AI-based unmanned aerial vehicle autonomous obstacle avoidance path planning method and system

Abstract

The application discloses an unmanned aerial vehicle autonomous obstacle avoidance path planning method and system based on AI, which comprises collecting path data and related data of a target area, preprocessing the path data and the related data; performing abnormal screening on the path data according to a traffic threshold to obtain abnormal point information, performing obstacle passing capacity analysis on the abnormal point information to obtain a first parameter variable; performing variable influence analysis on the related data to obtain a second parameter variable, and constructing a joint function according to the first parameter variable and the second parameter variable; constructing a target path optimization model according to the joint function, inputting to-be-optimized data into the target path optimization model, and outputting an optimization result. The method improves the precision of joint optimization of searching mode and flight path, has good interpretability, and can be directly applied to a joint optimization system of loading mode and flight path.

Description

Technical Field

[0001] This invention relates to the field of path optimization, and in particular to an AI-based method and system for autonomous obstacle avoidance path planning for unmanned aerial vehicles (UAVs). Background Technology

[0002] In flight search and control, efficient search methods and rational flight path planning are crucial for reducing costs, minimizing losses, and improving flight efficiency. Traditionally, search methods and flight paths are often optimized separately, neglecting their interrelationships, which makes it difficult to achieve optimal overall flight efficiency. Furthermore, flight is affected by various factors such as weather conditions, road conditions, and airspace traffic. Changes in these factors can lead to anomalies during flight, such as loss or damage to flight signals, further increasing flight costs and management complexity.

[0003] In recent years, the development of big data technology and intelligent optimization algorithms has provided new ideas for the joint optimization of search methods and flight paths. Genetic algorithms, as search algorithms that simulate natural evolution, have been widely used in the optimization of complex systems due to their strong global search capabilities and robustness. However, there is currently no mature AI-based autonomous obstacle avoidance path planning method for UAVs that can comprehensively consider multiple influencing factors and achieve coordinated optimization of loading methods and flight paths. Therefore, it is necessary to improve the AI-based autonomous obstacle avoidance path planning method for UAVs to enhance the accuracy, precision, and real-time performance of joint optimization. Summary of the Invention

[0004] The purpose of this invention is to provide an AI-based method and system for autonomous obstacle avoidance path planning for unmanned aerial vehicles (UAVs).

[0005] To achieve the above objectives, the present invention is implemented according to the following technical solution:

[0006] This invention includes the following steps:

[0007] Collect path data and related data for the target area, and preprocess the path data and related data;

[0008] The path data is filtered for anomalies based on the passage threshold to obtain anomaly point information, and the obstacle passage capability of the anomaly point information is analyzed to obtain the first parameter variable.

[0009] A variable influence analysis is performed on the relevant data to obtain the second parameter, and a joint function is constructed based on the first parameter and the second parameter.

[0010] Construct a target path optimization model based on the joint function, input the data to be optimized into the target path optimization model, and output the optimization results.

[0011] Furthermore, the method for obtaining anomaly point information by anomaly filtering of the path data based on a passage threshold includes:

[0012] The time points where path data shows dwell times and flight speeds below 30 km / h are designated as problem data points. The information weight of these problem data points is calculated based on their data characteristics and importance.

[0013]

[0014] The i-th path data of the u-th problem data point is u. i The path data of the y-th problem data point is u. y Path data u i The probability of occurrence is Path data u y The probability of occurrence is Path data u i and path data u y The probability of them occurring simultaneously is Observed path data u y Time path data u i The posterior probability is The number of path data is m, and the number of path data is u. i Information weight is

[0015] Problem data points with an information weight greater than 0.412 are designated as key data points, and optimization is performed on these key data points:

[0016] Calculate particle fitness:

[0017]

[0018] The probability parameter is chosen as sp, and the position of particle x is D. x Position D x Information weight is The fitness function of particle x is

