Stereoscopic traffic three-dimensional space conflict detection and resolution method based on centralized control

Abstract

The invention discloses a three-dimensional traffic three-dimensional space conflict detection and resolution method based on centralized control, and relates to the technical field of three-dimensional traffic management and wireless communication technology crossing, and the method comprises the steps: obtaining a target operation region and the grid position of each target object in the target operation region; respectively establishing a cylinder protection area and a sphere protection area of each target object; according to the sphere protection area of each target object, performing three-dimensional conflict detection on the target object by using a conflict detection method based on a boundary volume method; according to the cylinder protection area of each target object, performing three-dimensional conflict detection on the target object by using a conflict detection method based on a boundary volume method; based on a fuzzy comprehensive evaluation method, determining a risk degree priority of the conflict target object; and based on a priority rule, carrying out conflict resolution by adopting a speed resolution method, a directional resolution method or a high resolution method. According to the invention, a plurality of target objects in the cluster can be effectively managed and controlled, and ordered movement in the cluster is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of stereoscopic traffic management and wireless communication technology, and in particular to a method for detecting and resolving stereoscopic traffic three-dimensional space conflicts based on centralized control. Background Art

[0002] The deep intelligence of three-dimensional transportation systems is a major future development trend, with increasingly broad applications in areas such as resource exploration, environmental monitoring, urban traffic management, and military applications. Autonomous navigation technology is one of the key technologies underlying the intelligent nature of three-dimensional transportation systems. Improving autonomous navigation capabilities will not only enhance operational efficiency and safety, but also enable various unmanned platforms to perform more complex tasks in dynamic and unknown environments. Future three-dimensional transportation networks will transcend the boundaries of traditional transportation networks, enabling the interconnection of diverse vehicles, including ground vehicles, drones, and flying cars. Conflict detection and resolution technologies will provide an integrated, safe, and efficient solution for the autonomous operation of unmanned platforms in complex traffic environments. With the gradual establishment of standards, testing, and verification systems for intelligent transportation vehicles, the development of autonomous navigation technology and industry has entered a fast lane. While the advancement of intelligent unmanned platforms has made operations more efficient and safer, it has also led to increasingly diverse and complex traffic flow characteristics. Furthermore, there are currently no clear rules and regulations for conflict resolution between unmanned platforms in autonomous navigation scenarios. This makes automatic collision avoidance between unmanned platforms a key technical issue that needs to be addressed in the autonomous navigation of future intelligent transportation vehicles. In particular, due to limitations in communication technology, there are significant delays in information exchange between unmanned platforms and their control centers. This directly affects the speed of response to environmental changes and the timeliness of decision-making, thereby increasing the probability of conflicts and accidents during operation. Therefore, resolving the conflict issue between multiple unmanned platforms in autonomous navigation scenarios is of great significance for preventing collisions, maintaining traffic order, reducing collision risks, and improving mission execution efficiency.

[0003] Currently, the primary goal of conflict detection and conflict resolution (CDCR) technology is to identify and resolve potential conflicts within automated transportation and navigation systems, ensuring safe and efficient traffic flow within these systems. With the advancement of unmanned platforms and autonomous navigation technology, existing algorithms for tracking and predicting multiple targets within a swarm are in urgent need of improvement to enhance the system's adaptability and autonomous collision avoidance capabilities. Effectively managing and controlling multiple unmanned platforms within a swarm to achieve collaborative operation and information sharing has become a key issue. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for detecting and resolving conflicts in three-dimensional traffic space based on centralized control, which can ensure orderly movement within the cluster.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] In a first aspect, the present invention provides a method for detecting and resolving three-dimensional traffic conflicts based on centralized control, comprising:

[0007] Step 101: obtaining a target operation area and a grid position of each target object in the target operation area; the target operation area is a grid-divided area; the grid division adopts cubic grid division.

[0008] Step 102: establishing protection zones for each target object at a grid scale; the protection zones include cylindrical protection zones and spherical protection zones.

[0009] Step 103: Based on the spherical protection zone of each target object, a collision detection method based on the boundary volume method is used to perform three-dimensional collision detection on the target object; when a collision is detected for the target object, step 104 is executed; when no collision is detected for the target object, it is determined that there is no collision between the target object and other target objects.

[0010] Step 104: Based on the cylindrical protection zones of the target objects, a collision detection method based on the boundary volume method is used to perform three-dimensional collision detection on the target objects; when a collision is detected for the target objects, step 105 is executed; when no collision is detected for the target objects, it is determined that there is no collision between the target objects and other target objects.

[0011] Step 105: Determine the risk priority of the conflicting target objects based on a fuzzy comprehensive evaluation method; the conflicting target objects are the target objects where conflicts are detected.

[0012] Step 106: Based on the priority rule, a conflict resolution method is used to resolve the conflict target object; the conflict resolution method includes a speed resolution method, a direction resolution method, or a height resolution method.

[0013] Optionally, before step 101, the method further includes:

[0014] The working area is divided into a plurality of cubic grids to obtain a target working area; each of the cubic grids includes a grid center point and 8 vertices.

[0015] Optionally, the expression for each position point in the cube grid is specifically:

[0016]

[0017] Among them, x0, y0, z0 are the x-axis, y-axis and z-axis coordinates of the origin of the coordinate system respectively; S is the side length of the cube grid; P C (X, Y, Z), P FTL (X, Y, Z), P FTR (X, Y, Z), P FBL (X, Y, Z), P FBR (X, Y, Z), P BTL (X, Y, Z), P BTR (X, Y, Z), P BBL (X, Y, Z), P BBR (X, Y, Z) represent the coordinates of the center point, front upper left point, front upper right point, front lower left point, front lower right point, back upper left point, back upper right point, back lower left point, and back lower right point of the current cube grid respectively.

