Unmanned aerial vehicle cluster formation planning method, device and medium based on point cloud matching

Abstract

This invention discloses a method, device, and medium for drone swarm formation planning based on point cloud matching. For any intelligent agent in a drone swarm, the method includes the following steps: converting the acquired drone trajectory into a point cloud of actual positions to be matched; matching the point cloud of actual positions to be matched with the point cloud of desired positions to obtain the optimal formation position sequence; setting a trajectory optimization function, using the optimal formation position sequence as a constraint, and solving the trajectory optimization function to obtain the optimal formation trajectory.

Description

Technical Field

[0001] This invention belongs to the field of UAV swarm formation trajectory planning for ultra-large scale, and particularly relates to a method, device and medium for UAV swarm formation planning based on point cloud matching. Background Technology

[0002] Formation planning is a fundamental and core capability of drone swarms. Despite widespread attention in academia and industry, current technologies still face significant challenges and limitations when scaling agents to "ultra-large-scale" and applying them to complex (including obstacle-laden) environments: 1. Currently, most mainstream large-scale drone formation performances rely on pre-designed trajectories, which cannot handle temporary changes in formation movement. Furthermore, due to a lack of effective perception, they cannot avoid sudden obstacles in a timely manner, resulting in stiff and slow graphic changes. While advanced fully autonomous distributed formation planning methods can achieve autonomous obstacle avoidance for small-scale drone swarms, the complexity of formation constraints increases exponentially with the size of the swarm, leading to computational resource overload and making it unsuitable for larger-scale formations.

[0003] 2. To achieve large-scale formation with limited computing resources, some existing technologies tend to simplify formation constraints and motion planning algorithms, decoupling most of the cooperative connections between agents. Among these, the Artificial Potential Field (APF) method uses attractive and repulsive forces to guide agent self-assembly. However, this method has slow convergence speed and lacks a clear obstacle avoidance mechanism, failing to meet the real-time planning requirements of complex environments. The Probabilistic Distribution method generates formations by adjusting the agent density distribution through real-time feedback. However, its convergence to the target configuration is extremely slow and highly dependent on an obstacle-free environment. The Polyhedral Structure method only constrains the swarm to maintain a uniform distribution within the polyhedron, similarly lacking adaptability in complex environments and easily leading to severe shape deformation.

[0004] 3. In ultra-large-scale, long-duration formations, due to unsatisfactory communication conditions or agent malfunctions, agent disconnection or suboptimal trajectory are frequent occurrences. Summary of the Invention

[0005] To address the shortcomings of existing technologies, embodiments of the present invention provide a method, device, and medium for planning drone swarm formation based on point cloud matching.

[0006] In a first aspect, embodiments of the present invention provide a method for planning drone swarm formation based on point cloud matching. For any agent in the drone swarm, the method includes the following steps: The acquired drone trajectory is converted into a point cloud of the actual location to be matched; The actual position point cloud to be matched is matched with the desired position point cloud to obtain the optimal formation position sequence; Set a trajectory optimization function, use the optimal formation position sequence as a constraint, and solve the trajectory optimization function to obtain the optimal formation trajectory.

[0007] In a second aspect, embodiments of the present invention provide an electronic device, comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores one or more computer programs that can be executed by the at least one processor, and the one or more computer programs are executed by the at least one processor to enable the at least one processor to perform the above-described point cloud matching-based UAV swarm formation planning method.

[0008] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-described method for planning unmanned aerial vehicle (UAV) swarms based on point cloud matching.

[0009] Fourthly, embodiments of the present invention provide a computer program product, including a computer program / instruction, which, when executed by a processor, implements the above-described method for planning unmanned aerial vehicle (UAV) swarm formations based on point cloud matching.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention provides a method for planning UAV swarm formation based on point cloud matching. By transforming the problem of calculating the optimal position of the formation sequence into a point cloud matching problem, the target position of the intelligent agent is calculated by directly solving the transformation parameters (such as rotation, translation, and scaling) between point clouds, thereby enabling real-time formation planning of intelligent agents on a scale of tens of thousands.