[0019] We use roulette wheel selection to choose superior particles as the parents for crossover mutations, and calculate the selection probability:

[0020]

[0021] The z-th particle is x z Particle x z fitness is The number of particles is Particle x z The probability of selection is Particle x z The position is D(x)z );

[0022] Perform crossover and mutation on the particles, and calculate the crossover rate and mutation rate:

[0023]

[0024] The crossover rate is The rate of variation The parameters between 0 and 1 are β1, β2, β3, and β4, respectively, and the average fitness of the particle population is... The maximum fitness of the particle population is The fitness of the crossover individual is The fitness of the variant individual is

[0025] The gene segments of the particles after crossover mutation are reverse-sorted to obtain new particles. The fitness of the new particles is calculated, and the passage threshold is calculated based on the fitness.

[0026]

[0027] Where the passage threshold is η, and the new particle x new fitness is The flight speed of the u-th problem data point is b u The minimum flight speed of the u-th data point is b. min ;

[0028] The process iterates until the maximum number of iterations is reached, and then outputs key data points with a passage threshold greater than 0.374 as outlier information.

[0029] Furthermore, the method for obtaining the first parameter variable by obstacle capability analysis of the anomaly point information includes:

[0030] Input anomaly information into the obstacle capability analysis model, and calculate airspace traffic based on the anomaly information:

[0031] p = a·f

[0032] a=(k) -1

[0033] Where airspace traffic is p, congestion density is a, and the spacing between UAVs is k;

[0034] Calculate the relationship between drone traffic flow and speed:

[0035]

[0036] The relationship between drone flow and speed is ρ, the function curvature control parameter is μ, and the maximum density causing airspace congestion is a. maxThe maximum speed of the drone is f. v ;

[0037] Calculate the spacing between drones:

[0038]

[0039] The length of the drone fuselage is l, and the minimum safe distance of the drone is s. o The drone's flight speed is f, and the free flow velocity is f. o The safe head-to-head distance for manually piloted drones is H. o ;

[0040] Based on the spacing between UAVs and airspace traffic, the basic graph model analysis function for airspace flow is given:

[0041]

[0042] Anomaly information is analyzed using analytical functions to obtain airspace traffic and airspace congestion density. The first parameter is obtained by weighted summation of airspace traffic and airspace congestion density.

[0043] Furthermore, a method for obtaining a second parameter by performing variable influence analysis on the relevant data includes:

[0044] Construct a state matrix based on relevant data, using the time points of the relevant data as nodes, and calculate the contribution of each node:

[0045]

[0046] The contribution of the t-th node and the a-th related data is ψ. a (t), the flight score of the a-th related data at the t-th node is The flight score of the a-th relevant data at the (t-1)th node is The number of related data for the t-th node is The regulation coefficient of the a-th relevant data is γ a The a-th related data of the t-th node is h. a (t), where h is the a-th related data of the (t-1)-th node. a (t-1), the degree of influence is σ, and the variable of the a-th related data of the t-th node is Δη. a,t The flight score index value of the w-th related data of the t-th node is V. a,w (t), the number of flight score indicators is N, and the weight of the w-th flight score indicator is λ. w ;

[0047] Calculate the average probability of the node:

[0048]

[0049] The average probability of the a-th related data at the t-th node is: The number of nodes containing the a-th related data is M, and the probability of the a-th related data appearing in the t-th node is ζ. a (t);

[0050] Construct a weighted directed graph based on the directed weighted network, and calculate the importance index of the nodes:

[0051]

[0052] The importance index of the t-th node is ε. u The natural constant is e, and the weight of the connection sent out by the t-th node is C. ot (u), where the weight of the connection received by the t-th node is C. in (u), with an influence coefficient of υ;

[0053] Compute the weighted value of the nodes:

[0054]

[0055] The weighted value of the t-th node is: The probability that the a-th related data point at the t-th node is in an influenced state is ζ. a,yx (t), with a weighting factor of λ;

[0056] Calculate the second parameter:

[0057]

[0058] The flight score of the t-th node is The standard value of the a-th relevant data is h. ao The standard value for flight score is The second parameter is

[0059] Furthermore, the method for constructing a joint function based on the first parameter and the second parameter includes:

[0060] Priority data is obtained by filtering based on search methods and flight paths, and flight evaluation indicators are obtained based on the priority data.