[0018] Optionally, the cylindrical protection area is a minimum cylindrical area covering 8 small cylindrical protection areas; the small cylindrical protection area is a cylinder with a radius of r and a height of 2h established at the vertex of the cubic grid where the target object is located.

[0019] Optionally, the spherical protection zone is a minimum spherical area covering 8 small cylinder circumscribed spherical protection zones, and the small cylinder circumscribed spherical protection zones are protection zones established by the circumscribed spheres of the cylinders in the small cylinder protection zones.

[0020] Optionally, step 103 specifically includes:

[0021] It is determined whether the target object and its adjacent target objects are relatively stationary.

[0022] If the target object and its adjacent target objects are relatively stationary, the distance between the centers of the spherical protection zones of the target object and the adjacent target objects is determined according to the grid positions of the target object and its adjacent target objects.

[0023] According to the sphere center distance, based on the sphere radius of the sphere protection zone of the target object and the sphere radius of the sphere protection zone of the adjacent target object, it is determined whether the target object conflicts with the adjacent target object.

[0024] If the target object and its adjacent target objects are not relatively stationary, the distance between the centers of the spherical protection zones of the target object and the adjacent target objects is determined according to the grid positions of the target object and the adjacent target objects.

[0025] A first sphere center position function and a second sphere center position function of the target object and the adjacent target object are determined respectively according to the grid position and moving speed of the target object and the grid position and moving speed of the adjacent target object.

[0026] According to the first sphere center position function, the second sphere center position function and the sphere center distance, based on the sphere radius of the sphere protection zone of the target object and the sphere radius of the sphere protection zone of the adjacent target object, it is determined whether the target object conflicts with the adjacent target object.

[0027] Optionally, whether the target object conflicts with the adjacent target object is determined based on the first sphere center position function, the second sphere center position function, and the sphere center distance, and based on the sphere radius of the sphere protection zone of the target object and the sphere radius of the sphere protection zone of the adjacent target object, as follows:

[0028] According to the formula v 2 *t 2 +2vc*t+c 2 -r 2 =0, it is determined whether the target object conflicts with the adjacent target object.

[0029] When a solution exists for t, it is determined that the target object conflicts with the adjacent target object.

[0030] When there is no solution for t, it is determined that the target object does not conflict with the adjacent target object.

[0031] Where c is the distance from the center of the sphere, c = C A -C B , C A is the center position of the spherical protection zone of the target object, C B is the center position of the spherical protection zone of the adjacent target object; v = v a -v b , v is the speed difference; v a is the moving speed of the target object, v b is the moving speed of the adjacent target object; r=r A +r B ; r is the sum of the sphere radii, r A The radius of the spherical protection zone of the target object, r B is the spherical radius of the spherical protection zone of the adjacent target object; t is the time.

[0032] Optionally, step 104 specifically includes:

[0033] According to the cylindrical protection zone of the target object and the cylindrical protection zones of the target objects adjacent to the target object, a central axis S1 of the cylinder of the target object and a central axis S2 of the cylinder of the adjacent target object are determined.

[0034] A plane perpendicular to the central axis S1 of the cylinder is set, and an intersection point of the plane and the central axis S2 of the cylinder is calculated; the plane passes through at least one end point of the cylinder of the target object.

[0035] It is determined whether the intersection point is located on a line segment of the cylinder of the target object; the line segment is a line segment passing through the endpoints.

[0036] If the intersection point is located on a line segment, it is determined that the cylinders intersect in three-dimensional space.

[0037] If the intersection point is not located on the line segment, select different endpoints and repeat the steps: set a plane perpendicular to the central axis S1 of the cylinder, and calculate the intersection point of the plane and the central axis S2 of the cylinder.

[0038] After detecting all endpoints, if no intersection is found on the corresponding line segment, it is determined that there is no conflict between the target object and the adjacent target object.

[0039] Optionally, before executing step 106, the method further includes:

[0040] Determine the conflict type of the target object; the conflict type includes same-direction track passing conflict, intersecting track passing conflict and opposite-direction track passing conflict.

[0041] Optionally, step 106 specifically includes:

[0042] When the conflict type of the target object is a same-direction trajectory conflict, according to formula v Rx =v 1x +v 2x =v1cosθ1+v2cosθ2, to resolve the conflict; where v Rx is the relative velocity projected onto the x-axis, v Rx =v 1x -v 2x ; θ1 represents the angle between v1 and the line connecting the two targets, θ2 represents the angle between v2 and the line connecting the two targets; v1 is the velocity of the first target object; v 1x is the velocity of the first target object projected onto the x-axis; v2 is the velocity of the second target object; v 2x is the velocity of the second target object projected onto the x-axis.

[0043] When the conflict type of the target object is a cross-track passing conflict, the speed of the target object with the lower priority is adjusted according to the priority of the target object and the target object that conflicts with the target object.

[0044] When the conflict type of the target object is a reverse trajectory passing conflict, the height of the target object with the lower priority is adjusted according to the priority of the target object and the target object that conflicts with the target object.

[0045] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0046] The present invention provides a method for detecting and resolving conflicts in three-dimensional traffic flow based on centralized control. The method comprises the following steps: first, obtaining the target operation area and the location of each target object in a grid; then, establishing a protection zone for each target object at the grid scale; detecting conflicts within the spherical protection zone of the target object using a bounding volume method; and detecting conflicts within the cylindrical protection zone of the target object using a bounding volume method; determining the hazard priority of the conflicting target objects; and, based on priority rules, resolving the conflicting target objects using a speed resolution method, a direction resolution method, or a height resolution method. The present invention can effectively manage and control multiple target objects within a cluster, ensuring orderly movement. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 This is a flow chart of a method for detecting and resolving three-dimensional traffic conflicts based on centralized control provided in the first embodiment of the present invention.

