Course cone-based unmanned aerial vehicle collision probability calculation method

Abstract

The invention provides an unmanned aerial vehicle collision probability calculation method based on a course cone, and the method specifically comprises the steps: carrying out the design of an unmanned aerial vehicle collision region through a mathematical model, and taking the minimum circumscribed sphere of an unmanned aerial vehicle as the unmanned aerial vehicle collision region; combining the collision areas into a combined sphere collision area, and endowing the combined sphere collision area with the invaded unmanned aerial vehicle; based on collision rule setting of random heading, the heading of the two unmanned aerial vehicles is synthesized into a heading vector of the invading unmanned aerial vehicle; a concept of a collision course cone is used for representing all relative course sets which can be collided in a three-dimensional space, and a collision probability is converted into a ratio of a spherical crown area of an invading unmanned aerial vehicle course ball in the course cone to a surface area of the course ball to be used for measuring the collision probability of the unmanned aerial vehicle; and calculating the safety interval of the unmanned aerial vehicle through the safety level, and calibrating the safety interval. The invention provides a method for calculating the collision probability of the unmanned aerial vehicles, the collision course cone is used for identifying the course set with the collision risk, and quantification and evaluation of the collision risk of the two unmanned aerial vehicles are achieved.

Description

Technical Field

[0001] The invention relates to the field of low-altitude airspace management, and in particular to a method for calculating the collision probability of unmanned aerial vehicles based on a heading cone. Background Art

[0002] In recent years, with the gradual opening of the country to low-altitude airspace, various industries have increasingly used low-altitude airspace. Drones have become the main aircraft operating in low-altitude airspace due to their low price, strong portability, and high accessibility. They are widely used in logistics, aerial photography, reconnaissance, detection and other fields. As drones have gradually been integrated into the national airspace system in recent years, drones are about to become the main body of low-altitude airspace operations. However, with the expansion of the scale of drone operations, drone operation safety has become a major factor hindering the development of the industry.

[0003] Drone collision risk assessment is based on the performance characteristics of the drone, such as its heading and cruising speed, and considers various risk factors to determine the probability of a collision between an individual drone and other aircraft.

[0004] The collision probability problem stems from the randomness of the UAV's heading. The number of elements in the heading set is too large to be counted using the traversal method, and it is necessary to consider the geometric configuration of the heading for equivalent substitution. The existing technology is mostly based on the UAV dynamics model and three-dimensional trajectory for collision judgment, which is complex and time-consuming to calculate, and cannot quantify the collision risk well. In the context of increasingly busy low-altitude airspace, traditional judgment methods are difficult to meet the needs of actual development.

[0005] Therefore, this paper makes improvements on the traditional collision zone model and proposes a collision probability calculation method based on heading cone. Summary of the invention

[0006] In response to the current demand for a method for calculating the collision risk probability of unmanned aerial vehicles, the present invention proposes a method for calculating the collision probability based on a heading cone.

[0007] The present invention adopts the following technical solutions to solve the technical problems:

[0008] Step 1: Set the minimum circumscribed sphere of the drone as the drone collision zone.

[0009] Step 2: When studying the collision process of two drones, a combined spherical collision zone is formed based on the collision zone radius of each drone and assigned to the main drone;

[0010] Step 3: Perform kinematic analysis on the headings of the two drones to obtain the relative heading, and assign the relative heading to the other invading drone;

[0011] Step 4: Set the collision rules. When the relative heading passes through the collision zone of the combined sphere, it is determined that a collision has occurred, and the collision probability of the drones is calculated using the collision heading cone;

[0012] Step 5: Calibrate the drone interval based on the safety level.

[0013] The collision heading cone concept proposed in this invention uses surface area geometry to replace the heading set whose elements are too large to be counted using the traversal method, thereby quantifying and evaluating the collision risk between two drones. Compared with the traditional collision probability calculation method, the calculation is simpler and more timely. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a core flow chart of the content involved in the present invention;

[0015] Figure 2 A two-dimensional schematic diagram of the combined spherical safety zone

[0016] Figure 3 Schematic diagram of two-dimensional relative heading

[0017] Figure 4 Schematic diagram of the drone's heading

[0018] Figure 5 is the collision heading cone 2D map DETAILED DESCRIPTION

[0019] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings.

[0020] The method for calculating the collision probability of UAVs based on the collision heading cone is as follows: Figure 1 As shown, the specific steps are as follows.

