Unmanned aerial vehicle flight path analysis method and system for port logistics and unmanned aerial vehicle

Abstract

The invention discloses an unmanned aerial vehicle flight path analysis method and system for port logistics and an unmanned aerial vehicle. The method relates to the technical field of autonomous navigation, and comprises the following steps: collecting and processing flight path analysis data of an unmanned aerial vehicle, and analyzing, optimizing and adjusting the data. According to the method, the unmanned aerial vehicle flight path analysis data is collected and processed, the unmanned aerial vehicle flight path analysis data is analyzed, the navigation precision accuracy evaluation value and the unmanned aerial vehicle dynamic obstacle avoidance evaluation value are obtained, comprehensive analysis and optimization adjustment are carried out, the real-time performance of the unmanned aerial vehicle flight path analysis method for port logistics is improved, and the accuracy of port logistics is improved. The problem that in the prior art, an unmanned aerial vehicle flight path analysis method for port logistics is insufficient in real-time performance is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous navigation, and particularly to a method and system for analyzing the flight trajectory of an unmanned aerial vehicle (UAV) for port logistics and a UAV. Background Art

[0002] As a key node in global trade, the logistics efficiency of ports directly affects the smooth progress of international trade. The traditional port logistics methods can no longer meet the requirements of high efficiency, accuracy, and flexibility, and there is an urgent need for technological innovation. The introduction of UAV technology can significantly improve the automation and intelligence levels of port logistics, reduce operating costs, and enhance overall efficiency. The environment in port container terminals is complex and changeable, including conditions such as strong winds, thick fog, and bad weather, which pose challenges to the flight and data collection of UAVs. These environmental factors may cause unstable flight of UAVs and even malfunctions.

[0003] The existing UAV flight trajectory analysis system for port logistics uses a method for planning the inspection path of port UAVs based on the improved A algorithm, sets the flight starting point and all task points, and then uses the improved A algorithm to plan the path between the starting point and the task points; through polynomial trajectory optimization, on the basis of minimizing the snap method, the generated detection path points are preprocessed by constructing an isosceles triangle.

[0004] For example, a method and system for optimizing the path of a UAV for port logistics disclosed in the patent application with the publication number: CN115390586A includes: S1. Open-loop scheduling optimization, with the goal of minimizing the time for each task of the UAV, taking the UAV position and task position as inputs, and outputting the shortest path for each task; S2. Constraint calculation, imposing constraints on the carrying capacity, running speed, and path point capacity of the UAV; S3. Model predictive control, predicting the remaining battery power after task completion, predicting the amount of goods that can be transported per unit of battery power when performing other tasks after each task is completed, and performing optimization control with the goal of maximizing the amount of goods transported per unit of battery power; it can improve the working efficiency of the UAV by optimizing the task path and task time of the port logistics UAV, thereby extending the working time of the UAV and reducing the inaccuracy in prediction caused by the difference between the predicted power consumption and the actual power consumption, and achieving precise control of the UAV to meet the large-area use requirements of ports.

[0005] For example, a method for planning the flight path of a UAV and a UAV disclosed in the patent application with the publication number: CN103809597A includes: obtaining the depth information of the UAV flight environment, and generating a two-dimensional grid map of the flight environment based on the depth information, where the flight environment includes obstacles; constructing a potential function for each grid and the corresponding obstacle in the two-dimensional grid map according to the positions of each grid and the obstacles; obtaining a weighted graph of the grids in the two-dimensional grid map according to the position of the UAV and the potential function; and determining the flight path according to the weighted graph.

[0006] However, in the process of implementing the inventive technical solution in the embodiments of the present application, it is found that the above technologies have at least the following technical problems:

[0007] When the drone flies inside a warehouse without GPS, the gyroscope has zero-bias instability, resulting in an offset between the actual flight path and the planned path due to inertial navigation errors, and there is a problem of unreliable flight paths. When the drone detects multiple dynamic obstacles such as moving cranes and transport vehicles at the same time, the path replanning response time is long, and there is a problem of insufficient dynamic obstacle avoidance ability in dense scenarios. When the drone flies according to the initial plan, the environment has changed, such as the crane has moved to a new position. Since the obstacle avoidance system does not update the obstacle position in real time, collisions occur and the trajectory cannot be dynamically corrected, resulting in a problem of insufficient real-time performance of the method for analyzing the flight trajectory of drones in port logistics. Summary of the Invention

[0008] The embodiments of the present application provide a method, a system and a drone for analyzing the flight trajectory of drones in port logistics, which solve the problem of insufficient real-time performance of the method for analyzing the flight trajectory of drones in port logistics in the prior art, and improve the real-time performance of the method for analyzing the flight trajectory of drones in port logistics.

[0009] The embodiments of the present application provide a method for analyzing the flight trajectory of drones in port logistics, including the following steps: collecting and processing the data for analyzing the flight trajectory of drones; analyzing the data for analyzing the flight trajectory of drones to obtain an evaluation value of the accuracy of navigation precision and an evaluation value of the dynamic obstacle avoidance of the drone; comprehensively analyzing through the evaluation value of the accuracy of navigation precision and the evaluation value of the dynamic obstacle avoidance of the drone to obtain an evaluation value of the real-time performance of the analysis of the flight trajectory of the drone; comparing and analyzing the evaluation value of the accuracy of navigation precision, the evaluation value of the dynamic obstacle avoidance of the drone and the evaluation value of the real-time performance of the analysis of the flight trajectory of the drone with the first determination value of the evaluation value of the accuracy of navigation precision, the second determination value of the evaluation value of the dynamic obstacle avoidance of the drone and the third determination value of the evaluation value of the real-time performance of the analysis of the flight trajectory of the drone to obtain a method for optimizing and adjusting the analysis of the flight trajectory of the drone.

[0010] Further, the specific steps for collecting and processing the UAV flight trajectory analysis data are as follows: collecting the original UAV flight trajectory analysis data through the UAV detection and positioning device; cleaning and denoising the original UAV flight trajectory analysis data to obtain the UAV flight trajectory analysis data; collecting continuous image frames by using a binocular camera, extracting key points and their descriptors from the images by using a feature extraction algorithm, matching the feature points between consecutive frames by using a feature point matching algorithm, estimating the position and attitude of the UAV through visual odometry technology, analyzing the cumulative scale error caused by the feature point matching error, recording the change of the scale error over time to form time series data, and calculating the rate of increase of the scale error over time, i.e., the scale drift rate, by fitting the time series data; the UAV flight trajectory analysis data includes navigation accuracy data and UAV dynamic obstacle avoidance data.

[0011] Further, the specific steps for obtaining the navigation accuracy evaluation value are as follows: the navigation accuracy data includes the heading angle error rate, the scale drift rate, the obstacle avoidance response time, the abscissa of the actual UAV position, and the ordinate of the actual UAV position; obtaining the heading angle error rate threshold, the scale drift rate threshold, the obstacle avoidance response time threshold, the standard value of the UAV position abscissa, the standard value of the UAV position ordinate, the weight factor of the heading angle error rate, the weight factor of the UAV position error coefficient, the weight factor of the scale drift rate, and the weight factor of the obstacle avoidance response time from the UAV flight trajectory analysis database; performing mean value processing on the ratio analysis result of the heading angle error rate threshold and the heading angle error rate in different navigation accuracy detection segments, denoted as the first component of the navigation accuracy evaluation value; performing coupling processing on the abscissa of the actual UAV position and the standard value of the UAV position abscissa and the ordinate of the actual UAV position and the standard value of the UAV position ordinate to obtain the second component of the navigation accuracy evaluation value; denoting the ratio analysis result of the scale drift rate threshold and the scale drift rate as the third component of the navigation accuracy evaluation value; performing mean value processing on the ratio analysis result of the obstacle avoidance response time and the obstacle avoidance response time threshold in different navigation accuracy detection segments, denoted as the fourth component of the navigation accuracy evaluation value; performing ratio analysis on the coupling result of the first component and the third component of the navigation accuracy evaluation value and the coupling result of the second component and the fourth component of the navigation accuracy evaluation value to obtain the navigation accuracy evaluation value; the navigation accuracy evaluation value represents the quantitative data of the degree of proximity between the UAV position information and the true position jointly by the first component, the second component, the third component, and the fourth component of the navigation accuracy evaluation value.

