A method, system, device and medium for evaluating the risk of a drone collision

Abstract

The application discloses a kind of unmanned aerial vehicle impact risk assessment method, system, device and medium, comprising: by constructing the spatial proximity relationship of unmanned aerial vehicle and obstacle in airspace, the risk cost data of unmanned aerial vehicle air collision is obtained;According to gas model, the probability of unmanned aerial vehicle crash is calculated, and is combined with the severity model of crash consequence, and the risk cost data of unmanned aerial vehicle ground impact is obtained;Risk cost assessment model is constructed, and the risk cost data of considering city unmanned aerial vehicle air collision and the risk cost data of unmanned aerial vehicle ground impact are quantified and divided into risk grade.The application has the advantages that: the risk cost of unmanned aerial vehicle air and ground can be calculated, and complete risk assessment for unmanned aerial vehicle city operation is realized.According to the spatial proximity relationship, the risk cost of unmanned aerial vehicle air collision is obtained, and when calculating risk cost, the air collision and ground impact risk cost are weighted and distributed, so that they are considered simultaneously and more in line with reality.

Description

Technical Field

[0001] This invention relates to the field of drone collision risk assessment technology, and in particular to a method, system, device and medium for drone collision risk assessment in urban low-altitude airspace. Background Technology

[0002] Urban risk assessment is a prerequisite for drones to operate in urban airspace, and its rationality and accuracy directly determine the drones' ability to fly safely in cities. Drone safety assessment technology needs to comprehensively consider various scenarios of drone flight in cities, combine this with the actual urban air-ground situation, optimize relevant calculations, and construct specific risk assessment methods and standards to ultimately enable drones to operate safely and effectively in cities.

[0003] Existing research has established risk relationships between aircraft by constructing spatial proximity relationships based on ellipsoidal protected area models; determined the horizontal and vertical velocities of UAV crashes by considering relevant factors; determined the probability of mid-air collisions between aircraft by using air probability; and determined the parameters of various types of cover, the degree of injury and death corresponding to different impact energies, the impact area, and the probability of injury and death after a UAV collision.

[0004] However, most of them only consider the risk assessment of drones in a single situation, such as mid-air collision, threat from tall buildings, ground collision, or only consider ideal operating conditions. They rarely take into account real urban geographical and topographical data, such as important factors like urban buildings, no-fly zones, ground population density and distribution. At the same time, they do not involve specific safety assessment models and assessment methods, which leads to limited application scope, low operability, and poor reliability of risk assessment results.

[0005] Existing Technology 1: [Zhang Honghong, Gan Xusheng, Li Shuangfeng, Feng Zheng, Jin Yang. UAV Route Planning Considering Regional Risk Assessment in Complex Low-Altitude Environments] To address the low operational safety of UAVs in complex low-altitude environments, this paper proposes a UAV route planning method considering regional risk assessment, which can quickly generate routes with lower operational risks. First, the complex low-altitude environment is simplified and risk assessed to obtain a low-altitude three-dimensional risk map. Then, the path risk value is used as a comprehensive cost, and an improved ant colony algorithm is used to plan the three-dimensional spatial route, effectively reducing the redundancy of the generated path. Finally, cubic B-splines are used to smooth the planned discrete path, generating a flyable path with continuous curvature and pitch angle. Simulation and experimental results show that the path generated by the proposed algorithm can obtain a low-risk path with a small path cost, while also exhibiting low redundancy and continuous variation in curvature and pitch angle, thus satisfying the performance constraints of UAVs.

[0006] Disadvantages of existing technology 1

[0007] 1. Incomplete evaluation criteria: Only formulas expressing the properties are provided, without specific evaluation criteria.

[0008] 2. The effectiveness of the assessment method is not good: the mid-air collision and ground collision of drones are considered separately, and divided into those that do not consider regional risks and those that do consider regional risks, which does not reflect the actual situation.

[0009] 3. Inadequate assessment conditions: The drone crash probability only considers the probability of crashing due to its own malfunction, and there is no specific calculation formula or value; the formula definition is too idealistic and lacks consideration of the actual air and ground situation.