[0011] Furthermore, this invention employs a point cloud registration method that incorporates outlier removal capabilities (such as the RANSAC algorithm). In actual flight, this automatically identifies and filters out "abnormal" agents that deviate from the predetermined trajectory due to dynamic constraints, hardware failures, or control malfunctions. This mechanism fundamentally severs the propagation path of fault states within the cooperative network, ensuring that the anomalies of local individuals do not adversely affect the formation decisions of the entire cluster, thus guaranteeing the stability of overall formation performance.

[0012] Because this invention employs point cloud matching technology, it does not require the acquisition of the real-time positions of all agents. Even in the event of a small number of agent communication failures, data packet loss, or missing sensor observations, the algorithm can still utilize the remaining perceptible agent point cloud data to complete high-precision registration and position estimation. This characteristic of "local missing data not affecting global matching" enables the cluster to possess extremely strong operational survivability in complex electromagnetic environments or environments with large-scale obstruction, exhibiting strong communication fault tolerance and resilience, and adapting to non-ideal communication environments. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 A flowchart illustrating the UAV swarm formation planning method based on point cloud matching provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] It should be noted that, unless otherwise specified, the features in the following embodiments and implementation methods can be combined with each other.

[0017] like Figure 1 As shown, this embodiment of the invention provides a method for drone swarm formation planning based on point cloud matching. For any agent in the drone swarm, the method includes the following steps: Step S1: Convert the acquired drone trajectory into a spatiotemporal point cloud to be matched.

[0018] Specifically, step S1 includes: Consider a drone as an intelligent agent, and assume that there are N intelligent agents in a drone swarm. For the i-th agent, the i-th agent receives the other N... p The trajectory of N agents p <N; In timestamps {0,...,m,..., } for N p The trajectories of each agent are sampled uniformly over time to obtain N. p The set of spatiotemporal locations of an agent at each timestamp, denoted as . Where m represents the current timestamp m, This indicates the total number of time steps.

[0019] in, Indicates other N p The point cloud showing the actual location of an agent at time m. Indicates other N p An intelligent agent at time The actual location point cloud; , , This represents the position of the j-th agent at time m.

[0020] Step S2: Match the spatiotemporal point cloud to be matched with the point cloud at the desired location to obtain the optimal formation position sequence.

[0021] Specifically, step S2 includes the following steps: For other N p The actual location point cloud of the agent at time m Rather than the expected location point cloud Perform matching and solve for the optimal matching parameters; the expression is as follows: In the formula, This represents the desired position corresponding to the j-th agent. This represents the rotation parameters between the actual point cloud and the desired point cloud. This represents the translation parameter between the point cloud at the actual location and the point cloud at the desired location. This represents the scale transformation parameter between the actual location point cloud and the desired location point cloud.

[0022] The rotation parameters between the actual and desired point clouds are obtained by solving. The optimal translation parameters between the actual location point cloud and the desired location point cloud Scale transformation parameters between the actual location point cloud and the desired location point cloud , as the optimal matching parameter.

[0023] Based on the optimal matching parameters, the optimal formation position of the i-th agent at time m is obtained through coordinate transformation; the expression is as follows: In the formula, Let m represent the optimal formation position of the i-th agent at time m. This represents the desired position corresponding to the i-th agent.

[0024] By analogy, the optimal formation position of the i-th agent at each timestamp is obtained, resulting in the optimal formation position sequence corresponding to the i-th agent, denoted as... .

[0025] Furthermore, in this example, the RANSAC point cloud matching algorithm is used to match the actual position point cloud of the i-th agent at time m with its expected position point cloud, and the optimal matching parameters, including rotation parameters, translation parameters and scaling transformation parameters, are solved.

[0026] It should be noted that this invention does not use the optimal formation position corresponding to the i-th agent as the optimization variable to be directly optimized. Instead, it directly calculates the matching parameters by using the position points of other agents, and then uses the matching parameters to calculate the transformation of the expected formation point of the corresponding agent. This bypasses the complex optimization process of the optimal formation position and achieves efficient calculation of the optimal formation position.