[0012] Further, the specific steps for obtaining the UAV dynamic obstacle avoidance evaluation value are as follows: The UAV dynamic obstacle avoidance data includes the obstacle avoidance response time, path replanning response time, scale drift rate, actual abscissa of the UAV obstacle avoidance action, and actual ordinate of the UAV obstacle avoidance action; Obtain the obstacle avoidance response time threshold, path replanning response time threshold, scale drift rate threshold, standard abscissa value of the UAV obstacle avoidance action, standard ordinate value of the UAV obstacle avoidance action, weight factor of the obstacle avoidance response time, weight factor of the path replanning response time, weight factor of the scale drift rate, and weight factor of the dynamic trajectory deviation coefficient from the UAV flight trajectory analysis database; Average the ratio analysis results of the obstacle avoidance response time and the obstacle avoidance response time threshold in different UAV dynamic obstacle avoidance detection segments, denoted as the first component of the UAV dynamic obstacle avoidance evaluation value; Average the ratio analysis results of the path replanning response time and the path replanning response time threshold in different UAV dynamic obstacle avoidance detection segments, denoted as the second component of the UAV dynamic obstacle avoidance evaluation value; Denote the ratio analysis result of the scale drift rate and the scale drift rate threshold as the third component of the UAV dynamic obstacle avoidance evaluation value; Calculate the ratios of the actual abscissa of the UAV obstacle avoidance action to the standard abscissa value of the UAV obstacle avoidance action and the actual ordinate of the UAV obstacle avoidance action to the standard ordinate value of the UAV obstacle avoidance action to obtain the fourth component of the UAV dynamic obstacle avoidance evaluation value; Couple the first component of the UAV dynamic obstacle avoidance evaluation value, the second component of the UAV dynamic obstacle avoidance evaluation value, the third component of the UAV dynamic obstacle avoidance evaluation value, and the fourth component of the UAV dynamic obstacle avoidance evaluation value to obtain the UAV dynamic obstacle avoidance evaluation value; The UAV dynamic obstacle avoidance evaluation value represents the quantitative data of the ability to avoid obstacles in complex and dynamic environments by the first component of the UAV dynamic obstacle avoidance evaluation value, the second component of the UAV dynamic obstacle avoidance evaluation value, the third component of the UAV dynamic obstacle avoidance evaluation value, and the fourth component of the UAV dynamic obstacle avoidance evaluation value together.

[0013] Further, the specific steps for comprehensively analyzing and obtaining the real-time evaluation value of the UAV flight trajectory analysis are as follows: The UAV dynamic obstacle avoidance data includes the threshold of the obstacle data update frequency, the weight factor of the navigation accuracy evaluation value, the weight factor of the obstacle data update frequency, and the weight factor of the UAV dynamic obstacle avoidance evaluation value; The navigation accuracy evaluation value is averaged at different navigation accuracy detection points, denoted as the first component of the real-time evaluation value of the UAV flight trajectory analysis; The ratio analysis result of the obstacle data update frequency to the threshold of the obstacle data update frequency is averaged at different real-time detection points of the UAV flight trajectory analysis, denoted as the second component of the real-time evaluation value of the UAV flight trajectory analysis; The UAV dynamic obstacle avoidance evaluation value is averaged at different UAV dynamic obstacle avoidance detection points, denoted as the third component of the real-time evaluation value of the UAV flight trajectory analysis; The ratio analysis is performed on the coupling result of the first component and the second component of the real-time evaluation value of the UAV flight trajectory analysis and the third component of the real-time evaluation value of the UAV flight trajectory analysis to obtain the real-time evaluation value of the UAV flight trajectory analysis; The real-time evaluation value of the UAV flight trajectory analysis represents the quantization degree of the first component, the second component, and the third component of the real-time evaluation value of the UAV flight trajectory analysis on the real-time data processing of the UAV during flight.

[0014] Further, the specific steps for obtaining the optimization and adjustment method of the UAV flight trajectory analysis are as follows: Obtain the first determination value of the navigation accuracy evaluation value, the second determination value of the UAV dynamic obstacle avoidance evaluation value, and the third determination value of the real-time evaluation value of the UAV flight trajectory analysis from the UAV flight trajectory analysis database; The optimization and adjustment method of the UAV flight trajectory analysis includes the navigation accuracy optimization and adjustment method, the UAV dynamic obstacle avoidance optimization and adjustment method, and the real-time optimization and adjustment method of the UAV flight trajectory analysis; If the navigation accuracy evaluation value is greater than or equal to the first determination value of the navigation accuracy evaluation value, there is no need to optimize and adjust the navigation accuracy; If the navigation accuracy evaluation value is lower than the first determination value of the navigation accuracy evaluation value, the Kalman filter is used to fuse the visual SLAM data to suppress the angle error; Connect directly to the port BIM system to obtain the coordinates of the dynamic obstacles, such as cranes and containers, in real time, generate a 4D flight corridor, the digital twin platform receives the crane positioning data, dynamically updates the flight restricted area, and the UAV adjusts the path according to the real-time corridor to avoid the signal occlusion area.

[0015] Further, the specific steps of the UAV dynamic obstacle avoidance optimization and adjustment method are as follows: If the UAV dynamic obstacle avoidance evaluation value is lower than or equal to the second determination value of the UAV dynamic obstacle avoidance evaluation value, there is no need to optimize and adjust the UAV dynamic obstacle avoidance; if the UAV dynamic obstacle avoidance evaluation value is greater than the second determination value of the UAV dynamic obstacle avoidance evaluation value, the ultrasonic data is processed based on the rule engine to trigger an emergency avoidance; the radar and visual data are fused, and deep reinforcement learning is used to generate an optimal path.

[0016] Further, the specific steps of the UAV flight trajectory analysis real-time optimization and adjustment method are as follows: If the UAV flight trajectory analysis real-time evaluation value is greater than or equal to the third determination value of the UAV flight trajectory analysis real-time evaluation value, there is no need to optimize and adjust the UAV flight trajectory analysis real-time; if the UAV flight trajectory analysis real-time evaluation value is lower than the third determination value of the UAV flight trajectory analysis real-time evaluation value, the surrounding obstacles are detected by millimeter-wave radar and vision, the data is uploaded to the blockchain, and based on the pheromone concentration and real-time obstacle data, the weights of each flight segment are calculated, and the optimal path such as the path with the smallest total weight is output, bypassing the mobile crane and selecting the container gap. If the priority of the new task is higher than the current task, path replanning is triggered.

[0017] The embodiment of the present application provides a UAV flight trajectory analysis system for port logistics, including a UAV flight trajectory analysis data acquisition and processing module, a UAV flight trajectory analysis data analysis module, a comprehensive analysis module, and an optimization and adjustment module: The UAV flight trajectory analysis data acquisition and processing module is used to collect and process UAV flight trajectory analysis data; the UAV flight trajectory analysis data analysis module is used to analyze the UAV flight trajectory analysis data to obtain a navigation accuracy evaluation value and a UAV dynamic obstacle avoidance evaluation value; the comprehensive analysis module is used to comprehensively analyze the navigation accuracy evaluation value and the UAV dynamic obstacle avoidance evaluation value to obtain a UAV flight trajectory analysis real-time evaluation value; the optimization and adjustment module is used to compare and analyze the navigation accuracy evaluation value, the UAV dynamic obstacle avoidance evaluation value, and the UAV flight trajectory analysis real-time evaluation value with the first determination value of the navigation accuracy evaluation value, the second determination value of the UAV dynamic obstacle avoidance evaluation value, and the third determination value of the UAV flight trajectory analysis real-time evaluation value to obtain the UAV flight trajectory analysis optimization and adjustment method.

[0018] The embodiment of the present application provides a UAV, including that the UAV cargo hold adopts a carbon fiber honeycomb sandwich structure, is equipped with an adaptive electromagnetic suspension system, and the damping of the cargo hold is adjusted in real time through electrorheological fluid; the UAV fuselage is sprayed with a polyurethane coating containing microcapsules; the UAV motor adopts a closed magnetic suspension bearing; the belly of the UAV is coated with a superhydrophobic material.

[0019] One or more technical solutions provided in the embodiment of the present application have at least the following technical effects or advantages:

[0020] 1. By collecting and processing the UAV flight trajectory analysis data, analyzing the UAV flight trajectory analysis data, obtaining the navigation accuracy evaluation value and the UAV dynamic obstacle avoidance evaluation value, comprehensively analyzing and optimizing and adjusting, the real-time performance of the UAV flight trajectory analysis method for port logistics is improved, and the problem of insufficient real-time performance of the UAV flight trajectory analysis method for port logistics in the existing technology is solved.

[0021] 2. By analyzing the UAV flight trajectory analysis data, based on the cold atom interferometer to measure the angular velocity and acceleration, eliminating the zero-bias instability of the traditional MEMS gyroscope, and then fusing the visual SLAM data through Kalman filtering to suppress the cumulative error, thereby optimizing the navigation accuracy.

[0022] 3. By comprehensively analyzing, the real-time evaluation value of the UAV flight trajectory analysis is obtained, so as to be able to provide an instant flight trajectory analysis result, quickly respond to various changes during flight, and then prevent accidents and optimize the flight path, which can reduce the operation and maintenance costs of the UAV. Brief Description of the Drawings

[0023] Figure 1 It is a flowchart of the UAV flight trajectory analysis method for port logistics provided by the embodiment of the present application;

[0024] Figure 2 It is a schematic diagram of the real-time evaluation value function of the UAV flight trajectory analysis provided by the embodiment of the present application;

[0025] Figure 3 It is a schematic diagram of the structure of the UAV flight trajectory system for port logistics provided by the embodiment of the present application. Detailed Embodiments

[0026] The embodiment of the present application provides a UAV flight trajectory analysis method, system and UAV for port logistics, which solves the problem of insufficient real-time performance of the UAV flight trajectory analysis method for port logistics in the existing technology. By collecting and processing the UAV flight trajectory analysis data, analyzing the UAV flight trajectory analysis data, obtaining the navigation accuracy evaluation value and the UAV dynamic obstacle avoidance evaluation value, comprehensively analyzing and optimizing and adjusting, the real-time performance of the UAV flight trajectory analysis method for port logistics is improved.