[0010] Existing technology 2: [Yan Shaokun. Safety Risk Assessment of Unmanned Aerial Vehicle Operation [D]. Civil Aviation University of China, 2018]

[0011] In recent years, the civilian drone industry has experienced explosive growth, especially with a dramatic increase in the number of small and light drones, which now account for the vast majority of civilian drones. However, due to the imperfect drone regulatory system in my country and the difficulty of drone control, safety accidents caused by illegal drone flights are frequent. Some drones have even affected the normal operation of transport aircraft. It is estimated that there are approximately 20,000 drones flying illegally in my country. Therefore, risk assessment of drones is essential, and this paper aims to play a positive role in reducing the operational risks of drones. This paper uses the probabilistic assessment method in quantitative risk assessment to evaluate the operational risks of drones. First, the classification of drones in my country was optimized, and safety evaluation indicators and corresponding safety objectives for the drone industry were determined. Then, a drone crash probability model was established, consisting of various causal models: a drone-transport aircraft collision probability model, a drone-drone collision probability model, and a drone self-failure probability model. The drone-transport aircraft collision and drone-drone collision probability models were established using a gas model, while the drone self-failure probability model was established using a series system reliability model. Subsequently, a crash severity model for unmanned aerial vehicles (UAVs) was established, and the severity of crash consequences for different types of UAVs in different regions of my country was analyzed. Based on the crash probability model and the crash severity model, a UAV operation risk assessment model was established to evaluate the operation risks of different types of UAVs. By comparing the results with predetermined safety targets for the UAV industry, it was concluded that the current safety level of my country's UAV industry is extremely low. Methods to improve safety levels include reducing the risk of collisions between UAVs and transport aircraft and the failure rate of UAVs themselves. Reasonable suggestions for UAV safety management in my country are provided based on these risk reduction methods. Finally, the UAV risk assessment model was applied to analyze the setting of restricted flight zones for UAV airports in my country, and optimization schemes were proposed.

[0012] Disadvantages of existing technology 2

[0013] 1. The effectiveness of the assessment method is inadequate: it only considers the ground damage caused by the drone crash, and lacks the mid-air impact component of the drone in actual operation.

[0014] 2. Incomplete constraints: The lack of real airspace and geographical data means that the formulas for collision area and fall velocity need further optimization; moreover, most fall probabilities are conservative estimates with low accuracy.

[0015] 3. The assessment results have limitations: the risk assessment only applies to different types of drone operations, and there is no specific assessment method for drones operating in real terrain.

[0016] Existing Technology 3: [Hu Xinting, Dai Fuqing. Risk Assessment of UAV Operation Based on Pedestrian Safety in Urban Areas] To ensure the safe operation of unmanned aerial vehicles (UAVs) in urban areas, a risk assessment model for UAV operation based on pedestrian safety in urban areas is established. First, the threat posed to pedestrians by UAVs during urban operation is analyzed, and risk modeling is performed. Second, risk measurement standards are introduced, and risk cost calculation indicators are established. Then, a UAV urban operation risk map is generated. Finally, using risk-avoidance-based UAV path planning as an application scenario, the flight path of UAVs in urban areas is solved. Results show that the risk costs differ significantly in different areas, and the degree of risk is related to population density and environmental shielding effects. Compared with the traditional shortest path planning method, the path planning method considering operational risks can enable UAVs to avoid areas with higher risk levels, improving the safety of UAV operation in urban areas.

[0017] Disadvantages of existing technology three

[0018] 1. The assessment process is incomplete: Firstly, the formula lacks a method for calculating the exposure area; secondly, it lacks a method for determining densely populated and general areas; and finally, it only provides a specific risk value but lacks further criteria for its classification.

[0019] This would result in a lower practicality of the assessment.

[0020] 2. The effectiveness of the assessment method is not good: First, it does not take into account the impact of airspace structure on flight safety and ignores the airspace risks of UAVs; second, it lacks consideration of real data and is too idealistic. Summary of the Invention

[0021] This invention addresses the shortcomings of existing technologies by providing a method, system, device, and medium for assessing drone collision risks. By considering both ground and aerial collisions in the context of urban airspace and ground conditions, the invention obtains risk cost values ​​and classifies risk levels, effectively achieving risk assessment in complex urban low-altitude airspace.

[0022] To achieve the above-mentioned objectives, the technical solution adopted by the present invention is as follows:

[0023] A method for assessing drone collision risk includes the following steps:

[0024] S1. By constructing the spatial proximity relationship between the UAV and obstacles in the airspace, we obtain the risk cost data of UAV mid-air collision;

[0025] S2. Calculate the probability of drone crash based on the gas model, and combine it with the crash consequence severity model to obtain the risk cost data of drone ground impact.

[0026] S3. Construct a risk cost assessment model to quantify and classify the risk cost data considering urban drone aerial collisions and drone ground collisions.