[0021] Step 1: Set the minimum circumference of the drone as the drone collision zone

[0022] (1) Obtaining UAV shape parameters: Collect the size parameters of the UAV in various postures, analyze the size parameters of the UAV in each posture, and take the maximum value as the data for constructing the UAV collision zone.

[0023] (2) Calculate the minimum circumscribed sphere radius λ and set the drone collision zone based on λ. The formulas are shown in (1)(2)(3).

[0024] Step 2: Merge the collision zones of the two drones into a combined sphere collision zone

[0025] (1) By establishing the UAV sphere collision zone, the UAV collision problem can be converted into the position relationship between two spheres in three-dimensional space. In order to simplify the entire research process, the UAV safety zone is transferred to one of the UAVs. The transferred safety zone is called the combined sphere collision zone, and its radius is λ1+λ2.

[0026] (2) One of the drones in the collision zone of the combined sphere can be considered as the invaded drone, and the other drone is regarded as a particle and can be considered as the invading drone. The schematic diagram is shown in Figure 2 shown.

[0027] Step 3: Perform kinematic analysis on the headings of the two drones to obtain the relative headings

[0028] (1) The motion of the two drones is considered as the mutual motion between them. The heading vectors are synthesized into a relative heading vector, which is assigned to the invading drone. The invaded drone is considered to be relatively stationary, such as Figure 3 shown.

[0029] (2) Due to the flexibility of the UAV, it is difficult to predict the flight direction of the UAV at the next moment. The flight direction of the UAV at the next moment can be simplified to be randomly and uniformly distributed in three-dimensional space, such as Figure 4 As shown, formula (6) is as follows:

[0030]

[0031] in Represents the polar angle, ranging from [0,π], Indicates the azimuth angle, ranging from [0,2π].

[0032] Step 4: Set collision rules and use collision heading cone to calculate the probability of drone collision

[0033] (1) The collision judgment when the spherical collision zones of two drones touch each other is converted into a collision risk when the heading of the invading drone points to the invaded drone.

[0034] (2) The collision probability of UAVs is calculated using the collision heading cone method, such as Figure 5 As shown, the specific method is as follows:

[0035] (a) Considering that it is difficult to count the heading of a UAV in three-dimensional space, the geometric configuration of the heading is used as an equivalent replacement, as shown in formula (7):

[0036]

[0037] Where S collision represents the surface area of ​​the intersection between the collision heading cone and the invading drone heading sphere, S directionRepresents the surface area of ​​the invading drone's heading sphere.

[0038] (b) Calculate the surface area of ​​the geometric configuration. The formula is derived as follows:

[0039]

[0040] λ1+λ2=R (10)

[0041]

[0042] r=d×tanθ (12)

[0043] S collision = = 2πdh (13)

[0044] h=d(1-cosθ) (14)

[0045] S direction =4πd 2 (15)

[0046]

[0047] Among them, X1, Y1, and Z1 represent the length, width, and height of the main drone, λ1 represents the radius of the sphere collision zone of the main drone, X2, Y2, and Z2 represent the length, width, and height of the invading drone, and λ2 represents the radius of the sphere collision zone of the invading drone. R Represents the radius of the combined sphere collision zone. d represents the height of the collision heading cone, θ represents the separation angle of the collision heading cone, and the bottom radius of the collision heading cone is r. collision S represents the surface area of ​​the spherical crown where the collision heading cone intersects the particle heading sphere. direction represents the surface area of ​​the heading sphere of the invading UAV, and ε represents the collision probability of the two UAVs.

[0048] Step 5: Calibrate the drone interval based on the safety level

[0049] The collision probability is set at 1% as the safety level, from which the safety interval of drones is inferred. When the collision probability is less than 1%, the drones are considered to be at a safe distance. When the collision probability is greater than 1%, the distance between drones is considered to be too close, posing a safety risk.

Claims

1. A method for calculating the collision probability of a UAV based on a heading cone, comprising the following steps: Step 1: Set the minimum circumscribed sphere of the drone as the drone collision zone. Step 2: When studying the collision process of two drones, a combined spherical collision zone is formed based on the collision zone radius of each drone and assigned to one of the drones, which is regarded as the invaded drone; Step 3: Perform kinematic analysis on the headings of the two drones to obtain the relative heading, and assign the relative heading to the other invading drone; Step 4: Set the collision rules. When the relative heading passes through the collision zone of the combined sphere, it is determined that a collision has occurred, and the collision probability of the drones is calculated using the collision heading cone; Step 5: Calibrate the drone interval based on the safety level.