[0049] FIG2 is a flowchart of conflict detection and conflict resolution according to the first embodiment of the present invention; FIG2(a) is a conflict detection diagram, and FIG2(b) is a conflict resolution diagram.

[0050] Figure 3 Schematic diagram of the cylindrical protection zone provided in Example 1 of the present invention.

[0051] Figure 4 A schematic diagram of the spherical protection zone provided in Example 1 of the present invention.

[0052] Figure 5 A schematic diagram of the conflict of the cylindrical protection zone of infinite height provided in the first embodiment of the present invention.

[0053] Figure 6 This is a schematic diagram of the terminal test provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0055] Under a centralized control platform, solving collision avoidance requires comprehensive consideration of multiple key factors, including perception, prediction, coordination, and emergency response. First, accurate environmental perception is required, using sensors such as radar and lidar to acquire environmental information and localization technology to determine the vehicle's own position. Second, based on this environmental perception data, the movement of other unmanned platforms and obstacles must be predicted to determine if there are any conflicts on the path issued by the centralized control platform. Furthermore, unmanned platforms regularly send their location information to the centralized control platform, which then checks for similarity with the predicted path and verifies any conflicts. Finally, if a conflict exists, the centralized control platform will issue a conflict resolution strategy, avoiding restricted areas, and recheck the new path for conflicts. In scenarios involving a large number of transportation platforms within a cluster, ensuring their coordination is crucial. For emergencies, such as sensor failure, emergency stop or avoidance strategies are necessary to mitigate potential risks.

[0056] A dead reckoning method based on the fusion of an unmanned platform's kinematic model and navigation information combines the kinematic model with navigation information collected from various sensors, such as Doppler odometers, inertial navigation systems, attitude sensors, and depth sensors. In complex environments with unknown dynamics, this method uses data fusion algorithms such as Kalman filtering to address the uncertainty and errors in the disparate sensor data, thereby estimating position and orientation. This approach aims to improve positioning accuracy in environments without high-precision satellite navigation. Its advantage lies in using the unmanned platform's kinematic model rather than its dynamic model, enabling simple and effective fusion of data from multiple sensor sources when environmental parameters are uncertain and vary significantly. It also reduces errors and drift in the presence of noise, accurately compensating for navigation sensor errors, and improving the accuracy and reliability of the dead reckoning system. Relying on inertial navigation provides independence from external signals and provides fast, real-time positioning and attitude information, providing reliable positioning information for subsequent conflict detection and resolution.

[0057] The primary goal of conflict detection and conflict resolution (CDCR) technology is to identify and resolve potential conflicts within automated traffic and navigation systems, ensuring safe and efficient traffic flow within these systems. With the advancement of unmanned platforms and autonomous navigation technology, existing algorithms for tracking and predicting multiple targets within a swarm are in urgent need of improvement to enhance the system's adaptability and autonomous collision avoidance capabilities. Effectively managing and controlling multiple unmanned platforms within a swarm to achieve collaborative operation and information sharing has become a key issue.

[0058] The purpose of the present invention is to provide a method for detecting and resolving conflicts in three-dimensional traffic space based on centralized control, which can ensure orderly movement within the cluster.

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0060] Example 1

[0061] like Figure 1 As shown, this embodiment provides a method for detecting and resolving three-dimensional traffic conflicts based on centralized control, including:

[0062] Step 101: obtaining a target operation area and a grid position of each target object in the target operation area; the target operation area is a grid-divided area; the grid division adopts cubic grid division.

[0063] Step 102: establishing protection zones for each target object at a grid scale; the protection zones include cylindrical protection zones and spherical protection zones.

[0064] Step 103: Based on the spherical protection zone of each target object, a collision detection method based on the boundary volume method is used to perform three-dimensional collision detection on the target object; when a collision is detected for the target object, step 104 is executed; when no collision is detected for the target object, it is determined that there is no collision between the target object and other target objects.

[0065] Step 104: Based on the cylindrical protection zones of the target objects, a collision detection method based on the boundary volume method is used to perform three-dimensional collision detection on the target objects; when a collision is detected for the target objects, step 105 is executed; when no collision is detected for the target objects, it is determined that there is no collision between the target objects and other target objects.

[0066] Step 105: Determine the risk priority of the conflicting target objects based on a fuzzy comprehensive evaluation method; the conflicting target objects are the target objects where conflicts are detected.

[0067] Step 106: Based on the priority rule, a conflict resolution method is used to resolve the conflict target object; the conflict resolution method includes a speed resolution method, a direction resolution method, or a height resolution method.

[0068] In some implementations of this embodiment, before executing step 101, the process further includes:

[0069] The working area is divided into a plurality of cubic grids to obtain a target working area; each of the cubic grids includes a grid center point and 8 vertices.

[0070] Specifically, the work area is divided into a series of neat cubic grids. For a three-dimensional grid, there are 9 main location points, which are 1 grid center point and 8 vertices. Assuming that the grid side length is S and the actual coordinates of the origin of the coordinate system are (x0, y0, z0), then the actual coordinates of the 9 main location points of the grid C (X, Y, Z) can be expressed as:

[0071]

[0072] Among them, x0, y0, z0 are the x-axis, y-axis and z-axis coordinates of the origin of the coordinate system respectively; S is the side length of the cube grid; P C (X, Y, Z), P FTL (X, Y, Z), P FTR (X, Y, Z), P FBL (X, Y, Z), P FBR (X, Y, Z), P BTL (X, Y, Z), P BTR (X, Y, Z), P BBL (X, Y, Z), P BBR (X, Y, Z) represent the coordinates of the center point, front upper left point, front upper right point, front lower left point, front lower right point, back upper left point, back upper right point, back lower left point, and back lower right point of the current cube grid respectively.