[0061] Based on the construction of the joint function, the expression is:

[0062]

[0063] The second parameter is The first parameter is The optimization coefficient is The actual arrival time is T st Agreed arrival time The number of the nth type of unmanned aerial vehicle is B. n The energy consumption per kilometer of the nth type of unmanned aerial vehicle is g. n The flight speed and range of the nth type of unmanned aerial vehicle is S. n The sign function is sgn(·), and the fixed cost of flight is R. gd The joint function is

[0064] Furthermore, the method for constructing a target path optimization model based on the joint function includes:

[0065] The objective function is constructed based on the joint function, and its expression is:

[0066]

[0067] Where the control factor is The loss function is The objective function is

[0068] Target path optimization models include particle swarm optimization algorithm, random forest algorithm, and recurrent neural network;

[0069] Particle swarm optimization (PSO) algorithms adjust the search direction and position by sharing information and accumulating individual experience among particles, adapting to the joint function requirements of flight, and achieving efficient search and optimization in a multi-dimensional feature space to obtain improved flight schemes.

[0070] The random forest algorithm constructs multiple decision trees and obtains prediction results based on improved flight schemes, captures the nonlinear relationships of the prediction results, and extracts effective features based on the nonlinear relationships;

[0071] Recurrent neural networks update their internal state at sequential time steps through a recursive loop structure, capturing the dynamic features of the data to be optimized, and predicting the joint function value based on the effective features.

[0072] Secondly, embodiments of this application also provide an electronic device, including:

[0073] A processor; and a memory arranged to store computer-executable instructions, which, when executed, cause the processor to perform the steps of the method described in the first aspect.

[0074] Thirdly, embodiments of this application also provide a computer-readable storage medium storing one or more programs that, when executed by an electronic device including multiple applications, cause the electronic device to perform the steps of the method described in the first aspect.

[0075] The beneficial effects of this invention are:

[0076] This invention relates to an AI-based autonomous obstacle avoidance path planning method and system for drones. Compared with existing technologies, this invention has the following technical advantages:

[0077] This invention, through preprocessing, anomaly screening, obstacle clearance capability analysis, variable impact analysis, joint function construction, model building, and model optimization, can improve the accuracy of user intent recognition in multi-dimensional human-computer interaction scenarios, thereby improving the accuracy of joint optimization of search methods and flight paths. Joint optimization of search methods and flight paths can greatly save resources and improve work efficiency. It can realize intelligent joint optimization of search methods and flight paths, and perform obstacle clearance capability analysis and multi-data fusion in real time for joint optimization of search methods and flight paths. It is of great significance for obstacle clearance capability analysis, and can adapt to obstacle clearance capability analysis of different standards and different obstacle clearance capability analysis needs, thus having a certain degree of universality. Attached Figure Description

[0078] Figure 1 This is a flowchart illustrating the steps of the AI-based autonomous obstacle avoidance path planning method and system for unmanned aerial vehicles (UAVs) of the present invention.

[0079] Figure 2 This is a schematic diagram of the structure of an electronic device in an embodiment of this specification. Detailed Implementation

[0080] The present invention will be further described below through specific embodiments. The illustrative embodiments and descriptions herein are used to explain the present invention, but are not intended to limit the present invention.