2. The method for calculating the collision probability based on the heading cone according to claim 1, characterized in that: A mathematical model of the UAV sphere collision zone is established, and the collision zone of two UAVs is converted into a combined sphere collision zone based on the collision zone radius, including: The smallest circumscribed sphere of the drone body is used as the drone collision zone. The physical parameters of the sphere in the drone collision zone are as follows: Assuming that the horizontal length of the drone is x, the horizontal width is y, and the height is z, the radius λ, volume v, and surface area s of the spherical model can be obtained using mathematical formulas. By establishing the drone sphere collision zone, the drone collision problem can be converted to the position relationship between two spheres in three-dimensional space. In order to simplify the entire research process, the drone safety zone is transferred to one of the drones. The transferred safety zone is called the combined sphere collision zone. The drone in the combined sphere collision zone is regarded as the invaded drone, and the other drone is regarded as a particle as the invading drone: λ1+λ2=R (3) Among them, X1, Y1, and Z1 represent the length, width, and height of the invaded UAV, λ1 represents the radius of the spherical collision zone of the invaded UAV, X2, Y2, and Z2 represent the length, width, and height of the invading UAV, λ2 represents the radius of the spherical collision zone of the invading UAV, and R represents the radius of the combined spherical collision zone.

3. The method for calculating collision probability based on heading cone according to claim 1, characterized in that: Based on the collision rule setting of random headings, the headings of the two UAVs are synthesized into the heading vector of the invading UAV. The method of synthesizing the relative random vectors includes: Assuming that the heading of the drone is randomly and evenly distributed in three-dimensional space, when studying the collision problem of two drones, the headings of the two drones are all subject to random and uniform distribution in three-dimensional space. In order to improve the calculation efficiency, this paper intends to convert the headings of the two drones into relative headings. It can be considered that the heading of the invading drone obeys random and uniform distribution in three-dimensional space, and the invaded main drone is in a relatively static state. If the spherical collision zones of the two drones overlap, it can be determined that a collision has occurred between the two drones. Based on the concept of the combined spherical collision zone proposed in claim 2, it can also be determined that a collision has occurred between the two drones when the intruding drone enters the combined spherical collision zone. When the relative speed heading of the invading drone points to the combined sphere collision zone of the invaded drone, it is considered that the two drones are at risk of collision at this moment.

4. The method for calculating collision probability based on heading cone according to claim 1, characterized in that: The concept of collision heading cone is used to represent the relative heading set of all possible collisions in three-dimensional space. The collision probability is converted into the ratio of the crown area of ​​the heading ball of the invading drone to the surface area of ​​the heading ball in the heading cone, so as to quantify and evaluate the collision probability of two drones. This method includes: Probability problems stem from the uncertainty of events, and collision probability stems from the randomness of the drone's heading. Therefore, studying the probability of drone collisions is equivalent to studying the probability of the drone's heading when a drone collides: Among them, ε represents the probability of a UAV collision, l0 represents the set of headings where the UAV will collide, and l1 represents the set of all headings of the UAV. The number of elements in the l1 and l0 heading sets is too large to be counted using the traversal method, and it is necessary to consider the geometric configuration of the heading for equivalent substitution. Based on this, this paper proposes the concept of a collision heading cone, which is a ray from the invading drone to the collision zone of the combined sphere in three-dimensional space so that the ray is tangent to the collision zone of the combined sphere. The tangent ray is combined into a cone side. The cone surrounded by the cone side includes all relative heading sets that may collide in three-dimensional space. The headings of the invading drones are distributed spherically in three-dimensional space. The sphere contains the heading set of the invading drones. The drone collision probability can be regarded as the ratio between the geometric surface areas: Where S collision represents the surface area of ​​the intersection between the collision heading cone and the invading drone heading sphere, S direction Represents the surface area of ​​the invading drone's heading sphere. The collision probability is set to 1% as the UAV collision safety level, from which the UAV safety interval can be inferred. This paper provides a method for calculating the collision probability of two UAVs with randomly distributed headings. The collision zone and heading vector are combined respectively, and the concepts of heading sphere and heading cone of the invading UAV are proposed. The collision probability is quantified as the ratio between geometric surface areas, and the calculation of the instantaneous collision probability with strong timeliness is realized. Through this method, a reasonable collision probability reference can be provided for the safety risk assessment of UAVs and the determination of safety intervals, which will help in the future development of airspace performance and low-altitude management for large-scale UAV operations.

Images