[0073] In this embodiment, when executing step 102, the specific steps may be as follows:

[0074] This embodiment will be described using an unmanned platform as the target object. Taking into account the communication delay and the error in prediction accuracy, when the unmanned platform passes through a certain grid area, since the location coordinates issued by the centralized control platform are all the center points of the grid, and the location of the unmanned platform at this time cannot be determined, the boundary of the grid where the unmanned platform is located is used as the center of the platform, and a cylinder with a radius of r and a height of 2h is expanded as a protection zone, namely the cylindrical protection zone; Figure 3 As shown, the cylindrical protection zone is the smallest cylindrical area covering 8 small cylindrical protection zones; the small cylindrical protection zone is a cylinder with a radius of r and a height of 2h established at the vertex of the cube grid where the target object is located.

[0075] Among them, the sphere protection area is as follows Figure 4 As shown, the spherical protection zone is the smallest spherical area covering the 8 small cylinder circumscribed sphere protection zones. The small cylinder circumscribed sphere protection zone is a protection zone established by the circumscribed sphere of the cylinder in the small cylinder protection zone.

[0076] In some implementations of this embodiment, when executing step 103, the following steps may be specifically included:

[0077] It is determined whether the target object and its adjacent target objects are relatively stationary.

[0078] If the target object and its adjacent target objects are relatively stationary, the distance between the centers of the spherical protection zones of the target object and the adjacent target objects is determined according to the grid positions of the target object and its adjacent target objects.

[0079] According to the sphere center distance, based on the sphere radius of the sphere protection zone of the target object and the sphere radius of the sphere protection zone of the adjacent target object, it is determined whether the target object conflicts with the adjacent target object.

[0080] If the target object and its adjacent target objects are not relatively stationary, the distance between the centers of the spherical protection zones of the target object and the adjacent target objects is determined according to the grid positions of the target object and the adjacent target objects.

[0081] A first sphere center position function and a second sphere center position function of the target object and the adjacent target object are determined respectively according to the grid position and moving speed of the target object and the grid position and moving speed of the adjacent target object.

[0082] According to the first sphere center position function, the second sphere center position function and the sphere center distance, based on the sphere radius of the sphere protection zone of the target object and the sphere radius of the sphere protection zone of the adjacent target object, it is determined whether the target object conflicts with the adjacent target object.

[0083] Specifically, this embodiment is illustrated using an unmanned platform as the target object. As shown in the flowchart in Figure 2(a), a new spherical protection zone is first created with the center point of the grid as the center of the sphere and a radius of R1 to ensure that the spherical protection zone can cover the cylindrical circumscribed sphere protection zone centered on the grid boundary and expanding outward. Three-dimensional conflict detection is first performed based on the circumscribed sphere protection zone.

[0084]

[0085] Assume that the initial position of the center of the sphere of unmanned platform A and unmanned platform B is C A and C B , radius r A and r B , uniform moving speed v a and v b When unmanned platform A and unmanned platform B are relatively stationary, determine whether the two balls collide: the distance between the two ball centers is less than or equal to the sum of the two radii, that is, the two intersect. The specific formula is as follows:

[0086] |C A -C B |≤r A +r B .

[0087] When there is a relative displacement between unmanned platform A and unmanned platform B, the positions of the centers of the two uniformly moving spheres are expressed as a function of time:

[0088]

[0089] The formula for determining whether the sphere bounding boxes intersect is converted to:

[0090]

[0091] At this time, let v = v a -v b , c=C A -C B , r=r A +r B Then we can get the formula: |vt+c|=r. The modulo symbol can be removed by squaring |vt+c| 2 =r 2 , we get v 2 *t 2 +2vc*t+c 2-r 2 = 0. Treat this formula as a quadratic equation with parameter t and solve it using the root-finding formula method. If a solution exists, a conflict has been detected; otherwise, no conflict has occurred and the next step is stopped.

[0092] That is, according to the formula v 2 *t 2 +2vc*t+c 2 -r 2 =0, it is determined whether the target object conflicts with the adjacent target object.

[0093] When a solution exists for t, it is determined that the target object conflicts with the adjacent target object.

[0094] When there is no solution for t, it is determined that the target object does not conflict with the adjacent target object.

[0095] Where c is the distance from the center of the sphere, c = C A -C B , C A is the center position of the spherical protection zone of the target object, C B is the center position of the spherical protection zone of the adjacent target object; v = v a -v b , v is the speed difference; v a is the moving speed of the target object, v b is the moving speed of the adjacent target object; r=r A +r B ; r is the sum of the sphere radii, r A The radius of the spherical protection zone of the target object, r B is the spherical radius of the spherical protection zone of the adjacent target object; t is the time.

[0096] In some implementations of this embodiment, when executing step 104, the specific steps may be as follows:

[0097] According to the cylindrical protection zone of the target object and the cylindrical protection zones of the target objects adjacent to the target object, a central axis S1 of the cylinder of the target object and a central axis S2 of the cylinder of the adjacent target object are determined.

[0098] A plane perpendicular to the central axis S1 of the cylinder is set, and an intersection point of the plane and the central axis S2 of the cylinder is calculated; the plane passes through at least one end point of the cylinder of the target object.

[0099] It is determined whether the intersection point is located on a line segment of the cylinder of the target object; the line segment is a line segment passing through the endpoints.

[0100] If the intersection point is located on a line segment, it is determined that the cylinders intersect in three-dimensional space.