[0081] The AI-based autonomous obstacle avoidance path planning method for unmanned aerial vehicles (UAVs) of this invention includes the following steps:

[0082] like Figure 1 As shown, this embodiment includes the following steps:

[0083] Collect path data and related data for the target area, and preprocess the path data and related data;

[0084] In the actual assessment, a batch of electronic products that needed to be transported autonomously from the warehouse to a distribution center 20km away in a logistics park was taken as the research object;

[0085] The path data includes flight speed, flight altitude, flight path, temperature, humidity, wind speed, and traffic control information; related data includes the weight, volume, and fragility of electronic products.

[0086] The path data is filtered for anomalies based on the passage threshold to obtain anomaly point information, and the obstacle passage capability of the anomaly point information is analyzed to obtain the first parameter variable.

[0087] In the actual assessment, there were 15 outliers, with the first parameter being 0.6918;

[0088] A variable influence analysis is performed on the relevant data to obtain the second parameter, and a joint function is constructed based on the first parameter and the second parameter.

[0089] In the actual evaluation, the second parameter was 0.4471;

[0090] Construct a target path optimization model based on the joint function, input the data to be optimized into the target path optimization model, and output the optimization results;

[0091] In the actual assessment, the optimization result was to change the flight altitude and route to avoid traffic-controlled areas and high-risk airspace. The flight distance became 22km, but the flight time was shortened by 15 minutes and energy consumption was reduced by 10%.

[0092] In this embodiment, the method for obtaining anomaly point information by anomaly filtering of the path data based on a passage threshold includes:

[0093] The time points where path data shows dwell times and flight speeds below 30 km / h are designated as problem data points. The information weight of these problem data points is calculated based on their data characteristics and importance.

[0094]

[0095] The i-th path data of the u-th problem data point is u. i The path data of the y-th problem data point is u. y Path data u i The probability of occurrence is Path data u y The probability of occurrence is Path data u i and path data u y The probability of them occurring simultaneously is Observed path data u y Time path data u i The posterior probability is The number of path data is Path data u i Information weight is

[0096] Problem data points with an information weight greater than 0.412 are designated as key data points, and optimization is performed on these key data points:

[0097] Calculate particle fitness:

[0098]

[0099] The probability parameter is chosen as sp, and the position of particle x is D. x Position D x Information weight is The fitness function of particle x is

[0100] We use roulette wheel selection to choose superior particles as the parents for crossover mutations, and calculate the selection probability:

[0101]

[0102] The z-th particle is x z Particle x z fitness is The number of particles is Particle x z The probability of selection is Particle x z The position is D(x) z );

[0103] Perform crossover and mutation on the particles, and calculate the crossover rate and mutation rate:

[0104]

[0105] The crossover rate is The rate of variation The parameters between 0 and 1 are β1, β2, β3, and β4, respectively, and the average fitness of the particle population is... The maximum fitness of the particle population is The fitness of the crossover individual is The fitness of the variant individual is

[0106] The gene segments of the particles after crossover mutation are reverse-sorted to obtain new particles. The fitness of the new particles is calculated, and the passage threshold is calculated based on the fitness.

[0107]

[0108] Where the passage threshold is η, and the new particle x new fitness is The flight speed of the u-th problem data point is bu The minimum flight speed of the u-th data point is b. min ;

[0109] The process iterates until the maximum number of iterations is reached, and then outputs key data points with a passage threshold greater than 0.374 as outlier information.

[0110] In this embodiment, the method for obtaining the first parameter variable by performing obstacle clearance capability analysis on the anomaly point information includes:

[0111] Input anomaly information into the obstacle capability analysis model, and calculate airspace traffic based on the anomaly information:

[0112] p = a·f

[0113] a=(k) -1

[0114] Where airspace traffic is p, congestion density is a, and the spacing between UAVs is k;

[0115] Calculate the relationship between drone traffic flow and speed:

[0116]

[0117] The relationship between drone flow and speed is ρ, the function curvature control parameter is μ, and the maximum density causing airspace congestion is a. max The maximum speed of the drone is f. v ;

[0118] Calculate the spacing between drones:

[0119]

[0120] The length of the drone fuselage is l, and the minimum safe distance of the drone is s. o The drone's flight speed is f, and the free flow velocity is f. o The safe head-to-head distance for manually piloted drones is H. o ;

[0121] Based on the spacing between UAVs and airspace traffic, the basic graph model analysis function for airspace flow is given:

[0122]

[0123] Anomaly information is analyzed using analytical functions to obtain airspace traffic and airspace congestion density. The first parameter is obtained by weighted summation of airspace traffic and airspace congestion density.