[0027] The technical solution in the embodiment of the present application is to solve the above problem of insufficient real-time performance of the UAV flight trajectory analysis method for port logistics, and the general idea is as follows:

[0028] By collecting and processing the analysis data of the UAV flight trajectory, analyzing the analysis data of the UAV flight trajectory, obtaining the navigation accuracy evaluation value and the UAV dynamic obstacle avoidance evaluation value, comprehensively analyzing and optimizing and adjusting, the real-time performance of the UAV flight trajectory analysis method for port logistics is improved.

[0029] To better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings of the specification and specific implementation manners.

[0030] As Figure 1 shown, it is a flowchart of the UAV flight trajectory analysis method for port logistics provided by an embodiment of the present application. The method includes the following steps: collecting and processing the analysis data of the UAV flight trajectory; analyzing the analysis data of the UAV flight trajectory to obtain the navigation accuracy evaluation value and the UAV dynamic obstacle avoidance evaluation value; obtaining the real-time evaluation value of the UAV flight trajectory analysis through comprehensive analysis of the navigation accuracy evaluation value and the UAV dynamic obstacle avoidance evaluation value; comparing and analyzing the navigation accuracy evaluation value, the UAV dynamic obstacle avoidance evaluation value, and the real-time evaluation value of the UAV flight trajectory analysis with the first determination value of the navigation accuracy evaluation value, the second determination value of the UAV dynamic obstacle avoidance evaluation value, and the third determination value of the real-time evaluation value of the UAV flight trajectory analysis to obtain the optimization and adjustment method for the UAV flight trajectory analysis.

[0031] In this embodiment, a UAV automatic battery replacement station is deployed on the top of the quay crane of the wharf. The mechanical arm and visual positioning are used to achieve battery replacement in 90 seconds. The battery pack integrates a supercapacitor module, supports instantaneous large current discharge (peak power 20kW), and can cope with sudden gusts for emergency climbing. The proton exchange membrane fuel cell (power density 3kW / kg) has a battery life of 350km. The port hydrogen production station provides green hydrogen energy, and the carbon footprint is reduced by 92% compared with diesel trailers.

[0032] Further, the specific steps of collecting and processing the analysis data of the UAV flight trajectory are as follows: collecting the original data of the UAV flight trajectory analysis through the UAV detection and positioning device; cleaning and denoising the original data of the UAV flight trajectory analysis to obtain the analysis data of the UAV flight trajectory; collecting continuous image frames through the use of binocular cameras, extracting key points and their descriptors from the images by using feature extraction algorithms, matching the feature points between consecutive frames by using feature point matching algorithms, estimating the position and attitude of the UAV through visual odometry technology, analyzing the cumulative scale error caused by feature point matching errors, recording the change of the scale error over time to form time series data, and calculating the rate of increase of the scale error over time, that is, the scale drift rate, by fitting the time series data; the analysis data of the UAV flight trajectory includes navigation accuracy data and UAV dynamic obstacle avoidance data.

[0033] In this embodiment, large obstacles such as cranes and trucks are identified by millimeter-wave radar, with a detection range of 200 m; the gaps between containers are identified by binocular camera AI, with a minimum passable width of 1.2 m; a 360° near-field detection is covered by an ultrasonic array to identify small obstacles such as barbed wire and cables, with a response time of less than 50 ms. The millimeter-wave radar pre-screens obstacles and demarcates a safety boundary; the binocular camera identifies passable areas such as container gaps; the ultrasonic array monitors real-time near-field sudden obstacles such as suddenly appearing forklifts.

[0034] Further, the specific steps to obtain the navigation accuracy evaluation value are as follows: The navigation accuracy data includes the heading angle error rate, scale drift rate, obstacle avoidance response time, the abscissa of the actual position of the UAV, and the ordinate of the actual position of the UAV; Obtain the heading angle error rate threshold, scale drift rate threshold, obstacle avoidance response time threshold, the standard value of the abscissa of the UAV position, the standard value of the ordinate of the UAV position, the weight factor of the heading angle error rate, the weight factor of the UAV position error coefficient, the weight factor of the scale drift rate, and the weight factor of the obstacle avoidance response time from the UAV flight trajectory analysis database; Perform mean value processing on the ratio analysis results of the heading angle error rate threshold and the heading angle error rate in different navigation accuracy detection segments, which is denoted as the first component of the navigation accuracy evaluation value; Perform coupling processing on the abscissa of the actual position of the UAV and the standard value of the abscissa of the UAV position and the ordinate of the actual position of the UAV and the standard value of the ordinate of the UAV position to obtain the second component of the navigation accuracy evaluation value; Denote the ratio analysis result of the scale drift rate threshold and the scale drift rate as the third component of the navigation accuracy evaluation value; Perform mean value processing on the ratio analysis results of the obstacle avoidance response time and the obstacle avoidance response time threshold in different navigation accuracy detection segments, which is denoted as the fourth component of the navigation accuracy evaluation value; Perform ratio analysis on the coupling result of the first component and the third component of the navigation accuracy evaluation value and the coupling result of the second component and the fourth component of the navigation accuracy evaluation value to obtain the navigation accuracy evaluation value; The navigation accuracy evaluation value represents the quantitative data of the closeness between the UAV position information and the real position jointly by the first component, the second component, the third component, and the fourth component of the navigation accuracy evaluation value.

[0035] In this embodiment, the specific method for analyzing and obtaining the navigation accuracy evaluation value is as follows:

[0036]

[0037] ρ1 + ρ2 + ρ3 + ρ4 = 1;

[0038] Number the preset navigation accuracy detection points in sequence. S0 represents the number of the navigation accuracy detection point under the T0-th navigation accuracy detection segment, and S0 = 1, 2,..., S, where S represents the total number of navigation accuracy detection points.

[0039] Divide the preset navigation accuracy time into navigation accuracy detection segments of the same time length. T0 represents the number of the navigation accuracy detection segment, and T0 = 1, 2,..., T, where T represents the total number of navigation accuracy detection segments.

[0040] Represents the navigation accuracy evaluation value of the S0-th navigation accuracy detection point.

[0041] W1 represents the threshold of the heading angle error rate, which is a preset heading angle error rate threshold obtained from the UAV flight trajectory analysis database and can be the average value of the heading angle error rates under the preset navigation accuracy detection segments in the historical database.

[0042] Represents the heading angle error rate under the T0-th navigation accuracy detection segment, which is the maximum deviation angle between the UAV heading angle and the preset trajectory under the navigation accuracy detection segment and reflects the influence of gyro zero bias instability on direction control.

[0043] Represents the UAV position error coefficient under the S0-th navigation accuracy detection point.

[0044] ω1 represents the threshold of the scale drift rate, which is a preset scale drift rate threshold obtained from the UAV flight trajectory analysis database and can be the average value of the scale drift rates under the preset navigation accuracy detection segments in the historical database.

[0045] Represents the scale drift rate under the T0-th navigation accuracy detection segment, which is the rate at which the scale of the system changes with the path due to error accumulation and is the ratio of the absolute difference between the preset path length and the actual path length of the UAV within the navigation accuracy detection segment to the actual path length.

[0046] Represents the obstacle avoidance response time under the T0-th navigation accuracy detection segment.

[0047] B1 represents the obstacle avoidance response time threshold, which is a preset obstacle avoidance response time threshold obtained from the UAV flight trajectory analysis database and can be the average value of the obstacle avoidance response times under the preset navigation accuracy detection segments in the historical database.

[0048] Represents the abscissa of the actual position of the drone at the S0th navigation accuracy detection point.

[0049] X1 represents the standard value of the abscissa of the drone position, which is a preset standard value of the abscissa of the drone position obtained from the drone flight trajectory analysis database and can be the average value of the abscissa of the drone position at the preset navigation accuracy detection point in the historical database.

[0050] Represents the ordinate of the actual position of the drone at the S0th navigation accuracy detection point.

[0051] Y1 represents the standard value of the ordinate of the drone position, which is a preset standard value of the ordinate of the drone position obtained from the drone flight trajectory analysis database and can be the average value of the ordinate of the drone position at the preset navigation accuracy detection point in the historical database.

[0052] ρ1 is a preset heading angle error rate weight factor obtained from the drone flight trajectory analysis database.

[0053] ρ2 is a preset drone position error coefficient weight factor obtained from the drone flight trajectory analysis database.

[0054] ρ3 is a preset scale drift rate weight factor obtained from the drone flight trajectory analysis database.

[0055] ρ4 is a preset obstacle avoidance response time weight factor obtained from the drone flight trajectory analysis database.