[0101] If the intersection point is not located on the line segment, select different endpoints and repeat the steps: set a plane perpendicular to the central axis S1 of the cylinder, and calculate the intersection point of the plane and the central axis S2 of the cylinder.

[0102] After detecting all endpoints, if no intersection is found on the corresponding line segment, it is determined that there is no conflict between the target object and the adjacent target object.

[0103] Specifically, a cylindrical protection zone is established with the center point of the grid as the center of the new cylindrical protection zone, with a height of H and a radius of R2. The cylindrical protection zone surrounds all the small cylindrical protection zones at the grid boundary. 3D conflict detection is performed based on the cylindrical protection zone. Figure 4 As shown, the spherical protection zone is the smallest spherical area covering the 8 small cylinder circumscribed sphere protection zones. The small cylinder circumscribed sphere protection zone is a protection zone established by the circumscribed sphere of the cylinder in the small cylinder protection zone.

[0104] H=S+2h.

[0105]

[0106] Where S is the side length of the cube grid.

[0107] First, we detect the violation of the cylinder protection zone of infinite height, and use the Prucker coordinate system and dual vectors to represent these lines in space. Figure 5 As shown, S1 and S2 are the central axes of the two cylinders respectively. yes and The normal of the cylinder, vectors f, c, g, d represent points on the cylinder axis, which do not overlap. These points define the cylinder axis. Vectors s and w are the direction vectors of the cylinder axis, defined by vectors f, c, g, d. The solution formulas for S1 and S2 are lines expressed in Planck coordinates. The solution formula, The solution formula is a line represented by a double vector (for ease of calculation), using the dual vector representation, where ε 2 = 0. Vector n represents the common normal between the two cylinder axes, and p and q represent the intersection points of the common normal and the cylinder axes.

[0108]

[0109] If the dsinθ term is not equal to zero, then the lines do not intersect and are not parallel. If dsinθ is equal to zero and the cosθ term is not equal to 1, then the lines intersect and are not parallel. If the cosθ term of the dot product is equal to 1, then the lines are parallel and the resultant dual vector of the cross product will have a real component of 0. When the lines are identical, the dual component of the cross product will be 0. If the dcosθ term is nonzero, then the distance d can be calculated.

[0110] If the resulting distance is greater than the sum of the two radii, then a collision is unlikely, regardless of the height of the finite cylinder. If the result is not greater than the sum of the two radii, then a collision is possible, and a finite height cylinder protection zone model is used in the next stage of the collision detection algorithm.

[0111] Next, a parallel test is performed to determine a plane perpendicular to the centerline of the cylinders, which passes through at least one of the cylinder's endpoints. The intersection of the plane and the centerline of the other cylinder is calculated. A determination is made as to whether the intersection lies on the line segment defined by the cylinder's endpoints. If the intersection lies on the line segment, the cylinders are determined to intersect in three-dimensional space. If the intersection does not lie on the line segment, the above steps are repeated with different endpoints. If, after testing all possible endpoints, no intersection is found on the corresponding line segment, the cylinders are determined to not intersect in three-dimensional space, i.e., no collision has occurred.

[0112] In addition, a non-parallel test can be performed, requiring the angle θ between the axes to be acute. The shortest distance between two non-parallel lines is along their common normal. We first determine where the common normal intersects the axis of each cylinder. These axes are defined by their endpoints. and and non-unit direction vectors and To describe, and Equal to the height of their corresponding cylinders, line N intersects lines S1 and S2 at points and

[0113]

[0114] For each cylinder, determine the test point along its axis that is closest to the common normal, either inside or outside the finite cylinder. These test points are called TP1 and TP2. and Before or after the central axis section.

[0115] If t1≤0, then Located at the start of this segment or earlier, so the start of the cylinder is used as TP1. If t1 ≥ 1, then Located at the end of this segment or further, so this endpoint is used as TP1. If 0 < t1 < 1, then Located on the line segment, so Can be used as TP1. Repeat the above steps for cylinder 2 to determine TP2. From the determined points And Whether it is located on or outside the cylinder, there are three possible cases to consider. If 0 ≤ t i ≤ 1, where i = 1, 2, And Are both on the line segment, then the On-On test needs to be carried out; if And One of them but not both are on their respective central axis segments, then the On test is carried out; if And Are both not on the central axis segment, then the Off test needs to be carried out. These cases will be discussed below.

[0116] The first case, On-On test: If And Are both located on the central axis of the finite cylinder, then a conflict has occurred and no further inspection is required. Because in the infinite height cylinder detection, this distance has already been detected.

[0117] The second case, On test: First, find the point on the axis of cylinder 1 that is closest to TP2. This point P1 is the intersection of lines S1 and N1. Calculate t3 to determine whether P1 is on the cylinder or outside the cylinder. If the distance from TP2 to P1 is greater than the sum of the radii, then no conflict occurs. No further detection is required. Otherwise, find the point TP2' on cylinder 2 that is closest to the axis of cylinder 1, t3' and P1', where P1' is the point on the axis of cylinder 1 that is closest to TP2'. Calculate t3' generated by P1' to determine whether P1' is inside or outside the cylinder. If both P1 and P1' are outside cylinder 1, then there is no conflict and no further detection is required; if only P1 is on cylinder 1, then the ends of the cylinders may intersect and an end intersection detection needs to be carried out. If only P1' is on cylinder 1 or both P1 and P1' are on cylinder 1, then the distance is calculated from TP2' and P1'. If the distance is less than the radius of cylinder 1, then there is a conflict.

