A UAV formation collision avoidance method based on artificial potential field

By constructing an artificial potential field model and calculating the relative distance and pitch angle between UAVs, and assigning collision avoidance commands, the problems of large computational load and unsatisfactory collision avoidance effect in large-scale UAV formations are solved, achieving a more efficient collision avoidance effect.

CN115291627BActive Publication Date: 2024-11-29HARBIN ENG UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202210943702.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-08-08
Publication Date
2024-11-29
Estimated Expiration
2042-08-08

AI Technical Summary

Technical Problem

Existing collision avoidance methods for UAV formations based on artificial potential fields are computationally intensive and have unsatisfactory collision avoidance effects in large-scale formations, especially in the horizontal and vertical directions.

Method used

By constructing an artificial potential field model, the relative distance and relative pitch angle between UAVs are calculated. Using the relative projection value, pitch angle and judgment conditions, collision avoidance commands are assigned, including adjusting the UAV's heading angle and overload in the horizontal and vertical directions, reducing the number of times the artificial potential field is triggered and improving the collision avoidance effect.

Benefits of technology

It improves the collision avoidance effect of drone formations, reduces the amount of computation, and increases the flexibility and accuracy of collision avoidance. In particular, it can avoid collisions in time when the drone is moving forward, avoiding the need for a single-direction collision avoidance strategy and improving the collision avoidance effect.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115291627B_ABST
    Figure CN115291627B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for avoiding collision of unmanned aerial vehicle formation based on artificial potential field, which specifically includes the following steps: step 1, artificial potential field model, the artificial potential field is divided into an anti-collision area and a collision area, when another unmanned aerial vehicle enters the anti-collision area, the artificial potential field starts to calculate the corresponding artificial force, when another unmanned aerial vehicle enters the collision area, it is considered that the unmanned aerial vehicle has collided; step 2, calculate the relative distance and relative angle between unmanned aerial vehicles; step 3, calculate the artificial force, and execute the anti-collision scheme for the unmanned aerial vehicle entering the anti-collision area. The present invention improves the existing anti-collision scheme based on artificial potential field, by calculating the relative projection between unmanned aerial vehicles, realizing the anti-collision step when there is only a unmanned aerial vehicle in the forward direction of the speed, reducing the number of triggering times of the artificial potential field, using the relative pitch angle between unmanned aerial vehicles, realizing the anti-collision instruction distribution, no longer single selection of height anti-collision or heading angle anti-collision, further improving the anti-collision effect.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The invention belongs to the technical field of unmanned aerial vehicles, and in particular relates to an unmanned aerial vehicle formation anti-collision method based on an artificial potential field. Background Art

[0002] UAV formation flight aims to arrange a certain number of UAVs in a certain formation and keep the formation unchanged during the flight. UAV collision avoidance is the premise of safe flight of UAV formation. The uncertainty of UAV position during the process of UAV assembly and formation change leads to the risk of mutual collision between UAVs. Existing UAV collision avoidance is mainly achieved through path planning, artificial potential field method and geometric method. For large-scale UAV formation, it is very difficult and lacks flexibility to carry out path planning in advance. Geometric method and artificial potential field method can be implemented online, but the geometric method requires a large amount of calculation. Therefore, artificial potential field is more widely used in large-scale UAV formation. Existing UAV collision avoidance methods based on artificial potential field are mainly divided into height direction and horizontal direction collision avoidance, and the collision avoidance effect is not ideal.

[0003] In combination with the existing technical achievements, this application proposes a UAV collision avoidance method based on artificial potential field. Summary of the invention

[0004] The purpose of the present invention is to provide a UAV formation collision avoidance method based on artificial potential field to solve the problems raised in the above background technology.

[0005] A UAV formation collision avoidance method based on artificial potential field specifically comprises the following steps:

[0006] Step 1: Construct an artificial potential field model. The artificial potential field is divided into an anti-collision area and a collision area. When another drone enters the anti-collision area, the artificial potential field starts to calculate the corresponding artificial force. When another drone enters the collision area, it is considered that the drones collide.