[0056] Obtain the mapping table of weight factors through the database, and quickly extract the corresponding weight factors according to the current heading angle error rate, drone position error coefficient, scale drift rate, and obstacle avoidance response time, such as the weight factor of the heading angle error rate, the weight factor of the drone position error coefficient, the weight factor of the scale drift rate, and the weight factor of the obstacle avoidance response time. This mapping table defines a clear set of association rules that convert the specific values of the heading angle error rate, drone position error coefficient, scale drift rate, and obstacle avoidance response time into their corresponding weight factors. Under this mechanism, whether it is to achieve a one-to-one exact match or a one-to-many relationship where multiple parameters converge into a single weight, the dynamic acquisition of weight factors can be effectively achieved.

[0057] The heading angle error rate represents the deviation between the actual heading angle of the UAV and the predetermined heading angle. The higher the heading angle error rate, the less precise the heading control of the UAV, which will prolong the reaction time after the UAV detects an obstacle, and the obstacle avoidance response time becomes longer. The heading angle error will cause the UAV to have a position deviation during flight, thereby affecting the scale drift rate. The scale drift rate refers to the deviation rate between the actual path length and the expected path length of the UAV during flight. The higher the heading angle error rate, the higher the scale drift rate. The heading angle error will directly cause the actual flight path of the UAV to deviate from the predetermined path, thereby affecting the actual position abscissa and ordinate of the UAV. The higher the heading angle error rate, the greater the deviation between the actual position and the predetermined position of the UAV. The obstacle avoidance response time affects the execution of the UAV's obstacle avoidance action, and thus indirectly affects the actual position of the UAV. The longer the obstacle avoidance response time, the UAV cannot avoid obstacles in time, resulting in a deviation between the actual position and the predetermined position.

[0058] There is a negative correlation between the heading angle error rate and the evaluation value of navigation accuracy. It means that the greater the deviation between the actual heading of the UAV and the predetermined heading, the more the UAV will deviate from the predetermined flight path. The greater the heading angle error rate, the smaller the evaluation value of navigation accuracy. There is a negative correlation between the obstacle avoidance response time and the evaluation value of navigation accuracy. The longer the reaction time of the UAV after detecting an obstacle, the more the UAV will deviate from the predetermined path during obstacle avoidance. The longer the obstacle avoidance response time, the smaller the evaluation value of navigation accuracy. There is a negative correlation between the scale drift rate and the evaluation value of navigation accuracy. It means that the greater the position deviation of the UAV, the higher the scale drift rate, and the smaller the evaluation value of navigation accuracy. There is a negative correlation between the sum of the square of the difference between the actual position abscissa of the UAV and the standard value of the UAV position abscissa and the sum of the square of the difference between the actual position ordinate of the UAV and the standard value of the UAV position ordinate and the evaluation value of navigation accuracy. It means that the greater the deviation between the flight path of the UAV and the predetermined path, the greater the sum of the square of the difference between the actual position abscissa of the UAV and the standard value of the UAV position abscissa and the sum of the square of the difference between the actual position ordinate of the UAV and the standard value of the UAV position ordinate, and the smaller the evaluation value of navigation accuracy.

[0059] Further, the specific steps to obtain the UAV dynamic obstacle avoidance evaluation value are as follows: The UAV dynamic obstacle avoidance data includes the obstacle avoidance response time, path replanning response time, scale drift rate, actual abscissa of the UAV obstacle avoidance action, and actual ordinate of the UAV obstacle avoidance action; Obtain the obstacle avoidance response time threshold, path replanning response time threshold, scale drift rate threshold, standard abscissa value of the UAV obstacle avoidance action, standard ordinate value of the UAV obstacle avoidance action, weight factor of the obstacle avoidance response time, weight factor of the path replanning response time, weight factor of the scale drift rate, and weight factor of the dynamic trajectory deviation coefficient from the UAV flight trajectory analysis database; Average the ratio analysis results of the obstacle avoidance response time to the obstacle avoidance response time threshold in different UAV dynamic obstacle avoidance detection segments, denoted as the first component of the UAV dynamic obstacle avoidance evaluation value; Average the ratio analysis results of the path replanning response time to the path replanning response time threshold in different UAV dynamic obstacle avoidance detection segments, denoted as the second component of the UAV dynamic obstacle avoidance evaluation value; Denote the ratio analysis result of the scale drift rate to the scale drift rate threshold as the third component of the UAV dynamic obstacle avoidance evaluation value; Perform ratio processing on the actual abscissa of the UAV obstacle avoidance action and the standard abscissa value of the UAV obstacle avoidance action and the actual ordinate of the UAV obstacle avoidance action and the standard ordinate value of the UAV obstacle avoidance action to obtain the fourth component of the UAV dynamic obstacle avoidance evaluation value; Couple the first component of the UAV dynamic obstacle avoidance evaluation value, the second component of the UAV dynamic obstacle avoidance evaluation value, the third component of the UAV dynamic obstacle avoidance evaluation value, and the fourth component of the UAV dynamic obstacle avoidance evaluation value to obtain the UAV dynamic obstacle avoidance evaluation value; The UAV dynamic obstacle avoidance evaluation value represents the quantitative data of the first component of the UAV dynamic obstacle avoidance evaluation value, the second component of the UAV dynamic obstacle avoidance evaluation value, the third component of the UAV dynamic obstacle avoidance evaluation value, and the fourth component of the UAV dynamic obstacle avoidance evaluation value jointly on the ability to avoid obstacles in a complex and dynamic environment.

[0060] In this embodiment, the specific method for analyzing and obtaining the UAV dynamic obstacle avoidance evaluation value is as follows:

[0061]

[0062] Number the preset UAV dynamic obstacle avoidance detection points in sequence. R0 represents the number of the UAV dynamic obstacle avoidance detection point under the K0th UAV dynamic obstacle avoidance detection segment, and R0 = 1, 2,..., R, where R represents the total number of UAV dynamic obstacle avoidance detection point numbers.

[0063] Divide the preset UAV dynamic obstacle avoidance time into UAV dynamic obstacle avoidance detection segments of the same time length M. K0 represents the number of the UAV dynamic obstacle avoidance detection segment, and K0 = 1, 2,..., K, where K represents the total number of UAV dynamic obstacle avoidance detection segment numbers.

[0064] Represents the UAV dynamic obstacle avoidance evaluation value of the R0th UAV dynamic obstacle avoidance detection point.

[0065] Represents the obstacle avoidance response time under the K0th UAV dynamic obstacle avoidance detection segment.

[0066] G1 represents the obstacle avoidance response time threshold, which is a preset obstacle avoidance response time threshold obtained from the UAV flight trajectory analysis database and can be the average value of the obstacle avoidance response time preset for the UAV dynamic obstacle avoidance detection segment in the historical database.

[0067] Represents the path replanning response time under the K0th UAV dynamic obstacle avoidance detection segment, which is the total time from detecting a new obstacle to generating a new path and starting to execute.

[0068] U1 represents the path replanning response time threshold, which is a preset path replanning response time threshold obtained from the UAV flight trajectory analysis database and can be the average value of the path replanning response time preset for the UAV dynamic obstacle avoidance detection segment in the historical database.

[0069] Represents the scale drift rate under the K0th UAV dynamic obstacle avoidance detection segment.

[0070] V1 represents the scale drift rate threshold, which is a preset scale drift rate threshold obtained from the UAV flight trajectory analysis database and can be the average value of the scale drift rate preset for the UAV dynamic obstacle avoidance detection segment in the historical database.

[0071] Represents the dynamic trajectory deviation coefficient under the K0th UAV dynamic obstacle avoidance detection segment.

[0072] Represents the actual abscissa of the UAV obstacle avoidance action under the K0th UAV dynamic obstacle avoidance detection segment.

[0073] Represents the standard value of the UAV obstacle avoidance action abscissa, which is a preset standard value of the UAV obstacle avoidance action abscissa obtained from the UAV flight trajectory analysis database and can be the average value of the UAV obstacle avoidance action abscissa preset for the UAV dynamic obstacle avoidance detection segment in the historical database.

[0074] Represents the actual ordinate of the UAV obstacle avoidance action under the K0th UAV dynamic obstacle avoidance detection segment.

[0075] Indicates the standard value of the vertical coordinate of the UAV obstacle avoidance action, which is a preset standard value of the vertical coordinate of the UAV obstacle avoidance action obtained from the UAV flight trajectory analysis database. It can be the average value of the vertical coordinates of the UAV obstacle avoidance actions under the preset UAV dynamic obstacle avoidance detection section in the historical database.

[0076] Is a preset obstacle avoidance response time weight factor obtained from the UAV flight trajectory analysis database.

[0077] Is a preset path replanning response time weight factor obtained from the UAV flight trajectory analysis database.

[0078] Is a preset scale drift rate weight factor obtained from the UAV flight trajectory analysis database.

[0079] Is a preset dynamic trajectory deviation coefficient weight factor obtained from the UAV flight trajectory analysis database.