[0118] In the third case, the Off test, the same tests as in the On test are performed for each detection point. The distance from TP2 to S1 is calculated. If the distance is greater than the sum of the radii, there is no collision. If the distance is not greater than the sum of the radii, the process returns to the second case and performs the On test. This same test is also performed on the other end of cylinder 2. Finally, the cylinder data is swapped and the Off test is repeated using the new data.

[0119] End test: Figure 6 As shown, end-of-line testing is required when the circular bases of the cylinders may intersect. The first step is to find planes π1 and π2 perpendicular to each cylinder axis and find the parametric equation for the intersection of these two planes, N'. By setting the distance from TP1 to N' to r1, we obtain the equation for t4. Similarly, by setting the distance from TP2 to N' to r2, we can find t5.

[0120]

[0121] Among them, n x , n y , n z Represents the length of the projection of vector n on the x, y, and z axes, TP 1x It represents the length of the projection of TP1 on the x-axis. The positive direction of the x-axis is the direction from TP2 to TP2', and the z-axis is the direction of the central axis of cylinder 2.

[0122] If the quadratic equation produces complex roots, then the cylinder end circle and line N' do not intersect, and a collision is impossible. The multiple roots in the quadratic formula indicate that the cylinder end circle is tangent to line N', and there is no collision. If both roots are real, then t4 and t5 have a range, and if these ranges overlap, then a collision has occurred.

[0123] In some implementations of this embodiment, when executing step 105, the specific steps may be as follows:

[0124] The fuzzy comprehensive evaluation method is used to determine the traffic conflict risk ROC of unmanned platforms under three-dimensional traffic.

[0125] Taking the five influencing factors of unmanned platform speed ratio k, relative orientation θ, relative distance d, minimum encounter time TCPA and closest encounter distance DCPA as evaluation parameters, the fuzzy evaluation method is used to quantify the traffic conflict risk of unmanned platform, and the traffic conflict risk of unmanned platform under multi-factor conditions ROC is obtained. The factor set of target platform collision risk is:

[0126] u={k,θ,d,TCPA,DCPA}.

[0127] Through statistics and analysis of a large amount of data, it is found that the relationship between the weight values ​​of each influencing factor is as follows: Moreover, the weight values of TCPA and DCPA are stable between 0.3 and 0.4. Therefore, two random numbers between 0.3 and 0.4 are taken. The smaller value is the weight value of TCPA, and the larger value is the weight value of DCPA. The other three are randomly generated according to the size relationship, and finally, it is ensured that the sum of the weight values is 1.

[0128] Let X = V A / V B ≥0, W = 2, 0° < C < π. The expression of the membership function of the conflict risk degree of the speed ratio k of the unmanned platform is:

[0129]

[0130] In the formula, C—the collision angle.

[0131] The expression of the membership function of the conflict risk degree of the relative azimuth is:

[0132]

[0133] Among them, π ≤ θ ≤ 2π; 0° ≤ θ ≤ π.

[0134] The expression of the membership function of the conflict risk degree of the relative distance is as follows:

[0135]

[0136] In the formula, 0 ≤ D ≤ D1; D1 ≤ D ≤ D2; D2 < D; D1—the worst conflict resolution distance; D2—the distance at which conflict resolution can be carried out. The values of D1 and D2 are affected by the traffic area conditions, visibility conditions, and human factors, as follows:

[0137] D1 = K1·K2·K3·D L .

[0138] D2 = K1·K2·K3·R.

[0139] In the formula, K1 is affected by visibility; K2 depends on the complexity of the current area environment; K3 is determined by human factors; in the visual field of the autonomous passage of the unmanned platform in three-dimensional traffic, the values of K1, K2, and K3 are 1. DL is the latest trajectory change distance, generally taking 12 times the length of the avoidance platform; R is the corrected safe passing distance at any time.

[0140] The expression of the membership function of the conflict risk degree of TCPA is as follows:

[0141]

[0142] In the formula, 0 ≤ ∣TCPA∣ ≤ t1; d1 < ∣DCPA∣ ≤ d2; t2 < ∣TCPA∣.

[0143] The conflict risk membership function expression of DCPA is as follows:

[0144]

[0145] Where, d1 is the safe encounter distance of the unmanned platform; d2 is the absolutely safe encounter distance, defined as d2 = 2d1; |DCPA|≤d1; d1<|DCPA|≤d2; d2<|DCPA|.

[0146] Finally, calculate the collision risk ROC of the unmanned platform at any time:

[0147] ROC=w K u K +w θ u θ +w d u d +w TCPA u TCPA +w DCPA u DCPA .

[0148] Among them, w K is the weight of the factors affecting the speed ratio of the unmanned platform, w θ Relative orientation influencing factor weight value, w d Relative distance influencing factor weight value, w TCPA The weight value of the factors affecting the minimum encounter time and w DCPA The weight of the factor affecting the nearest encounter distance; u K is the unmanned platform speed ratio conflict risk membership, u θ Relative orientation conflict risk membership, u d Relative distance conflict risk membership, u TCPA Minimum encounter time conflict risk membership and u DCPA The nearest encounter distance conflict risk degree membership.

[0149] Establish the priority of unmanned platforms; that is, the higher the risk ROC, the higher the priority.

[0150] In some implementations of this embodiment, before executing step 106, the process further includes:

[0151] Determine the conflict type of the target object; the conflict type includes same-direction track passing conflict, intersecting track passing conflict and opposite-direction track passing conflict.

[0152] As shown in the flowchart in Figure 2(b), there are three types of conflicts at the same altitude level during the passage process: same-direction trajectory conflict, intersecting trajectory conflict, and opposite-direction trajectory conflict. Each conflict type has more than one resolution strategy. During the conflict resolution process, speed resolution is prioritized, followed by altitude resolution. Because changing trajectory direction increases energy consumption and costs, this strategy is considered last.