[0027] Furthermore, the sub-steps for calculating the cost of mid-air collisions between drones in S1 are as follows:

[0028] S11. The point within a certain area closest to the drone that is a static obstacle will be considered a critical point. Static obstacles include: buildings, tall trees, and large road signs.

[0029] Let i be the key point corresponding to UAV i at time t. p Then the drone i and the key point i p The spatial proximity relationship between them is Represented as:

[0030]

[0031] In the formula: For time t, the drone i reaches the key point i p Horizontal distance, km; D ip For the i-th p The reference distance for each key point is 1m.

[0032] The closer the distance between the drone and the key point is to the set reference distance, the larger the corresponding spatial proximity value. When it is less than or equal to the set reference distance, the spatial proximity value reaches the maximum value of 1.

[0033] Furthermore, take D ip It is 15km.

[0034] S12, Let Let i be the drone and i be the key point. p The rate of change of the spatial proximity relationship between them at time t is expressed as:

[0035]

[0036] S13. Based on the spatial proximity and proximity rate between the UAV and key points, propose the risk cost between the UAV and key points. The calculation method is as follows:

[0037]

[0038] Where: β p A is an adjustment factor for the proximity rate between the drone and key points; d The adjustment factor is set to 10. -6 .

[0039] The risk cost R1 of a mid-air collision with a drone is expressed as:

[0040]

[0041] Furthermore, the calculation of the cost of UAV ground collision risk in S2 involves the following sub-steps:

[0042] S21. The probability of a drone crash is expressed as follows:

[0043] P = P(C1) + P(C2) (5)

[0044] In the formula: P(C1) is the probability of the drone crashing due to a collision with an obstacle; P(C2) is the probability of the drone crashing due to its own failure.

[0045] The probability model for a drone colliding with an obstacle is expressed as follows:

[0046]

[0047] In the formula: A is the spatial area of ​​the horizontal region under study, in meters. 2 N is the number of drones in the airspace; g is the horizontal length of the drone (m); h is the vertical length of the drone (m); H is the airspace altitude (m); E(V) r () represents the expected relative velocity, in m / s;

[0048]

[0049] In the formula: V0 is the operating speed of the UAV, m / s; β r The relative directional angle between the drone and the obstacle is expressed in degrees.

[0050] The failure probability model of an unmanned aerial vehicle (UAV) system is expressed as follows:

[0051]

[0052] P(C2)=λ s =6.04×10 -5 (8)

[0053] S22, The severity model for the consequences of a plane crash is as follows:

[0054] S = N exp FI (9)

[0055] Where: N exp F represents the number of people who collide with the drone; I represents the probability of injury or death after a drone collision; and F represents the severity of injury or death. The severity of a drone crash is the degree of injury or death caused to people on the ground during the collision between the drone and the ground. The higher the degree of injury or death, the higher the severity, and vice versa.

[0056]

[0057] In the formula: p s As a shielding factor, trees, buildings, and industrial areas can buffer the speed of a drone during a crash. s The larger the value, the greater the buffering effect. s The smaller the value, the less effective the buffering effect; E imp For the impact kinetic energy, J, by Calculated, v x v y Let be the instantaneous horizontal and vertical velocities at the moment of impact, respectively, in m / s; α be the velocity of p. s =0.5 is the impact energy required to achieve a 50% mortality rate, J; β is the impact energy required when p s The impact energy threshold, J, required for death to approach 0.

[0058] Based on the force profile of the drone during its descent, and by incorporating the initial velocity boundary conditions, the crash zone is determined. Then, the instantaneous horizontal and vertical velocities v at the moment of impact are calculated. x v y :

[0059]

[0060] In the formula, m is the mass of the UAV (kg); g is the acceleration due to gravity (m / s²). 2 v0 is the initial velocity value, m / s; C D ρ is the air drag coefficient. A air density, kg / m³ 3 t is the falling time, in seconds; A x A y These are the lateral and vertical windward areas, respectively, in meters. 2 .

[0061]

[0062] In the formula, h is the operating altitude of the UAV, in meters (m).

[0063] N exp =A exp ρ (13)

[0064] In the formula: ρ is the population density, people / km 2 A exp The ground impact area of ​​the drone is m. 2 .

[0065] During the drone's impact with the ground, the area of ​​impact is calculated as the drone continues to glide when it reaches the average height of a pedestrian. A exp Represented as:

[0066] A exp =2×(r) uav +r p )×d+π(r uav +r p ) 2 14

[0067] In the formula: r p Let r be the average pedestrian radius, in meters (m); uav The maximum radius of the drone is m; d = H P cotθ,H P Let θ represent the average height of pedestrians, and θ represent the angle at which the drone veers and falls, which can be expressed by the formula... Sure.