[0124] In this embodiment, the method for obtaining the second parameter by performing variable influence analysis on the relevant data includes:

[0125] Construct a state matrix based on relevant data, using the time points of the relevant data as nodes, and calculate the contribution of each node:

[0126]

[0127] The contribution of the t-th node and the a-th related data is ψ. a (t), the flight score of the a-th related data at the t-th node is The flight score of the a-th relevant data at the (t-1)th node is The number of related data for the t-th node is The regulation coefficient of the a-th relevant data is γ a The a-th related data of the t-th node is h. a (t), where h is the a-th related data of the (t-1)-th node. a (t-1), the degree of influence is σ, and the variable of the a-th related data of the t-th node is Δη. a,t The flight score index value of the w-th related data of the t-th node is V. a,w (t), the number of flight score indicators is N, and the weight of the w-th flight score indicator is λ. w ;

[0128] Calculate the average probability of the node:

[0129]

[0130] The average probability of the a-th related data at the t-th node is: The number of nodes containing the a-th related data is M, and the probability of the a-th related data appearing in the t-th node is ζ. a (t);

[0131] Construct a weighted directed graph based on the directed weighted network, and calculate the importance index of the nodes:

[0132]

[0133] The importance index of the t-th node is ε. u The natural constant is e, and the weight of the connection sent out by the t-th node is C. ot (u), where the weight of the connection received by the t-th node is C. in (u), with an influence coefficient of υ;

[0134] Compute the weighted value of the nodes:

[0135]

[0136] The weighted value of the t-th node is: The probability that the a-th related data point at the t-th node is in an influenced state is ζ. a,yx (t), with a weighting factor of λ;

[0137] Calculate the second parameter:

[0138]

[0139] The flight score of the t-th node is The standard value of the a-th relevant data is h. ao The standard value for flight score is The second parameter is

[0140] In this embodiment, the method for constructing a joint function based on the first parameter and the second parameter includes:

[0141] Priority data is obtained by filtering based on search methods and flight paths, and flight evaluation indicators are obtained based on the priority data.

[0142] Based on the construction of the joint function, the expression is:

[0143]

[0144] The second parameter is The first parameter is The optimization coefficient is The actual arrival time is T st Agreed arrival time The number of the nth type of unmanned aerial vehicle is B. n The energy consumption per kilometer of the nth type of unmanned aerial vehicle is g. n The flight speed and range of the nth type of unmanned aerial vehicle is S. n The sign function is sgn(·), and the fixed cost of flight is R. gd The joint function is

[0145] In this embodiment, the method for constructing a target path optimization model based on the joint function includes:

[0146] The objective function is constructed based on the joint function, and its expression is:

[0147]

[0148] Where the control factor is χ, and the loss function is... The objective function is

[0149] Target path optimization models include particle swarm optimization algorithm, random forest algorithm, and recurrent neural network;

[0150] Particle swarm optimization (PSO) algorithms adjust the search direction and position by sharing information and accumulating individual experience among particles, adapting to the joint function requirements of flight, and achieving efficient search and optimization in a multi-dimensional feature space to obtain improved flight schemes.

[0151] The random forest algorithm constructs multiple decision trees and obtains prediction results based on improved flight schemes, captures the nonlinear relationships of the prediction results, and extracts effective features based on the nonlinear relationships;

[0152] Recurrent neural networks update their internal state at sequential time steps through a recursive loop structure, capturing the dynamic features of the data to be optimized, and predicting the joint function value based on the effective features. Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Please refer to it. Figure 2 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and memory. The memory may include main memory, such as high-speed random-access memory (RAM), or non-volatile memory, such as at least one disk drive. Of course, the electronic device may also include other hardware required for other business operations.