[0027] It should be noted that, to address the limitation of traditional formation planning algorithms in adaptively handling anomalous agents, this example introduces a Randomized Randomized Matching Algorithm (RANSAC) with outlier removal to process anomalous trajectories in the trajectories of other agents received by the i-th agent that affect formation coordination. In this example, anomalous agents include failed agents, agents with abnormal positions, and agents with communication failures. Without an outlier removal algorithm, the abnormal position information of failed and abnormally positioned agents will affect the formation coordination of other individuals through the cooperative network, leading to a decrease in overall formation performance and potentially causing local optima in the trajectory optimization process of other agents. In this example, the RANSAC algorithm can remove anomalous individual positions, and even for ultra-large-scale cluster formation positions (greater than 1000 positions), the RANSAC algorithm can calculate the optimal matching parameters in real time (matching 1000 points takes less than 10ms). Furthermore, this example calculates the optimal formation position of the i-th agent at time m. At that time, only the data already collected by the i-th agent was used. The trajectory information of other agents does not affect the calculation of point cloud matching. Therefore, even if the trajectories of all agents are not collected, this instance can still obtain the optimal position of the sequence formation.

[0028] Step S3: Set the trajectory optimization function, and solve the trajectory optimization function to obtain the optimal formation trajectory, using the optimal formation position sequence as a constraint.

[0029] Specifically, step S3 includes the following steps: Construct the trajectory optimization function, with the following expression: In the formula, q represents the midpoint of the trajectory. t represents the time allocation of the trajectory. , Indicates the start time of the trajectory. Indicates the end time of the trajectory. Indicates the total time. This represents the time regularization parameter. This represents the trajectory to be optimized. Represents the trajectory characterized by q and t. Represents the trajectory to be optimized The s-th derivative with respect to time, where s denotes the s-th derivative. This represents the initial state of the trajectory. Indicates the terminal state of the trajectory. This includes dynamic constraints, obstacle avoidance constraints, and formation constraints.

[0030] The expression for the formation constraint is as follows: In the formula, This represents the position on the trajectory at time m. Let m represent the optimal formation position of the i-th agent at time m.

[0031] In summary, this invention provides a method for planning UAV swarm formation based on point cloud matching. By transforming the problem of calculating the optimal position of the formation sequence into a point cloud matching problem, the target position of the intelligent agent is calculated by directly solving the transformation parameters (such as rotation, translation, and scaling) between point clouds, thereby enabling real-time formation planning for thousands of intelligent agents.

[0032] According to embodiments of the present invention, the present invention also provides an electronic device and a readable storage medium.

[0033] Figure 2A schematic block diagram of an electronic device that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0034] The electronic device includes a computing unit 101, which can perform various appropriate actions and processes according to a computer program stored in ROM 102 or a computer program loaded into RAM 103 from storage unit 108. RAM 103 may also store various programs and data required for the operation of the electronic device. The computing unit 101, ROM 102, and RAM 103 are interconnected via bus 104. I / O interface 105 is also connected to bus 104.

[0035] Multiple components in the electronic device are connected to the I / O interface 105, including: an input unit 106, such as a keyboard, mouse, etc.; an output unit 107, such as various types of displays, speakers, etc.; a storage unit 108, such as a disk, optical disk, etc.; and a communication unit 109, such as a network card, modem, wireless transceiver, etc. The communication unit 109 allows the electronic device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0036] The computing unit 101 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 101 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 101 performs the various methods and processes described above. For example, in some embodiments, the methods in the multidimensional early warning system for pressure injuries can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 108. In some embodiments, part or all of the computer program can be loaded and / or installed on an electronic device via ROM 102 and / or communication unit 109. When the computer program is loaded into RAM 103 and executed by the computing unit 101, one or more steps of the methods in the multidimensional early warning system for pressure injuries described above can be performed. Alternatively, in other embodiments, the computing unit 101 can be configured to perform the methods in the multidimensional early warning system for pressure injuries by any other suitable means (e.g., by means of firmware).