[0068] shading coefficients p for various types s The settings are as follows:

[0069]

[0070] The degree of injury or death (I) is determined by the impact energy as shown in the table below:

[0071]

[0072] The ground risk cost R2 for drones is the product of the probability of a drone crash and the severity of casualties, expressed as:

[0073] R2 = PS

[0074] Furthermore, the risk cost calculation and risk level classification in S3 involve the following sub-steps:

[0075] S31. Construct a risk cost assessment model:

[0076]

[0077] In the formula: R is the risk value, composed of the air collision risk cost R1 and the ground collision risk cost R2, but R = δ1R1 + δ2R2; k is an adjustment coefficient; q is the hazard coefficient of densely populated areas; R' is the safety level of UAV operation, in people / hour. The larger the risk cost C, the lower the safety and the higher the degree of danger. The distinction between densely populated areas and general areas is based on a population density of 2 × 10⁻⁶. 4 people / km 2 This serves as a boundary to divide dense and general areas.

[0078] S32. The urban open space situation is quantified as shown in the following table:

[0079]

[0080] S33. Risk Level Classification:

[0081] The calculated risk cost is greater than or equal to the population density, where ρ = 2 × 10. 4 people / km 2 , , obscuring coefficient f = 0.2, flight speed u = 20 m / s, R1 = 6 × 10 -7 The area with the highest risk cost is considered a high-risk area.

[0082] The calculated risk cost is lower than that of high risk but greater than or equal to ρ = 1 × 10 3 people / km 2 , obscuring coefficient f = 0.5, flight speed u = 15 m / s, R1 = 3 × 10 -7 The area with the highest risk cost is the medium-risk area;

[0083] The calculated risk cost is lower than that of medium risk but greater than or equal to ρ = 200 people / km 2 , obscuring factor f = 0.75, flight speed u = 5 m / s, R1 = 1 × 10 -7 The area with the highest risk cost is considered a low-risk area.

[0084] The risk cost is less than ρ = 200 people / km 2 , obscuring factor f = 0.75, flight speed u = 5 m / s

[0085] R1 = 1 × 10 -7 The area where the risk cost value is located is risk-free.

[0086] The present invention also discloses a drone collision risk assessment system for urban low-altitude airspace. The system can be used to implement the above-mentioned drone collision risk assessment method for urban low-altitude airspace. Specifically, it includes: a drone air collision risk cost assessment module, a drone ground collision risk cost assessment module, and a risk level structure assessment module.

[0087] The risk and cost assessment module for drone mid-air collisions: By constructing the spatial proximity relationship between the drone and obstacles in the airspace, the risk and cost data of drone mid-air collisions are obtained;

[0088] The risk and cost assessment module for drone ground collisions calculates the probability of drone crashes based on a gas model and combines it with a crash severity model to obtain risk and cost data for drone ground collisions.

[0089] Risk level structure assessment module: Constructs a risk cost assessment model to quantify and classify the risk cost data considering urban drone aerial collisions and drone ground collisions.

[0090] The present invention also discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-mentioned method for assessing the collision risk of unmanned aerial vehicles (UAVs) in urban low-altitude airspace.

[0091] The present invention also discloses a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for assessing drone collision risks in urban low-altitude airspace.

[0092] Compared with the prior art, the advantages of the present invention are as follows:

[0093] 1. It can actually calculate the air and ground risk costs of drones, obtain risk cost data, and has complete risk assessment standards. It can achieve a complete risk assessment of drone operation in cities, breaking through the complex problems of difficulty in classifying risk types and quantifying risk values ​​in the traditional sense.

[0094] 2. The risk cost of mid-air collision with UAVs is derived based on spatial proximity, and the risk cost is weighted to account for both mid-air collision and ground collision risks when calculating the risk cost, so that it is considered simultaneously and is more in line with reality.

[0095] 3. The urban airspace situation is realistically considered, including population density, air resistance coefficient, flight altitude, etc., and the formulas such as collision area and impact speed are optimized accordingly to make them conform to the actual situation. Attached Figure Description

[0096] Figure 1 This is an overall flowchart of a drone collision risk assessment method according to an embodiment of the present invention;

[0097] Figure 2 This is a schematic diagram illustrating the spatial proximity of a mid-air collision between drones according to an embodiment of the present invention.