[0080] Obtain the mapping table of weight factors through the database, and quickly extract the corresponding weight factors according to the current obstacle avoidance response time, path replanning response time, scale drift rate, and dynamic trajectory deviation coefficient, such as the weight factor of the obstacle avoidance response time, the weight factor of the path replanning response time, the weight factor of the scale drift rate, and the weight factor of the dynamic trajectory deviation coefficient. This mapping table defines a clear set of association rules, which converts the specific values of the obstacle avoidance response time, path replanning response time, scale drift rate, and dynamic trajectory deviation coefficient into their corresponding weight factors. Under this mechanism, whether it is to achieve one-to-one exact matching or the many-to-one relationship where multiple parameters converge into a single weight, the dynamic acquisition of weight factors can be effectively achieved.

[0081] The obstacle avoidance response time refers to the time interval from when the UAV detects an obstacle to when it starts to execute the obstacle avoidance action. The path replanning response time refers to the time required for the UAV to replan the flight path after detecting an obstacle. The shorter the obstacle avoidance response time, the faster the UAV can make an obstacle avoidance decision, thus providing more time for path replanning, and the shorter the path replanning response time. The higher the scale drift rate, the lower the accuracy of the UAV's position and obstacle detection, which will increase the obstacle avoidance response time and the path replanning response time. The UAV needs more time to process this inaccurate information, and the obstacle avoidance response time and the path replanning response time are longer. The actual abscissa and the actual ordinate of the UAV's obstacle avoidance action are the position coordinates where the UAV actually arrives after executing the obstacle avoidance action. They are the direct results of the obstacle avoidance response time and the path replanning response time. A short obstacle avoidance response time and path replanning response time mean that the UAV can complete the obstacle avoidance action faster and thus reach the predetermined obstacle avoidance position coordinates more accurately.

[0082] There is a positive correlation between the obstacle avoidance response time and the UAV's dynamic obstacle avoidance evaluation value. The slower the UAV's reaction to the obstacle, the more the obstacle avoidance performance deteriorates, and the longer the obstacle avoidance response time, the greater the UAV's dynamic obstacle avoidance evaluation value. There is a positive correlation between the path replanning response time and the UAV's dynamic obstacle avoidance evaluation value. The slower the UAV replans the path after encountering an obstacle, the higher the collision risk, and the longer the path replanning response time, the greater the UAV's dynamic obstacle avoidance evaluation value. There is a positive correlation between the scale drift rate and the UAV's dynamic obstacle avoidance evaluation value. The lower the accuracy of the UAV's estimation of its own position and the surrounding environment, the more it will affect the precision of the obstacle avoidance action. The higher the scale drift rate, the greater the UAV's dynamic obstacle avoidance evaluation value. There is a positive correlation between the sum of the square of the difference between the actual abscissa of the UAV's obstacle avoidance action and the standard value of the abscissa of the UAV's obstacle avoidance action and the sum of the square of the difference between the actual ordinate of the UAV's obstacle avoidance action and the standard value of the ordinate of the UAV's obstacle avoidance action and the UAV's dynamic obstacle avoidance evaluation value. A large deviation from the predetermined obstacle avoidance position indicates that the obstacle avoidance action is executed inaccurately. The greater the sum of the square of the difference between the actual abscissa of the UAV's obstacle avoidance action and the standard value of the abscissa of the UAV's obstacle avoidance action and the sum of the square of the difference between the actual ordinate of the UAV's obstacle avoidance action and the standard value of the ordinate of the UAV's obstacle avoidance action, the greater the UAV's dynamic obstacle avoidance evaluation value.

[0083] Further, the specific steps for comprehensively analyzing and obtaining the real-time evaluation value of UAV flight trajectory analysis are as follows: The UAV dynamic obstacle avoidance data includes the threshold of the obstacle data update frequency, the weight factor of the navigation accuracy evaluation value, the weight factor of the obstacle data update frequency, and the weight factor of the UAV dynamic obstacle avoidance evaluation value; The navigation accuracy evaluation value is averaged at different navigation accuracy detection points, denoted as the first component of the real-time evaluation value of UAV flight trajectory analysis; The ratio analysis result of the obstacle data update frequency to the threshold of the obstacle data update frequency is averaged at different real-time detection points of UAV flight trajectory analysis, denoted as the second component of the real-time evaluation value of UAV flight trajectory analysis; The UAV dynamic obstacle avoidance evaluation value is averaged at different UAV dynamic obstacle avoidance detection points, denoted as the third component of the real-time evaluation value of UAV flight trajectory analysis; The ratio analysis is performed on the coupling result of the first component and the second component of the real-time evaluation value of UAV flight trajectory analysis and the third component of the real-time evaluation value of UAV flight trajectory analysis to obtain the real-time evaluation value of UAV flight trajectory analysis; The real-time evaluation value of UAV flight trajectory analysis represents the quantization degree of the first component, the second component, and the third component of the real-time evaluation value of UAV flight trajectory analysis on the real-time data processing of the UAV during flight.

[0084] In this embodiment, the specific method for analyzing and obtaining the real-time evaluation value of UAV flight trajectory analysis is as follows:

[0085]

[0086] φ1 + φ2 + φ3 = 1;

[0087] The preset real-time detection points of UAV flight trajectory analysis are numbered in sequence. L0 represents the number of the real-time detection point of UAV flight trajectory analysis, and L0 = 1, 2,..., L, where L represents the total number of the numbers of the real-time detection points of UAV flight trajectory analysis.

[0088] represents the real-time evaluation value of UAV flight trajectory analysis.

[0089] represents the navigation accuracy evaluation value of the S0th navigation accuracy detection point.

[0090] represents the obstacle data update frequency at the L0th real-time detection point of UAV flight trajectory analysis, which is the refresh rate of the dynamic obstacle position information by the sensor or the digital twin platform.

[0091] ∈1 represents the threshold of the obstacle data update frequency, which is a preset obstacle data update frequency threshold obtained from the UAV flight trajectory analysis database and can be the average value of the obstacle data update frequency under the preset real-time detection points of the UAV flight trajectory analysis in the historical database.

[0092] Represents the UAV dynamic obstacle avoidance evaluation value of the R0th UAV dynamic obstacle avoidance detection point.

[0093] φ1 is the weight factor of the preset navigation accuracy evaluation value obtained from the UAV flight trajectory analysis database.

[0094] φ2 is the weight factor of the preset obstacle data update frequency obtained from the UAV flight trajectory analysis database.

[0095] φ3 is the weight factor of the preset UAV dynamic obstacle avoidance evaluation value obtained from the UAV flight trajectory analysis database.

[0096] By obtaining the mapping table of the weight factors from the database, the corresponding weight factors can be quickly extracted according to the current navigation accuracy evaluation value, obstacle data update frequency, and UAV dynamic obstacle avoidance evaluation value, such as the weight factor of the navigation accuracy evaluation value, the weight factor of the obstacle data update frequency, and the weight factor of the UAV dynamic obstacle avoidance evaluation value. This mapping table defines a clear set of association rules, which converts the specific values of the navigation accuracy evaluation value, obstacle data update frequency, and UAV dynamic obstacle avoidance evaluation value into their corresponding weight factors. Under this mechanism, whether it is to achieve an exact one-to-one match or a one-to-many relationship where multiple parameters converge into a single weight, the dynamic acquisition of the weight factors can be effectively realized.

[0097] Table 1 is an example table of the UAV flight trajectory analysis real-time evaluation value. The example parameters in Table 1 only take the parameters under one real-time detection point of the UAV flight trajectory analysis for illustration. The weight factor φ1 is set to 0.4, the weight factor φ2 is set to 0.2, the weight factor φ3 is set to 0.4, and the ∈1 obstacle data update frequency threshold is 2. Table 1 is as follows.

[0098] Table 1 Example Table of UAV Flight Trajectory Analysis Real-time Evaluation Value

[0099]

[0100]

[0101] The navigation accuracy evaluation value affects the dynamic obstacle avoidance ability of the UAV. The lower the navigation accuracy evaluation value, the greater the deviation between the actual flight of the UAV and the predetermined trajectory. As a result, when the UAV encounters an obstacle, its dynamic obstacle avoidance system cannot accurately judge the position and motion state of the obstacle, and the UAV dynamic obstacle avoidance evaluation value is larger; the higher the obstacle data update frequency, the more real-time the UAV can obtain the obstacle information in the surrounding environment, so as to more accurately predict the motion trajectory of the obstacle, and the UAV dynamic obstacle avoidance evaluation value is lower; the larger the navigation accuracy evaluation value, it helps the UAV to reach the predetermined position more accurately, thus providing a more stable environmental condition for the sensor system, which is conducive to increasing the obstacle data update frequency, and the larger the obstacle data update frequency.