[0153] The processor, network interface, and memory can be interconnected via an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 2 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0154] Memory is used to store programs. Specifically, programs may include program code, which includes computer operation instructions. Memory may include main memory and non-volatile memory, and provides instructions and data to the processor.

[0155] The processor reads the corresponding computer program from non-volatile memory into main memory and then runs it, forming an AI-based autonomous obstacle avoidance path planning device for drones at the logical level. The processor executes the program stored in memory and is specifically used to execute any of the aforementioned AI-based autonomous obstacle avoidance path planning methods for drones.

[0156] The above is as stated in this application. Figure 1 The AI-based autonomous obstacle avoidance path planning method for unmanned aerial vehicles (UAVs) disclosed in the illustrated embodiments can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0157] The electronic device can also perform Figure 1 A method for autonomous obstacle avoidance path planning for unmanned aerial vehicles based on AI was developed and implemented. Figure 1 The functions of the embodiments shown are not described in detail here.

[0158] This application also proposes a computer-readable storage medium that stores one or more programs, the programs including instructions that, when executed by an electronic device including multiple applications, perform any of the aforementioned AI-based drone autonomous obstacle avoidance path planning methods.

[0159] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0160] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0161] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0162] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0163] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0164] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0165] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0166] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0167] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0168] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An AI-based autonomous obstacle avoidance path planning method for unmanned aerial vehicles (UAVs), characterized in that, Includes the following steps: Collect path data and related data of the target area, and preprocess the path data and related data; the related data includes the weight, volume, and fragility of the electronic product; The path data is filtered for anomalies based on the passage threshold to obtain anomaly point information, and the obstacle passage capability of the anomaly point information is analyzed to obtain the first parameter variable. A variable influence analysis is performed on the relevant data to obtain a second parameter, and a joint function is constructed based on the first and second parameters; including: Construct a state matrix based on relevant data, using the time points of the relevant data as nodes, and calculate the contribution of each node: The contribution of the t-th node and the a-th related data is: The flight score of the t-th node and the a-th related data is The flight score of the a-th relevant data at the (t-1)-th node is The number of related data for the t-th node is The regulation coefficient of the a-th related data is The coefficients that adjust the degree of flight impact based on relevant data are as follows: the a-th relevant data point at the t-th node is... The a-th related data of the (t-1)-th node is The degree of impact is The variable of the a-th related data of the t-th node is The value of the w-th flight score index for the a-th related data of the t-th node is The number of flight score indicators is N, and the weight of the w-th flight score indicator is... ; Calculate the average probability of the node: The average probability of the a-th related data at the t-th node is: The number of nodes containing the a-th related data is M, and the probability of the a-th related data appearing in the t-th node is M. ; Construct a weighted directed graph based on the directed weighted network, and calculate the importance index of the nodes: The importance index of the t-th node is: The natural constant is e, and the weight of the connection sent out by the t-th node is... The weight of the connection received by the t-th node is The influence coefficient is ; Compute the weighted value of the nodes: The weighted value of the t-th node is: The probability that the a-th related data of the t-th node is in an influenced state is The weighting factor is ; Calculate the second parameter: The flight score of the t-th node is The standard value of the a-th relevant data is The standard value for flight score is The second parameter is ; Priority data is obtained by filtering based on search methods and flight paths, and flight evaluation indicators are obtained based on the priority data. Based on the construction of the joint function, the expression is: The second parameter is The first parameter is The optimization coefficient is The actual arrival time is Agreed arrival time , No. The number of flying drones is The energy consumption per kilometer of the nth type of unmanned aerial vehicle is The flight speed and range of the nth type of unmanned aerial vehicle are The symbolic function is Fixed flight costs are The joint function is ; A target path optimization model is constructed based on the joint function, the data to be optimized is input into the target path optimization model, and the optimization result is output; including: The objective function is constructed based on the joint function, and its expression is: Where the control factor is The loss function is The objective function is ; Target path optimization models include particle swarm optimization algorithm, random forest algorithm, and recurrent neural network; Particle swarm optimization (PSO) algorithms adjust the search direction and position by sharing information and accumulating individual experience among particles, adapting to the joint function requirements of flight, and achieving efficient search and optimization in a multi-dimensional feature space to obtain improved flight schemes. The random forest algorithm constructs multiple decision trees and obtains prediction results based on improved flight schemes, captures the nonlinear relationships of the prediction results, and extracts effective features based on the nonlinear relationships; Recurrent neural networks update their internal state at sequential time steps through a recursive loop structure, capturing the dynamic features of the data to be optimized, and predicting the joint function value based on the effective features.