[0098] Figure 3This is a schematic diagram of a gas model for the collision rate between an unmanned aerial vehicle and an obstacle in an embodiment of the present invention;

[0099] Figure 4 This is a schematic diagram showing the crash speed and ground impact area of ​​a drone according to an embodiment of the present invention. Detailed Implementation

[0100] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and examples.

[0101] A method for assessing drone collision risk includes the following steps:

[0102] S1. By constructing the spatial proximity relationship between the drone and obstacles in the airspace, the risk cost of drone mid-air collision is obtained.

[0103] S2. Calculate the probability of drone crash based on the gas model, and combine it with the crash consequence severity model to obtain the risk cost of drone ground impact.

[0104] S3. Construct a risk cost assessment model to quantify and classify the risk levels of urban open space.

[0105] The following is a detailed analysis:

[0106] Drone mid-air collisions and ground collisions are considered as two risk scenarios, together constituting the risk cost. A risk cost model is constructed, and finally, risk levels are classified, such as... Figure 1 As shown in the flowchart.

[0107] Calculation of the risk and cost of mid-air collision between drones:

[0108] S11. When drones fly in low-altitude urban environments, buildings, tall trees, large road signs, etc., are considered static obstacles that pose a threat to drones. The point within a certain area where the static obstacle is closest to the drone will be considered a critical point.

[0109] Let i be the key point corresponding to UAV i at time t. p Then the drone i and the key point i p The spatial proximity relationship between them is like Figure 2 As shown, it is represented as:

[0110]

[0111] In the formula: For time t, the drone i reaches the key point i p Horizontal distance, km; D ip For the i-th p Reference distances to key points, in km.

[0112] Because in actual area control handover processes, communication handover often takes place 15km from the handover point, therefore, we take D. ip The distance is 15km. The closer the distance between the UAV and the key point is to the set reference distance, the larger the corresponding spatial proximity value. When it is less than or equal to the set reference distance, the spatial proximity value reaches the maximum value of 1.

[0113] S12, Let Let i be the drone and i be the key point. p The rate of change of the spatial proximity relationship between them at time t is expressed as:

[0114]

[0115] S13. Based on the spatial proximity and proximity rate between the UAV and key points, propose the risk cost between the UAV and key points. The calculation method is as follows:

[0116]

[0117] Where: β p A is an adjustment factor for the proximity rate between the drone and key points; d The adjustment factor is set to 10. -6 .

[0118] The risk cost R1 of a mid-air collision with a drone is expressed as:

[0119]

[0120] Calculation of the cost of drone ground collision risk:

[0121] Calculating the risk cost of drone ground collisions requires a large amount of information. Tall buildings are the primary obstacles to drone airspace flight, and collision models need to be established considering building height. Ground population density and the area of ​​ground obstructions will affect the severity of drone crashes. Drone flight altitude and air drag coefficient will affect drone fall speed. In addition, there are air humidity, ground topography, wind disturbance and other air-ground situational factors. Based on the principles of flight safety, effectiveness, minimization of redundant information, and universality, the following six main air-ground situations are selected for risk assessment.

[0122] Table 1

[0123]

[0124]

[0125] S21. Collisions between drones and obstacles in the airspace and drone malfunctions are the main causes of drone crashes. In addition, human error and environmental factors also play a role, but because they are not easily quantifiable and have extremely low probabilities, they are not considered in the drone crash probability. Therefore, the drone crash probability is expressed as:

[0126] P = P(C1) + P(C2) (5)

[0127] In the formula: P(C1) is the probability of the drone crashing due to a collision with an obstacle; P(C2) is the probability of the drone crashing due to its own failure.

[0128] Gas models can be used to describe the expected mid-air collision rate between aircraft, such as... Figure 3 As shown, calculations show that when the distribution of UAVs is consistent within a height layer H, and H is much greater than the UAV's height h, the probability model for a UAV colliding with an obstacle can be expressed as:

[0129]

[0130] In the formula: A is the spatial area of ​​the horizontal region under study, in meters. 2 N is the number of drones in the airspace; g is the horizontal length of the drone (m); h is the vertical length of the drone (m); H is the airspace altitude (m); E(V) r () represents the expected relative velocity, in m / s;

[0131]

[0132] In the formula: V0 is the operating speed of the UAV, m / s; β r The relative directional angle between the drone and the obstacle is expressed in degrees.