[0102] As can be seen from Table 1, there is a positive correlation between the navigation accuracy evaluation value and the UAV flight trajectory analysis real-time evaluation value. The smaller the deviation between the actual flight trajectory of the UAV and the predetermined trajectory, the larger the navigation accuracy evaluation value, and the larger the UAV flight trajectory analysis real-time evaluation value; there is a negative correlation between the UAV dynamic obstacle avoidance evaluation value and the UAV flight trajectory analysis real-time evaluation value. When the UAV reacts slowly or the obstacle avoidance action is inaccurate during obstacle avoidance, the flight trajectory analysis system cannot obtain accurate flight trajectory information in time, and the larger the UAV dynamic obstacle avoidance evaluation value, the smaller the UAV flight trajectory analysis real-time evaluation value; there is a positive correlation between the obstacle data update frequency and the UAV flight trajectory analysis real-time evaluation value, enabling the flight trajectory analysis system to understand the changes in the surrounding environment more real-time and adjust the flight trajectory analysis results in time. The larger the obstacle data update frequency, the larger the UAV flight trajectory analysis real-time evaluation value.

[0103] As Figure 2 shown, it is a schematic diagram of the UAV flight trajectory analysis real-time evaluation value function provided by the embodiment of the present application; x is the positive half-axis of the abscissa, y is the positive half-axis of the ordinate, the weight factor φ1 is set to 0.4, the weight factor φ2 is set to 0.2, the weight factor φ3 is set to 0.4, and the ∈1 obstacle data update frequency threshold is 2.

[0104] Curve a indicates that if the UAV dynamic obstacle avoidance evaluation value is set to a fixed value of 1, the obstacle data update frequency is set to a fixed value of 1, the navigation accuracy evaluation value is x, and the UAV flight trajectory analysis real-time evaluation value is y, the UAV flight trajectory analysis real-time evaluation value increases as the navigation accuracy evaluation value increases.

[0105] Further, the specific steps for obtaining the method for optimizing and adjusting the UAV flight trajectory analysis are as follows: obtaining the first determination value of the navigation accuracy evaluation value, the second determination value of the UAV dynamic obstacle avoidance evaluation value, and the third determination value of the UAV flight trajectory analysis real-time evaluation value from the UAV flight trajectory analysis database; the method for optimizing and adjusting the UAV flight trajectory analysis includes the method for optimizing and adjusting the navigation accuracy, the method for optimizing and adjusting the UAV dynamic obstacle avoidance, and the method for optimizing and adjusting the UAV flight trajectory analysis real-time; if the navigation accuracy evaluation value is greater than or equal to the first determination value of the navigation accuracy evaluation value, there is no need to optimize and adjust the navigation accuracy; if the navigation accuracy evaluation value is lower than the first determination value of the navigation accuracy evaluation value, the Kalman filter is used to fuse the visual SLAM data to suppress the angle error; directly connect to the port BIM system to obtain the coordinates of the dynamic obstacles, such as cranes and containers, in real time, generate a 4D flight corridor, the digital twin platform receives the crane positioning data and dynamically updates the flight restricted area, and the UAV adjusts the path according to the real-time corridor to avoid the signal occlusion area.

[0106] In this embodiment, when the GPS signal is available, the quantum inertial navigation reference is calibrated. The quantum inertial navigation system is calibrated using the accurate position information provided by GPS, and the reference data of the quantum inertial navigation system is updated to ensure the accuracy of the high-precision pose information output by it, improve the accuracy of the quantum inertial navigation system, and reduce the cumulative error. When GPS is denied, the quantum module independently outputs high-precision pose. When the GPS signal is denied or unavailable, the quantum inertial navigation module independently outputs high-precision pose information to maintain the accuracy of the positioning and attitude estimation of the UAV, ensuring that the UAV can still maintain high-precision positioning and attitude control in the case of GPS unavailability. The visual SLAM data is fused through Kalman filtering. The Kalman filtering algorithm is used to fuse the pose information output by the quantum inertial navigation module and the environmental perception data provided by the visual SLAM system, and the fused data is processed to suppress the cumulative error, improve the stability and accuracy of the navigation system, and reduce the error accumulation during long-term flight. It is directly connected to the port BIM system to obtain dynamic obstacle information in real time. The UAV system is directly connected to the BIM (Building Information Modeling) system of the port to obtain the coordinate information of dynamic obstacles such as cranes and containers in real time, obtain real-time and accurate obstacle information, and provide data support for flight path planning. A 4D flight corridor is generated. According to the obtained dynamic obstacle information, a 4D flight corridor including the time dimension is generated to ensure that there are no obstacles in the flight corridor, provide a safe flight path for the UAV, plan a safe flight path, and avoid collisions with obstacles. The digital twin platform receives the crane positioning data. The digital twin platform receives the crane positioning data from the UWB base station, monitors the real-time position of the crane, obtains high-precision crane position information, and provides data support for dynamically updating the flight restricted area. The flight restricted area is dynamically updated. According to the real-time position and movement trajectory prediction of the crane, the flight restricted area is dynamically updated, and the flight restricted area information is transmitted to the UAV navigation system in real time to prevent the UAV from flying into the area of the crane movement trajectory and ensure flight safety. The UAV adjusts its path according to the real-time corridor. The UAV receives the real-time updated 4D flight corridor information, adjusts its flight path according to the flight corridor, avoids signal occlusion areas and flight restricted areas, ensures that the UAV always flies along a safe path, and avoids collisions and signal loss. Signal occlusion areas are avoided. Signal occlusion areas are identified, such as areas where buildings, large equipment, etc. may occlude GPS or communication signals, and the flight path is adjusted to avoid these areas, maintain the communication between the UAV and the ground control station, and ensure the effective transmission of flight control instructions.

[0107] Further, the specific steps of the UAV dynamic obstacle avoidance optimization and adjustment method are as follows: If the UAV dynamic obstacle avoidance evaluation value is lower than or equal to the second determination value of the UAV dynamic obstacle avoidance evaluation value, there is no need to optimize and adjust the UAV dynamic obstacle avoidance; if the UAV dynamic obstacle avoidance evaluation value is greater than the second determination value of the UAV dynamic obstacle avoidance evaluation value, the ultrasonic data is processed based on the rule engine to trigger an emergency avoidance; the radar and visual data are fused, and deep reinforcement learning is used to generate an optimal path.

[0108] In this embodiment, the millimeter-wave radar pre-screens the detected obstacles, differentiates different types of obstacles, delimits the safe flight boundary of the UAV according to the obstacle information, sets a safe flight area for the UAV, and avoids entering a dangerous area; the binocular camera continuously monitors the environment, identifies new passable areas, updates the passable area information in real time, transmits it to the UAV obstacle avoidance system, dynamically adjusts the flight path, and uses the identified passable areas for obstacle avoidance; the ultrasonic array continuously monitors the near-field environment, pays special attention to sudden obstacles such as suddenly appearing forklifts, responds quickly, transmits the sudden obstacle information to the UAV obstacle avoidance system, discovers and responds to near-field sudden obstacles in time to ensure flight safety; the rule engine processes the data transmitted by the ultrasonic array in real time. When an emergency obstacle is detected, an emergency avoidance action, such as lateral dodging, is triggered within 50 ms to achieve a quick response and avoid collision with sudden obstacles; the data of the millimeter-wave radar and the binocular camera are fused to obtain comprehensive obstacle information, and the deep reinforcement learning algorithm is used to generate an optimal obstacle avoidance path based on the fused data, dynamically plan the obstacle avoidance path, and ensure that the UAV safely bypasses the obstacles; receive the real-time environment data from the digital twin platform, combine the digital twin data, re-plan the long-term flight route, optimize the overall flight path, achieve global optimization, improve flight efficiency, and ensure long-term flight safety at the same time.

[0109] Further, the specific steps of the UAV flight trajectory analysis real-time optimization and adjustment method are as follows: If the UAV flight trajectory analysis real-time evaluation value is greater than or equal to the third determination value of the UAV flight trajectory analysis real-time evaluation value, there is no need to optimize and adjust the UAV flight trajectory analysis real-time; if the UAV flight trajectory analysis real-time evaluation value is lower than the third determination value of the UAV flight trajectory analysis real-time evaluation value, the surrounding obstacles are detected by the millimeter-wave radar and vision, the data is uploaded to the blockchain, and based on the pheromone concentration and real-time obstacle data, the weights of each flight segment are calculated, and the optimal path, such as the path with the smallest total weight, is output to bypass the mobile crane, select the container gap, and if the priority of the new task is higher than the current task, trigger path re-planning.