2. The AI-based autonomous obstacle avoidance path planning method for unmanned aerial vehicles according to claim 1, characterized in that, A method for obtaining anomaly point information by filtering the path data based on a passage threshold includes: The time points where path data shows dwell times and flight speeds below 30 km / h are designated as problem data points. The information weight of these problem data points is calculated based on their data characteristics and importance. The i-th path data of the u-th problem data point is... The path data of the y-th problem data point is: Path data The probability of occurrence is Path data The probability of occurrence is Path data and path data The probability of them occurring simultaneously is Observed path data Time path data The posterior probability is The number of path data is Path data Information weight is ; Problem data points with an information weight greater than 0.412 are designated as key data points, and optimization is performed on these key data points: Calculate particle fitness: Where the selection probability parameter is The position of particle x is ,Location Information weight is The fitness function of particle x is ; We use roulette wheel selection to choose superior particles as the parents for crossover mutations, and calculate the selection probability: The z-th particle is ,particle fitness is The number of particles is ,particle The probability of selection is ,particle The position is ; Perform crossover and mutation on the particles, and calculate the crossover rate and mutation rate: The crossover rate is The mutation rate is The parameters between 0 and 1 are respectively , , , The average fitness of the particle population is The maximum fitness of the particle population is The fitness of the crossover individuals is The fitness of the mutated individual is ; The gene segments of the particles after crossover mutation are reverse-sorted to obtain new particles. The fitness of the new particles is calculated, and the passage threshold is calculated based on the fitness. The passage threshold is New particles fitness is The flight speed of the u-th problem data point is The minimum flight speed of the u-th problem data point is ; The process iterates until the maximum number of iterations is reached, and then outputs key data points with a passage threshold greater than 0.374 as outlier information.

3. The AI-based autonomous obstacle avoidance path planning method for unmanned aerial vehicles according to claim 1, characterized in that, A method for obtaining a first parameter variable by performing obstacle capability analysis on the anomaly point information includes: Input anomaly information into the obstacle capability analysis model, and calculate airspace traffic based on the anomaly information: Where the airspace traffic is p, the congestion density is a, and the spacing between UAVs is... ; Calculate the relationship between drone traffic flow and speed: The relationship between drone traffic flow and speed is as follows: The function curvature control parameter is The maximum density that causes airspace congestion is The maximum speed of the drone is ; Calculate the distance between drones: The length of the drone fuselage is The minimum safe distance for drones is The drone's flight speed is The free flow velocity is The safe time distance for drones during manual intervention is ; Based on the spacing between UAVs and airspace traffic, the basic graph model analysis function for airspace flow is given: Anomaly information is analyzed using analytical functions to obtain airspace traffic and airspace congestion density. The first parameter is obtained by weighted summation of airspace traffic and airspace congestion density.

4. An electronic device, comprising: processor; as well as A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the method of any one of claims 1 to 3.

5. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of applications, cause the electronic device to perform the method of any one of claims 1 to 3.

Images