling method for autonomous operation scenarios according to claim 6, characterized in that, In step 4.1, the spatial hotspot grid marking is specifically as follows: Based on the existing simulation environment, calculate the risk values ​​of all grids corresponding to different simulation flights, repeat the experiment independently multiple times, record the 0.8 quantile of all grid risk values ​​in each experiment, take the average value as the risk threshold corresponding to the simulation flight, fit the functional relationship between the simulation flight and the risk threshold, and mark the grids with risk values ​​greater than the risk threshold as hotspot grids. In step 4.2, the visualization of airspace hotspot distribution by combining multiple grid scales is specifically as follows: based on the function fitting relationship between the number of simulated flights and the airspace risk value, the risk threshold corresponding to the corresponding number of simulated flights in the airspace is calculated, and the hotspots are visualized by combining three granular grid sizes of 10km, 5km, and 1km.

Images