[0037] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0038] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0039] In the context of this invention, a readable storage medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A readable storage medium can be a machine-readable signal medium or a machine-readable storage medium. A readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0040] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including voice input, speech input, or tactile input).

[0041] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0042] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0043] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only.

[0044] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for planning UAV swarm formation based on point cloud matching, characterized in that, For any agent in a drone swarm, the method includes the following steps: The acquired drone trajectory is converted into a point cloud of the actual location to be matched; The actual position point cloud to be matched is matched with the desired position point cloud to obtain the optimal formation position sequence; Set a trajectory optimization function, use the optimal formation position sequence as a constraint, and solve the trajectory optimization function to obtain the optimal formation trajectory.

2. The UAV swarm formation planning method based on point cloud matching according to claim 1, characterized in that, The process of converting a drone trajectory into a point cloud of actual locations to be matched includes: Assume there are N agents in the drone swarm; The i-th agent receives information from the other N... p The trajectory of N agents p <N; For N under timestamps {0,...,m,...Mc} p The trajectories of each agent are sampled uniformly over time to obtain N. p The set of spatiotemporal locations of an agent at various timestamps.

3. The UAV swarm formation planning method based on point cloud matching according to claim 2, characterized in that, The set of spatiotemporal location points is denoted as ;in, N represents p The point cloud showing the actual location of an agent at time m. N represents p An intelligent agent at time The actual location point cloud; N p The actual location point cloud of the agent at time m , , This represents the position of the j-th agent at time m.

4. The UAV swarm formation planning method based on point cloud matching according to claim 1, characterized in that, The process of matching the spatiotemporal point cloud to be matched with the point cloud at the desired location and calculating the optimal formation position sequence includes: For time m and the i-th agent, perform the following steps: For the other N trajectories received by the i-th agent... p The point cloud of the actual position of each agent at time m is matched with its expected position point cloud to solve for the optimal matching parameters; based on the optimal matching parameters, the optimal formation position of the i-th agent at time m is obtained through coordinate transformation. By iterating through each timestamp, we obtain the optimal formation position sequence corresponding to the i-th agent.

5. The UAV swarm formation planning method based on point cloud matching according to claim 4, characterized in that, For the i-th agent, the other N trajectories received p The process of matching the actual position point cloud of an agent at time m with its desired position point cloud and solving for the optimal matching parameters includes: The RANSAC point cloud matching algorithm is used to match the trajectory received by the i-th agent with the other N points. p The actual position point cloud of an agent at time m is matched with its expected position point cloud, and the optimal matching parameters, including rotation parameters, translation parameters, and scaling transformation parameters, are solved.

6. The UAV swarm formation planning method based on point cloud matching according to claim 1, characterized in that, The process of solving the trajectory optimization function to obtain the optimal formation trajectory, using the optimal formation position sequence as a constraint, includes: Construct a trajectory optimization function, which is used to minimize the integral of the square of the derivative of the trajectory and the weighted sum of the time allocation of each trajectory segment; Formation constraints are constructed based on the optimal formation position sequence. The trajectory optimization function is solved by combining initial state constraints, terminal state constraints, dynamic constraints, and obstacle avoidance constraints to obtain the optimal formation trajectory.

7. The UAV swarm formation planning method based on point cloud matching according to claim 6, characterized in that, The process of constructing formation constraints based on the optimal formation position sequence includes: The formation constraint is obtained by summing the distances between the position on the trajectory at the current time and the optimal formation position of the i-th agent at the current time for each timestamp.

8. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores one or more computer programs that can be executed by the at least one processor, the one or more of the computer programs being executed by the at least one processor to enable the at least one processor to perform the UAV swarm formation planning method based on point cloud matching as described in any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the UAV swarm formation planning method based on point cloud matching as described in any one of claims 1-7.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the UAV swarm formation planning method based on point cloud matching as described in any one of claims 1-7.

Images