[0007] Step 2: Calculate the relative distance and angle between drones:

[0008]

[0009]

[0010]

[0011] where x i ,y i ,z i is the coordinate of UAV i in the inertial coordinate system, v x ,v y ,v zis the speed of drone i in the inertial coordinate system, dxyz is the relative distance matrix between drones, yaw is the relative pitch angle matrix between drones, indicating the relative pitch angle between drone i and drone j, and projection is the projection matrix of drone j in the speed direction of drone i;

[0012] Step 3: Calculate the artificial force. For the drone entering the anti-collision area, perform the following calculation steps:

[0013] S1: Calculate the sum of each row of yaw and record it as yaw sum (i);

[0014] S2: Determine whether the location information of each drone meets the condition (dxyz(i,j)<r d )&&(projection(i,j)<0)do;

[0015] S3: For UAVs that meet the conditions, use the corresponding artificial potential field function to calculate the artificial force U x (i),U y (i),U z (i);

[0016] S4: Determine whether the drone that meets the conditions in the second step meets the following conditions: yaw sum (i)≥π / 4 The output control command u of the drone that meets the condition z (i) = sum(U z ), output control instruction u that does not meet the conditions xy (i) = atan(sum(U y ) / sum(U x ));

[0017] The input of the above algorithm is the relative distance dxyz between drones, the relative pitch angle dxyz, and the projection value projection; the output u z (i) is the overload instruction of the i-th UAV in the vertical direction, output u xy (i) is the heading angle command of the i-th UAV in the horizontal direction.

[0018] Compared with the prior art, the beneficial effects of the present invention are: improving the existing anti-collision scheme based on artificial potential field, by calculating the relative projection between drones, realizing the start of the anti-collision step only when there is a drone in the forward direction, reducing the number of triggering of the artificial potential field, and utilizing the relative pitch angle between drones to realize the allocation of anti-collision instructions, no longer single selection of height anti-collision or heading angle anti-collision, further improving the anti-collision effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the artificial potential field of the present invention;

[0020] Figure 2 This is a relative projection schematic diagram of the UAV of the present invention;

[0021] Figure 3 is the experimental initial condition data table;

[0022] Figure 4 It is the distance diagram between unmanned aircrafts of the present invention;

[0023] Figure 5 This is a formation flight result diagram of the present invention;

[0024] Figure 6 Pseudocode operation for calculating artificial force for the present invention. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0026] The present invention provides a UAV formation collision avoidance method based on artificial potential field, which specifically comprises the following steps:

[0027] Step 1: Construct an artificial potential field model. The artificial potential field defines a spatial area. UAVs entering this area will be affected by an abstract artificial force field. The effect of the force field on the UAV is reflected by the repulsive artificial force applied to the UAV. The repulsive artificial force is obtained by calculating the gradient of the potential function of the artificial potential field. The artificial potential field is divided into an anti-collision area and a collision area. When another UAV enters the anti-collision area, the artificial potential field starts to calculate the corresponding artificial force. When another UAV enters the collision area, it is considered that the UAV has collided.

[0028] Step 2: Calculate the relative distance and angle between drones:

[0029]

[0030]

[0031]

[0032] where x i ,y i ,z i is the coordinate of UAV i in the inertial coordinate system, v x ,v y ,v zis the speed of drone i in the inertial coordinate system, dxyz is the relative distance matrix between drones, yaw is the relative pitch angle matrix between drones, indicating the relative pitch angle between drone i and drone j, and projection is the projection matrix of drone j in the speed direction of drone i;

[0033] Step 3: Calculate the artificial force. For the drone entering the anti-collision area, perform the following calculation steps (the pseudo code of this step is as attached). Figure 6 shown):

[0034] S1: Calculate the sum of each row of yaw and record it as yaw sum (i);

[0035] S2: Determine whether the location information of each drone meets the condition (dxyz(i,j)<r d )&&(projection(i,j)<0)do;

[0036] S3: For UAVs that meet the conditions, use the corresponding artificial potential field function to calculate the artificial force U x (i),U y (i),U z (i);

[0037] S4: Determine whether the drone that meets the conditions in the second step meets the following conditions: yaw sum (i)≥π / 4 The output control command u of the drone that meets the condition z (i) = sum(U z ), output control instruction u that does not meet the conditions xy (i) = atan(sum(U y ) / sum(U x ));

[0038] The input of the above algorithm is the relative distance dxyz between drones, the relative pitch angle dxyz, and the projection value projection; the output u z (i) is the overload instruction of the i-th UAV in the vertical direction, output u xy (i) is the heading angle command of the i-th UAV in the horizontal direction.