[0133] The failure rates of each system were obtained through literature review:

[0134] Table 2

[0135]

[0136] The unmanned aerial vehicle (UAV) system is a series system, and the service life of each system follows an exponential distribution. Therefore, the failure probability model of the UAV system can be expressed as:

[0137]

[0138] P(C2)=λ s =6.04×10 -5 (8)

[0139] S22. Model for the severity of consequences of a plane crash:

[0140] S = Nexp FI (9)

[0141] Where: N exp F represents the number of people who collide with the drone; I represents the probability of injury or death after a drone collision; and F represents the severity of injury or death. The severity of a drone crash is the degree of injury or death caused to people on the ground during the collision between the drone and the ground. The higher the degree of injury or death, the higher the severity, and vice versa.

[0142]

[0143] In the formula: p s As a shielding factor, trees, buildings, and industrial areas can buffer the speed of a drone during a crash. s The larger the value, the greater the buffering effect. s The smaller the value, the less effective the buffering effect; E imp For the impact kinetic energy, J, by Calculated, v x v y Let be the instantaneous horizontal and vertical velocities at the moment of impact, respectively, in m / s; α be the velocity of p. s =0.5 is the impact energy required to achieve a 50% mortality rate, J; β is the impact energy required when p s The impact energy threshold, J, required for death to approach 0.

[0144] The descent of a drone typically follows a projectile motion. By analyzing the forces acting on the drone during its descent and applying initial velocity boundary conditions, the crash zone can be determined. From this, the instantaneous horizontal and vertical velocities v at the moment of impact can be calculated. x v y :

[0145]

[0146] In the formula, m is the mass of the UAV (kg); g is the acceleration due to gravity (m / s²). 2 v0 is the initial velocity value, m / s; C D ρ is the air drag coefficient. A air density, kg / m³ 3 t is the falling time, in seconds; A x A y These are the lateral and vertical windward areas, respectively, in meters. 2 .

[0147]

[0148] In the formula, h is the operating altitude of the UAV, in meters (m).

[0149] N exp =A exp ρ ⒀

[0150] In the formula: ρ is the population density, people / km 2 A exp The ground impact area of ​​the drone is m. 2 .

[0151] During the drone's impact with the ground, the area of ​​impact is calculated by the drone continuing to glide when it reaches the average height of a pedestrian on the ground. Figure 4 As shown, A exp Represented as:

[0152] A exp =2×(r) uav +r p )×d+π(r uav +r p ) 2 14

[0153] In the formula: r p Let r be the average pedestrian radius, in meters (m); uav The maximum radius of the drone is m; d = H P cotθ,H P Let θ represent the average height of pedestrians, and θ represent the angle at which the drone veers and falls, which can be expressed by the formula... Sure.

[0154] Different types of ground cover in urban areas have varying effects on buffering the crash velocity of drones; the cover coefficient p for each type... s The settings are as follows:

[0155] Table 3

[0156]

[0157] The degree of casualties (I) is determined by the impact energy:

[0158] Table 4

[0159]

[0160]

[0161] The ground risk cost R2 for drones is the product of the probability of a drone crash and the severity of casualties, expressed as:

[0162] R2 = PS

[0163] 3) Risk cost calculation and risk level classification:

[0164] Risk cost is a crucial indicator in safety risk assessment, representing the degree of danger posed by drone operation. Constructing a risk cost assessment model:

[0165]

[0166] In the formula: R is the risk value, composed of the air collision risk cost R1 and the ground collision risk cost R2, but R = δ1R1 + δ2R2; k is an adjustment coefficient; q is the hazard coefficient of densely populated areas; R' is the safety level of UAV operation, in people / hour. The larger the risk cost C, the lower the safety and the higher the degree of danger. The distinction between densely populated areas and general areas is based on a population density of 2 × 10⁻⁶. 4 people / km 2 This serves as a boundary to divide dense and general areas.

[0167] Quantifying the situation of urban open spaces:

[0168] Table 5

[0169]

[0170]

[0171] Risk level classification:

[0172] The calculated risk cost is greater than or equal to the population density, where ρ = 2 × 10. 4 people / km 2 , , obscuring coefficient f = 0.2, flight speed u = 20 m / s, R1 = 6 × 10 -7 The area with the highest risk cost is the high-risk area; the calculated risk cost is lower than that of high-risk areas but greater than or equal to ρ = 1 × 10. 3 people / km 2 , obscuring coefficient f = 0.5, flight speed u = 15 m / s, R1 = 3 × 10 -7 The area with the highest risk cost is considered a medium-risk area; the calculated risk cost is lower than that of medium-risk but greater than or equal to ρ = 200 people / km. 2 , obscuring factor f = 0.75, flight speed u = 5 m / s, R1 = 1 × 10 -7 The area with the highest risk cost is considered a low-risk area; the risk cost value is less than ρ = 200 people / km. 2 , obscuring factor f = 0.75, flight speed u = 5 m / s, R1 = 1 × 10 -7 The area where the risk cost value is located is risk-free.