[0110] In this embodiment, the wind speed of the port environment is monitored in real time. When the wind speed exceeds the preset wind speed, the rotor folding mechanism is automatically triggered to initiate gliding and forced landing. Under high wind speed conditions, the resistance is reduced by rotor folding to ensure the safe landing of the UAV. Based on the microscale wind field prediction, the gliding trajectory is dynamically adjusted. The data of the port meteorological station is accessed in real time to predict the wind field changes in the next 30 seconds. According to the predicted wind field changes, the gliding path is adjusted in real time to optimize the gliding trajectory and improve the stability and safety during the forced landing process. An UAV swarm collaborative search algorithm based on pheromone decision-making divides the task process into three stages: the search stage, the communication stage, and the decision-making stage. The UAVs continuously move to update the pheromone map of their own aircraft. The pheromone map reflects the UAV's perception and exploration of the environment. The fusion of the multi-aircraft pheromone map is achieved through the communication network, that is, the UAVs exchange pheromone data with each other to obtain more comprehensive environmental information, make decisions based on local and global information, and use the grid pheromone concentration as a decision function to guide the position of the UAV. The area with high pheromone concentration is regarded as the area more likely to have targets or need further exploration. Mark fixed obstacles such as lighthouses and warehouses, and assign a constant weight to these fixed obstacles, which are permanently considered in the path planning to avoid collisions. Use dynamic pheromones to mark temporary obstacles such as mobile cranes and vehicles. The weights of these pheromones decay with time, reflecting the dynamic changes of temporary obstacles and making the path planning more flexible. When the UAV flies, it releases pheromones. The lower the difficulty of passing the path, the higher the pheromone concentration. The surrounding obstacles are detected through on-board sensors (such as millimeter-wave radar and vision sensors), and the difficulty of passing the path is reflected through the pheromone concentration, providing a basis for path planning. The detected obstacle data is uploaded to the blockchain, and the blockchain records the real-time obstacle data and pheromone concentration to ensure the security and immutability of the data and provide reliable data support for path planning. Based on the pheromone concentration and real-time obstacle data, the weights of each flight segment are calculated. The weight of the flight segment is usually jointly determined by the pheromone concentration and obstacle data. The higher the pheromone concentration, the greater the probability that the flight segment has been successfully passed, so a smaller value should be assigned in the weight calculation. The obstacle influence factor can be set according to factors such as the type, size, distance, and moving speed of the obstacle. For example, if the obstacle is large and close to the flight segment, the influence factor should be large, indicating a higher risk for this flight segment. The flight path with the minimum total weight is output as the optimal path, such as bypassing the mobile crane and choosing the gap between containers to achieve the optimal path planning, avoiding obstacles, and improving the flight efficiency. Monitor the priority of new tasks. If the priority of the new task is higher than the current task, trigger path replanning to ensure that high-priority tasks can be executed in a timely manner and improve the task response ability. Monitor the density of UAVs in a certain area. For example, when the density exceeds five or more, increase the pheromone evaporation rate in this area to guide subsequent UAVs to bypass the congested area. Avoid over-concentration of UAVs in a specific area, reduce the risk of collision, and improve the flight efficiency.

[0111] As shown Figure 2 in the figure, it is a schematic structural diagram of an unmanned aerial vehicle (UAV) flight trajectory analysis system for port logistics provided by an embodiment of the present application. The UAV flight trajectory analysis system for port logistics provided by the embodiment of the present application includes: a UAV flight trajectory analysis data acquisition and processing module, a UAV flight trajectory analysis data analysis module, a comprehensive analysis module, and an optimization and adjustment module. The UAV flight trajectory analysis data acquisition and processing module is used to collect and process UAV flight trajectory analysis data. The UAV flight trajectory analysis data analysis module is used to analyze the UAV flight trajectory analysis data to obtain a navigation accuracy evaluation value and a UAV dynamic obstacle avoidance evaluation value. The comprehensive analysis module is used to comprehensively analyze the navigation accuracy evaluation value and the UAV dynamic obstacle avoidance evaluation value to obtain a UAV flight trajectory analysis real-time evaluation value. The optimization and adjustment module is used to compare and analyze the navigation accuracy evaluation value, the UAV dynamic obstacle avoidance evaluation value, and the UAV flight trajectory analysis real-time evaluation value with the first determination value of the navigation accuracy evaluation value, the second determination value of the UAV dynamic obstacle avoidance evaluation value, and the third determination value of the UAV flight trajectory analysis real-time evaluation value to obtain a UAV flight trajectory analysis optimization and adjustment method.

[0112] The UAV provided by the embodiment of the present application includes: the UAV cargo hold adopts a carbon fiber honeycomb sandwich structure, is equipped with an adaptive electromagnetic suspension system, and adjusts the cargo hold damping in real time through electrorheological fluid; the UAV fuselage is sprayed with a polyurethane coating containing microcapsules; the UAV motor adopts a closed magnetic suspension bearing; the UAV belly is coated with a superhydrophobic material.

[0113] In this embodiment, the combination of the carbon fiber honeycomb sandwich structure and the self-healing nano-coating not only realizes lightweight but also enhances the adaptability to extreme environments. The closed magnetic suspension bearing motor solves the problem of salt spray corrosion and greatly improves the service life. The application of the hydrogen-electric hybrid power system in the UAV field significantly extends the endurance and reduces carbon emissions. The supercapacitor module supports instantaneous 20 kW discharge to meet the demand for sudden gust emergency climb. The "swarm transfer" mode based on blockchain decentralizes task allocation and optimizes the flight route with a group intelligence algorithm, improving the efficiency by 30%. The on-board customs intelligent lock and the "flight declaration channel" achieve second-level cross-border customs clearance.

[0114] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented 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.

[0115] The present invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each flow and / or block in the flowchart illustrations and / or block diagrams, and combinations of flows and / or blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions may 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 executed by the processor of the computer or other programmable data processing apparatus create means for implementing the functions specified in the flowchart flow or flows and / or block or blocks. Figure 1 in a flow or flows and / or block or blocks Figure 1 or means for implementing the functions specified in a block or blocks.

[0116] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means that implement the functions specified in a flow or flows and / or block or blocks. Figure 1 in a flow or flows and / or block or blocks Figure 1 or means for implementing the functions specified in a block or blocks.

[0117] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in a flow or flows and / or block or blocks. Figure 1 in a flow or flows and / or block or blocks Figure 1 or means for implementing the functions specified in a block or blocks.

[0118] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0119] It is obvious that those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A UAV flight trajectory analysis method for port logistics, characterized in that: The following steps are involved: Collect and process drone flight trajectory analysis data; Analyze the UAV flight trajectory analysis data to obtain the navigation accuracy evaluation value and the UAV dynamic obstacle avoidance evaluation value; The real-time evaluation value of the UAV flight trajectory analysis is obtained through comprehensive analysis of the navigation accuracy evaluation value and the UAV dynamic obstacle avoidance evaluation value; The navigation accuracy evaluation value, the UAV dynamic obstacle avoidance evaluation value and the UAV flight trajectory analysis real-time evaluation value are compared and analyzed with the first judgment value of the navigation accuracy evaluation value, the second judgment value of the UAV dynamic obstacle avoidance evaluation value and the third judgment value of the UAV flight trajectory analysis real-time evaluation value to obtain the UAV flight trajectory analysis optimization and adjustment method.

2. The method for analyzing the flight trajectory of unmanned aerial vehicles for port logistics according to claim 1, characterized in that: The specific steps of collecting and processing the UAV flight trajectory analysis data are as follows: Collect raw data of drone flight trajectory analysis through drone detection and positioning equipment; The original data of the UAV flight trajectory analysis is cleaned and denoised to obtain the UAV flight trajectory analysis data; By using a binocular camera to collect continuous image frames, a feature extraction algorithm is used to extract key points and their descriptors from the image, and the feature points between continuous frames are matched by a feature point matching algorithm. The position and posture of the drone are estimated by visual odometer technology, and the cumulative scale error caused by the feature point matching error is analyzed. The change of the scale error over time is recorded to form time series data. By fitting the time series data, the rate at which the scale error grows over time, that is, the scale drift rate, is calculated; The UAV flight trajectory analysis data includes navigation accuracy data and UAV dynamic obstacle avoidance data.

3. The method for analyzing the flight trajectory of unmanned aerial vehicles for port logistics according to claim 1, characterized in that: The specific steps of obtaining the navigation accuracy evaluation value are as follows: The navigation accuracy data includes heading angle error rate, scale drift rate, obstacle avoidance response time, actual position abscissa of the drone and actual position ordinate of the drone; Obtaining the heading angle error rate threshold, scale drift rate threshold, obstacle avoidance response time threshold, standard value of the horizontal coordinate of the UAV position, standard value of the vertical coordinate of the UAV position, weight factor of the heading angle error rate, weight factor of the UAV position error coefficient, weight factor of the scale drift rate, and weight factor of the obstacle avoidance response time from the UAV flight trajectory analysis database; The heading angle error rate threshold and the heading angle error rate ratio analysis results are averaged in different navigation accuracy detection sections and recorded as the first component of the navigation accuracy evaluation value; The actual position abscissa of the drone and the standard value of the abscissa of the drone position and the actual position ordinate of the drone and the standard value of the ordinate of the drone position are coupled to obtain the second component of the navigation accuracy assessment value; The analysis result of the scale drift rate threshold and the scale drift rate ratio is recorded as the third component of the navigation accuracy evaluation value; The analysis results of the ratio of obstacle avoidance response time to obstacle avoidance response time threshold are averaged in different navigation accuracy detection sections and recorded as the fourth component of the navigation accuracy evaluation value; Performing a ratio analysis on the coupling result of the first component of the navigation accuracy assessment value and the third component of the navigation accuracy assessment value and the coupling result of the second component of the navigation accuracy assessment value and the fourth component of the navigation accuracy assessment value to obtain a navigation accuracy assessment value; The navigation accuracy assessment value represents the quantitative data of the degree of closeness between the drone position information and the actual position, which is jointly represented by the first component of the navigation accuracy assessment value, the second component of the navigation accuracy assessment value, the third component of the navigation accuracy assessment value and the fourth component of the navigation accuracy assessment value.