[0153] When the conflict type is a same-direction track conflict (0°~45°).

[0154] If the speed of the rear unmanned platform is greater than that of the front platform, that is, the rear platform will overtake the front platform, and the distance s ≥ l between the two unmanned platforms, the height can be changed to avoid traffic conflict.

[0155] If the distance s between the two unmanned platforms is less than the safety distance l, the direction is changed to avoid the traffic conflict. The release process only considers the speed and position relationship in the x-axis direction, then:

[0156] v Rx =v 1x +v 2x =v1cosθ1+v2cosθ2.

[0157]

[0158] When the conflict type of the target object is a cross-track conflict (45° to 135°), the speed of the target object with a higher priority is adjusted according to the priority of the target object and the target object that conflicts with the target object.

[0159] Specifically, if the unmanned platform speed range ( and ), then change the speed to avoid traffic conflicts. ⊥ Indicates the target given ascent / descent rate, v is the lower speed limit, The upper speed limit.

[0160] If it is not within the speed regulation range of the unmanned platform ( or ), then the high degree of release condition is changed to whether the following conditions are met:

[0161]

[0162] At this time, if s≥l, the height release strategy is adopted to release the conflict, otherwise the direction release strategy is adopted to release the conflict.

[0163] When the conflict type is reverse trajectory passing conflict (135° - 180°), according to the priorities of the target object and the target object that conflicts with the target object, the height of the target object with a lower priority is adjusted.

[0164] Specifically, if the distance s between two unmanned platforms is ≥ l and there is a reverse trajectory passing conflict, the height layer is first changed to avoid platform conflicts. The release process only considers the speed and position relationship in the x-axis direction, and the safety distance for height release is:

[0165]

[0166] If the distance s between two unmanned platforms is < l, a direction release strategy is adopted to avoid platform conflicts.

[0167] In summary, the beneficial effects of the present invention are as follows:

[0168] (1) The conflict detection method of the present invention based on the boundary volume method uses a bounding box to establish a protection area for unmanned platforms, and performs conflict detection from large to small at the grid scale. It not only considers the communication delay and prediction accuracy error conditions, but also saves a large amount of computing resources. In the case of conflicts in large grids, small grids are subdivided, the boundaries of the grids occupied by unmanned platforms are used as the limit offset positions of the platforms, the platform center is redefined, the protection area is expanded outward, and overlapping inspection is performed.

[0169] (2) The conflict detection method of the present invention based on the boundary volume method establishes a double protection area for unmanned platforms. First, a cylindrical protection area is established outside the unmanned platform, and then an external sphere protection area is established outside the cylindrical protection area for two-stage conflict detection. In the first stage, rough conflict detection is performed to quickly filter potential conflict objects in a large amount of data, significantly improving the speed of preliminary conflict detection. In the second stage, the selected conflicts are detected again in detail, further improving the detection accuracy and quality. This two-stage boundary volume method not only optimizes the application of computing resources, but also ensures high efficiency and high accuracy, and is especially suitable for dealing with targets with collaborative operations in a cluster in a complex dynamic traffic environment.

[0170] (3) The conflict resolution method of the present invention based on the velocity obstacle method introduces the concept of risk level and the priority division based on risk level, providing a practical and logically sound method for conflict resolution. By formulating clear rules, it is ensured that unmanned platforms with higher priorities can maintain or only make a small offset as much as possible, while platforms with lower priorities actively avoid, thus ensuring the orderly movement of unmanned platforms in the cluster and the coherence of task execution. This method not only improves the collaborative efficiency within the cluster, but also minimizes the delay and path deviation of task execution through a refined conflict resolution strategy, ensuring the high efficiency and safety of cluster operations.

[0171] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0172] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A method for detecting and resolving three-dimensional traffic conflicts based on centralized control, characterized in that: include: Step 101: Acquire a target operation area and a grid position of each target object in the target operation area; The target operation area is the area after grid division; The grid division adopts cubic grid division; Step 102: establishing protection zones for each target object at a grid scale; the protection zones include cylindrical protection zones and spherical protection zones; Step 103: performing three-dimensional collision detection on the target objects using a collision detection method based on a bounding volume method according to the spherical protection zones of the target objects; when a collision is detected for the target objects, executing step 104; when no collision is detected for the target objects, determining that no collision exists between the target objects and other target objects; Step 104: performing three-dimensional collision detection on the target objects using a collision detection method based on a boundary volume method according to the cylindrical protection zones of the target objects; When the target object detects a conflict, step 105 is executed; when the target object does not detect a conflict, it is determined that there is no conflict between the target object and other target objects; Step 105: Determine the risk priority of the conflict target object based on the fuzzy comprehensive evaluation method; The conflicting target objects are the target objects where conflicts are detected; Step 106: Based on the priority rule, a conflict resolution method is used to resolve the conflict target object; the conflict resolution method includes a speed resolution method, a direction resolution method, or a height resolution method.

2. The method for detecting and resolving three-dimensional traffic conflicts based on centralized control according to claim 1, characterized in that: Before step 101, the method further includes: The working area is divided into a plurality of cubic grids to obtain a target working area; each of the cubic grids includes a grid center point and 8 vertices.

3. The method for detecting and resolving three-dimensional traffic conflicts based on centralized control according to claim 2, characterized in that: The specific expression of each position point in the cube grid is: Among them, x0, y0, z0 are the x-axis, y-axis and z-axis coordinates of the origin of the coordinate system respectively; S is the side length of the cube grid; P C (X, Y, Z), P FTL (X, Y, Z), P FTR (X, Y, Z), P FBL (X, Y, Z), P FBR (X, Y, Z), P BTL (X, Y, Z), P BTR (X, Y, Z), P BBL (X, Y, Z), P BBR (X, Y, Z) represent the coordinates of the center point, front upper left point, front upper right point, front lower left point, front lower right point, back upper left point, back upper right point, back lower left point, and back lower right point of the current cube grid respectively.