[0039] In the above algorithm, the first step calculates the sum of the relative pitch angles, the second step starts to calculate the artificial force between UAV j and UAV i, and the third step gives the judgment conditions. The first judgment condition compares the distance between UAV i and UAV j, and the second judgment condition compares the projection of the position vector of UAV ij in the direction of the speed of UAV i. When the projection value is less than zero, UAV j is in the space in the direction of the positive semi-axis of the x-axis of the speed coordinate system of UAV i. When the conditions are met at the same time, it is considered that UAV j is in front of UAV i and the distance is less than the anti-collision distance. At this time, UAV i calculates the artificial force; the sixth step determines the sum of the relative pitch angles of UAV i and UAV j. When the sum of the relative pitch angles is greater than 45 degrees, anti-collision is achieved by adjusting the height. When the sum of the relative pitch angles is less than 45 degrees, anti-collision is achieved by adjusting the heading angle. The command heading angle is calculated according to step 9.

[0040] The following verification test is performed for the above-mentioned embodiment:

[0041] 1. Experimental Preparation

[0042] The experimental artificial potential field function is as follows:

[0043]

[0044] Where p represents the position vector of the drone [x i y i z i ]、r d =10m,r c =5m.

[0045] The kinematic model of the experimental formation UAV is as follows: (5)

[0046]

[0047] In formula (5), v xi , v yi , v zi Respectively represent the speed of the drone in the X, Y, and Z directions, a xi , a xi , a zi Represents the acceleration of the drone along the X, Y, and Z directions respectively.

[0048] The following control law is used to achieve formation stability

[0049]

[0050] In formula (6), u xi ,u yi ,u zi is the acceleration command of the drone in the x, y, and z directions, a ijRepresents the corresponding element in the topological matrix of the inter-UAV communication topology. If individual i can receive information from individual j, then a ij =1,i,j=1,2,L,n,i≠j, otherwise a ij =0, R x ,R y .R z is the expected distance between drones. The expected drone formation is arranged along the y-axis with a distance of 10m between adjacent drones, an expected speed of 10m / s, an expected heading angle of 0 degrees, and an expected height of 30m.

[0051] 2. Combining the invention content with the artificial potential field function to conduct experiments, the initial conditions of the experiment are shown in the attached Figure 3 :

[0052] 3. Experiment duration: 20s, experimental step length: 0.01s. Figure 4 , Figure 5 The distance between drones and their flight trajectories are given respectively. Figure 4 The horizontal dashed line in the middle represents r c =5m, which is the collision distance. It can be seen that the distance between drones is greater than the collision distance, which verifies the effectiveness of the anti-collision algorithm proposed in this patent. Figure 4 Figure 5 It can be seen that the distance between drones and the flight trajectory quickly reach the expected state, demonstrating that the anti-collision algorithm proposed in the present invention has little impact on the formation.

[0053] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A UAV formation collision avoidance method based on artificial potential field, Features: The specific steps include: Step 1: Construct an artificial potential field model. The artificial potential field is divided into an anti-collision area and a collision area. When another drone enters the anti-collision area, the artificial potential field starts to calculate the corresponding artificial force. When another drone enters the collision area, it is considered that the drones collide. Step 2: Calculate the relative distance and angle between drones: where x i ,y i ,z i is the coordinate of UAV i in the inertial coordinate system, v x ,v y ,v z is the speed of drone i in the inertial coordinate system, dxyz is the relative distance matrix between drones, yaw is the relative pitch angle matrix between drones, indicating the relative pitch angle between drone i and drone j, and projection is the projection matrix of drone j in the speed direction of drone i; Step 3: Calculate the artificial force. For the drone entering the anti-collision area, perform the following calculation steps: S1: Calculate the sum of each row of yaw and record it as yaw sum (i); S2: Determine whether the location information of each drone meets the conditions; (dxyz(i,j)<r d )&&(projection(i,j)<0); S3: For UAVs that meet the conditions, use the corresponding artificial potential field function to calculate the artificial force U x (i),U y (i),U z (i); S4: Determine whether the drone that meets the conditions in the second step meets the following conditions: yaw sum (i)≥π / 4 The output control command u of the drone that meets the condition z (i) = sum(U z ), output control instruction u that does not meet the conditions xy (i) = atan(sum(U y ) / sum(U x )); The inputs in the above calculation steps are the relative distance dxyz between drones, the relative pitch angle dxyz, and the projection value projection, and the output u z (i) is the overload instruction of the i-th UAV in the vertical direction, output u xy (i) is the heading angle command of the i-th UAV in the horizontal direction.

Citation Information

Patent Citations

  • Multi-area classified unmanned aerial vehicle formation route planning method based on distributed control

    CN109508035A

  • Unmanned aerial vehicle formation based on consistency algorithm and artificial potential field method and obstacle avoidance control method thereof

    CN111290429A