4. The method for analyzing the flight trajectory of unmanned aerial vehicles for port logistics according to claim 1, characterized in that: The specific steps of obtaining the dynamic obstacle avoidance evaluation value of the drone are as follows: The UAV dynamic obstacle avoidance data includes obstacle avoidance response time, path replanning response time, scale drift rate, actual horizontal coordinate of the UAV obstacle avoidance action, and actual vertical coordinate of the UAV obstacle avoidance action; Obtain the obstacle avoidance response time threshold, path replanning response time threshold, scale drift rate threshold, standard value of the horizontal coordinate of the obstacle avoidance action of the UAV, standard value of the vertical coordinate of the obstacle avoidance action of the UAV, weight factor of the obstacle avoidance response time, weight factor of the path replanning response time, weight factor of the scale drift rate and weight factor of the dynamic trajectory deviation coefficient from the UAV flight trajectory analysis database; The analysis results of the ratio of obstacle avoidance response time to obstacle avoidance response time threshold are averaged in different UAV dynamic obstacle avoidance detection sections and recorded as the first component of the UAV dynamic obstacle avoidance evaluation value; The analysis results of the ratio of the path replanning response time to the path replanning response time threshold are averaged in different UAV dynamic obstacle avoidance detection sections and recorded as the second component of the UAV dynamic obstacle avoidance evaluation value; The analysis result of the ratio of the scale drift rate to the scale drift rate threshold is recorded as the third component of the UAV dynamic obstacle avoidance evaluation value; The actual horizontal coordinate of the drone's obstacle avoidance action and the standard value of the horizontal coordinate of the drone's obstacle avoidance action, as well as the actual vertical coordinate of the drone's obstacle avoidance action and the standard value of the vertical coordinate of the drone's obstacle avoidance action, are processed to obtain the fourth component of the drone's dynamic obstacle avoidance evaluation value; The first component of the UAV dynamic obstacle avoidance evaluation value, the second component of the UAV dynamic obstacle avoidance evaluation value, the third component of the UAV dynamic obstacle avoidance evaluation value and the fourth component of the UAV dynamic obstacle avoidance evaluation value are coupled to obtain the UAV dynamic obstacle avoidance evaluation value; The UAV dynamic obstacle avoidance evaluation value represents the quantitative data of the UAV dynamic obstacle avoidance evaluation value first component, the UAV dynamic obstacle avoidance evaluation value second component, the UAV dynamic obstacle avoidance evaluation value third component and the UAV dynamic obstacle avoidance evaluation value fourth component on the ability to avoid obstacles in a complex and dynamic environment.

5. The method for analyzing the flight trajectory of unmanned aerial vehicles for port logistics as claimed in claim 1, characterized in that: The specific steps of obtaining the real-time evaluation value of the UAV flight trajectory analysis through comprehensive analysis are as follows: The UAV dynamic obstacle avoidance data includes an obstacle data update frequency threshold, a navigation precision accuracy evaluation value weighting factor, an obstacle data update frequency weighting factor, and a UAV dynamic obstacle avoidance evaluation value weighting factor; The navigation accuracy evaluation value is averaged at different navigation accuracy detection points and recorded as the first component of the real-time evaluation value of the UAV flight trajectory analysis; The analysis results of the proportion of obstacle data update frequency and obstacle data update frequency threshold are averaged at different UAV flight trajectory analysis real-time detection points, and recorded as the second component of the UAV flight trajectory analysis real-time evaluation value; The UAV dynamic obstacle avoidance evaluation value is averaged at different UAV dynamic obstacle avoidance detection points and recorded as the third component of the UAV flight trajectory analysis real-time evaluation value; The coupling result of the first component of the UAV flight trajectory analysis real-time evaluation value and the second component of the UAV flight trajectory analysis real-time evaluation value is analyzed with the third component of the UAV flight trajectory analysis real-time evaluation value to obtain the UAV flight trajectory analysis real-time evaluation value; The real-time evaluation value of the UAV flight trajectory analysis represents the quantification degree of the real-time data processing of the UAV during the flight by the first component of the UAV flight trajectory analysis real-time evaluation value, the second component of the UAV flight trajectory analysis real-time evaluation value and the third component of the UAV flight trajectory analysis real-time evaluation value.

6. The method for analyzing the flight trajectory of unmanned aerial vehicles for port logistics as claimed in claim 1, characterized in that: The specific steps of obtaining the method for analyzing and optimizing the flight trajectory of the unmanned aerial vehicle are as follows: Obtaining a first determination value of the navigation accuracy evaluation value, a second determination value of the drone dynamic obstacle avoidance evaluation value, and a third determination value of the drone flight trajectory analysis real-time evaluation value from a drone flight trajectory analysis database; The UAV flight trajectory analysis optimization and adjustment method includes a navigation accuracy optimization and adjustment method, a UAV dynamic obstacle avoidance optimization and adjustment method, and a UAV flight trajectory analysis real-time optimization and adjustment method; If the navigation accuracy evaluation value is greater than or equal to the first navigation accuracy evaluation value, then there is no need to optimize and adjust the navigation accuracy; If the navigation accuracy assessment value is lower than the first judgment value of the navigation accuracy assessment value, the visual SLAM data is fused through Kalman filtering to suppress the angle error; it is directly connected to the port BIM system to obtain the coordinates of dynamic obstacle cranes and containers in real time, generate a 4D flight corridor, and the digital twin platform receives the crane positioning data and dynamically updates the flight restricted area. The drone adjusts the path according to the real-time corridor to avoid the signal blocking area.

7. The method for analyzing the flight trajectory of unmanned aerial vehicles for port logistics as claimed in claim 6, characterized in that: The specific steps of the unmanned aerial vehicle dynamic obstacle avoidance optimization adjustment method are as follows: If the UAV dynamic obstacle avoidance evaluation value is lower than or equal to the second determination value of the UAV dynamic obstacle avoidance evaluation value, then there is no need to optimize and adjust the UAV dynamic obstacle avoidance; If the UAV dynamic obstacle avoidance evaluation value is greater than the second determination value of the UAV dynamic obstacle avoidance evaluation value, the ultrasonic data is processed based on the rule engine to trigger emergency avoidance; radar and visual data are integrated to generate the optimal path using deep reinforcement learning.

8. The method for analyzing the flight trajectory of unmanned aerial vehicles for port logistics as claimed in claim 6, characterized in that: The specific steps of the real-time optimization and adjustment method for UAV flight trajectory analysis are as follows: If the UAV flight trajectory analysis real-time evaluation value is greater than or equal to the UAV flight trajectory analysis real-time evaluation value third determination value, then there is no need to optimize and adjust the UAV flight trajectory analysis real-time performance; If the real-time evaluation value of the drone flight trajectory analysis is lower than the third judgment value of the real-time evaluation value of the drone flight trajectory analysis, the surrounding obstacles are detected through millimeter wave radar and vision, and the data is uploaded to the blockchain. Based on the pheromone concentration and real-time obstacle data, the weight of each flight segment is calculated, and the optimal path is output, such as the path with the smallest total weight, bypassing the mobile crane, and selecting the container gap. If the new task priority is higher than the current task, the path re-planning is triggered.

9. UAV flight trajectory analysis system for port logistics, characterized by: It includes UAV flight trajectory analysis data acquisition and processing module, UAV flight trajectory analysis data analysis module, comprehensive analysis module and optimization adjustment module: UAV flight trajectory analysis data acquisition and processing module: used to collect and process UAV flight trajectory analysis data; UAV flight trajectory analysis data analysis module: used to analyze the UAV flight trajectory analysis data, obtain the navigation accuracy evaluation value and the UAV dynamic obstacle avoidance evaluation value; Comprehensive analysis module: used to obtain the real-time evaluation value of the UAV flight trajectory analysis through comprehensive analysis of the navigation accuracy evaluation value and the UAV dynamic obstacle avoidance evaluation value; Optimization and adjustment module: used to compare and analyze the navigation accuracy evaluation value, the UAV dynamic obstacle avoidance evaluation value and the UAV flight trajectory analysis real-time evaluation value with the first judgment value of the navigation accuracy evaluation value, the second judgment value of the UAV dynamic obstacle avoidance evaluation value and the third judgment value of the UAV flight trajectory analysis real-time evaluation value, and obtain the UAV flight trajectory analysis optimization adjustment method.

10. Use of a drone as claimed in any one of claims 1 to 9, characterized in that: include: The drone cargo compartment adopts a carbon fiber honeycomb sandwich structure and is equipped with an adaptive electromagnetic suspension system, which uses electrorheological fluid to adjust the cargo compartment damping in real time; The drone fuselage is sprayed with a polyurethane coating containing microcapsules; The drone motor uses a closed magnetic bearing; The belly of the drone is coated with super-hydrophobic material.

Images