4. The method for detecting and resolving three-dimensional traffic conflicts based on centralized control according to claim 2, characterized in that: The cylindrical protection area is the smallest cylindrical area covering 8 small cylindrical protection areas; the small cylindrical protection area is a cylinder with a radius of r and a height of 2h established at the vertex of the cube grid where the target object is located.

5. The method for detecting and resolving three-dimensional traffic conflicts based on centralized control according to claim 4 is characterized in that: The spherical protection zone is the smallest spherical area covering the 8 small cylinder circumscribed sphere protection zones. The small cylinder circumscribed sphere protection zone is a protection zone established by the circumscribed sphere of the cylinder in the small cylinder protection zone.

6. The method for detecting and resolving three-dimensional traffic conflicts based on centralized control according to claim 1, characterized in that: The step 103 specifically includes: Determining whether the target object and its adjacent target objects are relatively stationary; If the target object and its adjacent target objects are relatively stationary, determining the distance between the centers of the spherical protection zones of the target object and the adjacent target objects according to the grid positions of the target object and the adjacent target objects; Determining whether a conflict occurs between the target object and the adjacent target object based on the spherical center distance, the spherical radius of the spherical protection zone of the target object, and the spherical radius of the spherical protection zone of the adjacent target object; If the target object and its adjacent target objects are not relatively stationary, determining the distance between the centers of the spherical protection zones of the target object and the adjacent target objects according to the grid positions of the target object and the adjacent target objects; Determining a first sphere center position function and a second sphere center position function of the target object and the adjacent target object, respectively, according to the grid position and movement speed of the target object and the grid position and movement speed of the adjacent target object; According to the first sphere center position function, the second sphere center position function and the sphere center distance, based on the sphere radius of the sphere protection zone of the target object and the sphere radius of the sphere protection zone of the adjacent target object, it is determined whether the target object conflicts with the adjacent target object.

7. The method for detecting and resolving three-dimensional traffic conflicts based on centralized control according to claim 6, characterized in that: Determine whether the target object conflicts with the adjacent target object based on the first sphere center position function, the second sphere center position function, and the sphere center distance, and based on the sphere radius of the sphere protection zone of the target object and the sphere radius of the sphere protection zone of the adjacent target object, specifically as follows: According to the formula v 2 *t 2 +2vc*t+c 2 -r 2 =0, determine whether the target object conflicts with the adjacent target object; When there is a solution for t, determining that the target object conflicts with the adjacent target object; When there is no solution for t, determining that the target object does not conflict with the adjacent target object; Where c is the distance from the center of the sphere, c = C A -C B , C A is the center position of the spherical protection zone of the target object, C B is the center position of the spherical protection zone of the adjacent target object; v = v a -v b , v is the speed difference; v a is the moving speed of the target object, v b is the moving speed of the adjacent target object; r=r A +r B ; r is the sum of the sphere radii, r A The radius of the spherical protection zone of the target object, r B is the spherical radius of the spherical protection zone of the adjacent target object; t is the time.

8. The method for detecting and resolving three-dimensional traffic conflicts based on centralized control according to claim 1, characterized in that: The step 104 specifically includes: Determining a central axis S1 of the cylinder of the target object and a central axis S2 of the cylinder of the adjacent target object according to the cylindrical protection zone of the target object and the cylindrical protection zones of the adjacent target objects; Setting a plane perpendicular to the central axis S1 of the cylinder, and calculating the intersection of the plane and the central axis S2 of the cylinder; the plane passes through at least one end point of the cylinder of the target object; Determining whether the intersection point is located on a line segment of the cylinder of the target object; the line segment is a line segment passing through the endpoints; If the intersection point is located on a line segment, it is determined that the cylinders intersect in three-dimensional space; If the intersection point is not located on the line segment, repeat the steps by selecting different endpoints: setting a plane perpendicular to the central axis S1 of the cylinder, and calculating the intersection point of the plane with the central axis S2 of the cylinder; After detecting all endpoints, if no intersection is found on the corresponding line segment, it is determined that there is no conflict between the target object and the adjacent target object.

9. The method for detecting and resolving three-dimensional traffic conflicts based on centralized control according to claim 1, characterized in that: Before executing step 106, the method further includes: Determine the conflict type of the target object; the conflict type includes same-direction track passing conflict, intersecting track passing conflict and opposite-direction track passing conflict.

10. The method for detecting and resolving three-dimensional traffic conflicts based on centralized control according to claim 9, characterized in that: The step 106 specifically includes: When the conflict type of the target object is a same-direction trajectory conflict, according to formula v Rx =v 1x +v 2x =v1cosθ1+v2cosθ2, to resolve the conflict; where v Rx is the relative velocity projected onto the x-axis, v Rx =v 1x -v 2x ; θ1 represents the angle between v1 and the line connecting the two targets, θ2 represents the angle between v2 and the line connecting the two targets; v1 is the velocity of the first target object; v 1x is the velocity of the first target object projected onto the x-axis; v2 is the velocity of the second target object; v 2x is the velocity of the second target object projected onto the x-axis; When the conflict type of the target object is a cross-track conflict, adjusting the speed of the target object with the lower priority according to the priority of the target object and the target object that conflicts with the target object; When the conflict type of the target object is a reverse trajectory passing conflict, the height of the target object with the lower priority is adjusted according to the priority of the target object and the target object that conflicts with the target object.

Images