[0173] In another embodiment of the present invention, a drone collision risk assessment system for urban low-altitude airspace is provided. This system can be used to implement the above-mentioned drone collision risk assessment method. Specifically, it includes: a drone air collision risk cost assessment module, a drone ground collision risk cost assessment module, and a risk level structure assessment module.

[0174] The risk and cost assessment module for drone mid-air collisions: By constructing the spatial proximity relationship between the drone and obstacles in the airspace, the risk and cost data of drone mid-air collisions are obtained;

[0175] The risk and cost assessment module for drone ground collisions calculates the probability of drone crashes based on a gas model and combines it with a crash severity model to obtain risk and cost data for drone ground collisions.

[0176] Risk level structure assessment module: Constructs a risk cost assessment model to quantify and classify the risk cost data considering urban drone aerial collisions and drone ground collisions.

[0177] In another embodiment of the present invention, a terminal device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to achieve a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used in the operation of a drone collision risk assessment method, including the following steps:

[0178] S1. By constructing the spatial proximity relationship between the UAV and obstacles in the airspace, we obtain the risk cost data of UAV mid-air collision;

[0179] S2. Calculate the probability of drone crash based on the gas model, and combine it with the crash consequence severity model to obtain the risk cost data of drone ground impact.

[0180] S3. Construct a risk cost assessment model to quantify and classify the risk cost data considering urban drone aerial collisions and drone ground collisions.

[0181] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory). This computer-readable storage medium is a memory device in a terminal device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and extended storage media supported by the terminal device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device.

[0182] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the drone collision risk assessment method in the above embodiments; one or more instructions in the computer-readable storage medium are loaded and executed by the processor to perform the following steps:

[0183] S1. By constructing the spatial proximity relationship between the UAV and obstacles in the airspace, we obtain the risk cost data of UAV mid-air collision;

[0184] S2. Calculate the probability of drone crash based on the gas model, and combine it with the crash consequence severity model to obtain the risk cost data of drone ground impact.

[0185] S3. Construct a risk cost assessment model to quantify and classify the risk cost data considering urban drone aerial collisions and drone ground collisions.

[0186] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0187] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0188] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0189] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0190] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the implementation methods of the present invention, and should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of the present invention.

Claims

1. A method for assessing the collision risk of unmanned aerial vehicles (UAVs), characterized in that, Includes the following steps: S1. By constructing the spatial proximity relationship between the UAV and obstacles in the airspace, we obtain the risk cost data of UAV mid-air collision; The sub-steps for calculating the risk cost of a mid-air collision with a drone are as follows: S11. The point within a certain area where a static obstacle is closest to the drone is considered a key point; Static obstacles include: buildings, tall trees, and large road signs. Let the drone be at time t. The corresponding key points are Then drone and key points The spatial proximity relationship between them is , represented as: （1） In the formula: for Time Drone To the key point Horizontal distance, ; For the first Reference distances to key points ; The closer the drone is to the key point, the larger the corresponding spatial proximity value. When the distance is less than or equal to the set reference distance, the spatial proximity value reaches the maximum value of 1. S12, Let For drones and key points are The rate of change of the spatial proximity relationship between them at time t is expressed as: （2） S13. Based on the spatial proximity and proximity rate between the UAV and key points, propose the risk cost between the UAV and key points. The calculation method is as follows: （3） In the formula: This is an adjustment factor for the proximity rate between the drone and key points; To adjust the coefficient, take the following values: ; Risks and costs of drone mid-air collisions Represented as: （4）； S2. Calculate the probability of drone crash based on the gas model, and combine it with the crash consequence severity model to obtain the risk cost data of drone ground impact. S3. Construct a risk cost assessment model to quantify and classify the risk cost data considering urban drone aerial collisions and drone ground collisions.

2. The method for assessing drone collision risk according to claim 1, characterized in that: Pick It is 15km.

3. The method for assessing drone collision risk according to claim 1, characterized in that: The calculation of the cost of UAV ground collision risk in S2 includes the following sub-steps: S21. The probability of a drone crash is expressed as follows: （5） In the formula: The probability of a drone colliding with an obstacle and crashing; The probability of a drone crashing due to its own failure; The probability model for a drone colliding with an obstacle is expressed as follows: （6） In the formula: The area of ​​the horizontal region under study. ; The number of drones in the airspace; The horizontal length of the drone. ; The vertical length of the drone. ; For airspace altitude, ; For the expected relative speed, ; （7） In the formula: For the drone's operating speed, ; The relative azimuth angle between the drone and the obstacle. The failure probability model of an unmanned aerial vehicle (UAV) system is expressed as follows: ， （8） S22, The severity model for the consequences of a plane crash is as follows: （9） In the formula: The number of people who collided with the drone; The probability of injury or death after a drone collision; The severity of casualties refers to the degree of injury or death caused to people on the ground during the collision between the drone and the ground. The higher the degree of casualties, the higher the severity, and the lower the degree of casualties, the lower the severity. （10） In the formula: As a shielding factor, trees, buildings, and industrial areas can act as a buffer against the speed of a drone during a crash. The larger the size, the greater the buffering effect. The smaller the value, the weaker the buffering effect; For impact kinetic energy, ,Depend on Calculated These are the instantaneous horizontal velocity and vertical velocity at the moment of impact. ; for The impact energy required to achieve a 50% fatality rate. ; For when The impact energy threshold required for death to approach zero. ; Based on the force profile of the drone during its descent, and by incorporating the initial velocity boundary conditions, the crash zone is determined. Then, the instantaneous horizontal and vertical velocities at the moment of impact are calculated. : （11） In the formula For UAV quality, ; It is the acceleration due to gravity. ; This is the initial velocity value. ; This refers to the air drag coefficient; air density, ; For the time of descent, ; , These are the lateral and vertical windward areas, respectively. ; （12） In the formula The operating altitude of the drone, ; （13） In the formula: For population density, ; The ground impact area of ​​the drone. ; The area of ​​impact with the ground is calculated by the drone continuing to glide when it reaches the average height of a pedestrian. Represented as: （14） In the formula: For the average pedestrian radius, ; The maximum radius of the drone ; , The average height of pedestrians, The angle at which a drone descends and falls can be expressed by the formula. Sure; Shading coefficients of various types The settings are as follows: ， Casualty level The impact energy is determined as shown in the table below: ， Ground risks and costs of drones This is the product of the probability of a drone crash and the severity of casualties, expressed as: 。 4. The method for assessing drone collision risk according to claim 1, characterized in that: The sub-steps for calculating risk costs and classifying risk levels in S3 are as follows: S31. Construct a risk cost assessment model: ， In the formula: The risk value is determined by the cost of the risk of an aerial collision. and the cost of ground impact risk Composition, but ; For adjustment coefficients; Risk factor for densely populated areas; To improve the safety level of drone operations, ; Risk Cost The larger the value, the lower the safety and the higher the risk of operation. Dense areas and general areas are distinguished by population density. Used as a boundary to divide dense and general areas; S32. The urban open space situation is quantified as shown in the following table: ， S33. Risk Level Classification: The calculated risk cost is greater than or equal to the population density. shading coefficient Flight speed , The area with the highest risk cost is considered a high-risk area. The calculated risk cost is lower than that of high risk but greater than or equal to that of high risk. shading coefficient Flight speed , The area with the highest risk cost is the medium-risk area; The calculated risk cost is lower than that of medium risk but greater than or equal to that of medium risk. shading coefficient Flight speed , The area with the highest risk cost is considered a low-risk area. Risk cost value is less than shading coefficient Flight speed , The area where the risk cost value is located is risk-free.

5. A drone collision risk assessment system for urban low-altitude airspace, characterized in that: The system can be used to implement the UAV collision risk assessment method according to any one of claims 1 to 4, including: a UAV air collision risk cost assessment module, a UAV ground collision risk cost assessment module, and a risk level structure assessment module; The risk and cost assessment module for drone mid-air collisions: By constructing the spatial proximity relationship between the drone and obstacles in the airspace, the risk and cost data of drone mid-air collisions are obtained; The risk and cost assessment module for drone ground collisions calculates the probability of drone crashes based on a gas model and combines it with a crash severity model to obtain risk and cost data for drone ground collisions. Risk level structure assessment module: Constructs a risk cost assessment model to quantify and classify the risk cost data considering urban drone aerial collisions and drone ground collisions.

6. A computer device, characterized in that: The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the drone collision risk assessment method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: It stores a computer program that, when executed by a processor, implements the drone collision risk assessment method according to any one of claims 1 to 4.

Images