A method for optimizing the flight path of an unmanned aerial vehicle

Abstract

A method for optimizing the flight path of an unmanned aerial vehicle belongs to the technical field of unmanned aerial vehicle flight path planning. In order to improve the flight safety performance of the unmanned aerial vehicle, the present invention collects information data of surrounding objects, environmental factor data, and information data of the unmanned aerial vehicle during the operation of the unmanned aerial vehicle; divides the standardized data into three groups of factor data sets; constructs a method for calculating the minimum safe distance for unmanned aerial vehicle flight, and calculates the minimum safe flight distance of the unmanned aerial vehicle at different time points; constructs a prediction model for the minimum safe distance for unmanned aerial vehicle flight, calculates the minimum safe distance for unmanned aerial vehicle flight under different single environmental factors, and calculates the comprehensive minimum safe predicted distance for unmanned aerial vehicle flight; constructs a collision warning method based on the comprehensive minimum safe predicted distance for unmanned aerial vehicle flight, and warns when the distance between the unmanned aerial vehicle and the surrounding objects does not exceed the comprehensive minimum safe predicted distance; and designs and optimizes the flight path of the unmanned aerial vehicle. The present invention dynamically adjusts the flight path of the unmanned aerial vehicle to maintain a safe distance.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicle flight path planning, and in particular relates to a method for optimizing the flight path of an unmanned aerial vehicle. Background Art

[0002] When a drone executes a predetermined flight path, even if the path is designed accurately, it is difficult to fully predict all obstacles and changes encountered in actual flight. This includes design deviations, objects that suddenly appear, etc. As a result, drones that lack flight path optimization and adjustment methods face multiple risks and potential problems. First, the risk of collision increases significantly, and drones may collide with objects at design deviation locations, protruding buildings, and natural obstacles (such as trees and mountains). These collisions may not only cause damage to the drone, but also damage valuable cargo or sensitive equipment carried by the drone, and even cause third-party property losses or personal injuries. Secondly, operating costs will rise due to frequent repairs and replacement of parts, and the insurance cost of drones may also rise due to increased flight risks. In addition to direct financial losses, the lack of effective flight path optimization will also lead to an increase in the failure rate of drone missions, affecting service quality and customer satisfaction. In the long run, this will affect the market competitiveness and brand reputation of operators, and limit the application and development of drone technology in a wider range of fields.

[0003] The invention patent with application number 202111131532.5 and invention name “A method for optimizing the control path of a drone” proposes a method for optimizing the control path of a drone. The method first obtains the path nodes that the drone needs to pass through according to the actual flight mission, and performs real number encoding on the path formed by these nodes to generate a random path. Then, the fitness value of the path is calculated, and the optimal path is selected using a genetic algorithm.

[0004] The invention patent with application number 202111300583.6 and invention name "A delivery drone, a delivery drone early warning method and device" proposes a drone inspection path optimization method, which obtains the longitude and latitude data of the monitoring points on the inspection track through a high-precision map, and uses the ant colony algorithm to solve the shortest inspection path. Through iterative optimization and adaptive pheromone concentration update mechanism, the path is continuously improved until the maximum number of iterations is reached, thereby determining the optimal inspection path.

[0005] However, existing studies have not fully considered the changes that may be encountered in actual flight, such as design deviations, sudden weather changes, unforeseen obstacles or GPS signal deviations. These unforeseen factors may cause the drone to deviate from the planned path when performing a flight mission, causing the drone to be too close to obstacles or ground objects, thereby increasing the risk of collision. Summary of the invention

[0006] The problem to be solved by the present invention is to improve the flight safety performance of a UAV and propose a method for optimizing the flight path of a UAV.

[0007] To achieve the above object, the present invention is implemented through the following technical solutions:

[0008] A method for optimizing a UAV flight path comprises the following steps:

[0009] S1. Collecting information data of surrounding objects, environmental factors, and drone information during the operation of the drone. The surrounding object information data includes object type, object height, and width. The environmental factors data include ambient wind speed, light intensity, and temperature. The drone information data includes drone length, drone width, and drone weight.

[0010] S2. The data collected in step S1 is standardized, and the standardized data is divided into three groups of factor data sets to obtain a data set for temperature analysis, a data set for light intensity analysis, and a data set for ambient wind speed analysis;

[0011] S3. Construct a method to calculate the minimum safe flight distance of UAVs, and calculate the minimum safe flight distance of UAVs at different time points;

[0012] S4. Construct a prediction model for the minimum safe distance of UAV flight based on different single environmental factors, calculate the minimum safe distance of UAV flight under different single environmental factors, and then calculate the comprehensive minimum safe prediction distance of UAV flight;

[0013] S5. Construct a collision warning method based on the comprehensive minimum safe predicted distance of UAV flight, and issue a warning when the distance between the UAV and surrounding objects does not exceed the comprehensive minimum safe predicted distance;

[0014] S6. Based on the warning information obtained in step S5, the flight path of the UAV is optimized.

[0015] Furthermore, in step S1, data on the height, width, ambient wind speed, light intensity, temperature, drone length, drone width, and drone weight of the object at the same time point are collected, and the selected time point is set as t 1, t 2…, you …, tn ,in, you For the i A point in time, n is the total number of time points.

[0016] Furthermore, in step S2, the standardized data are divided into three groups of factor data sets, to obtain a data set for temperature analysis, a data set for light intensity analysis, and a data set for ambient wind speed analysis;

[0017] The factors in the data set for temperature analysis include object type, object height, object width, ambient wind speed, light intensity, drone length, drone width, and drone weight;

[0018] The factors in the data set for analyzing light intensity include object type, object height, object width, ambient wind speed, temperature, drone length, drone width, and drone weight;

[0019] The factors in the data set for environmental wind speed analysis include object type, object height, object width, light intensity, temperature, drone length, drone width, and drone weight.

[0020] Furthermore, the specific implementation method of step S3 includes the following steps:

[0021] S3.1. According to i Time point you , the corresponding object type, object height, width, ambient wind speed, light intensity, temperature, drone length, drone width, and drone weight data are used to establish a simulated drone flight field. In the simulated drone flight field, the distance between the drone and the object is set to , m Refers to the total number of distances between the drone and the object when it is set to fly. Point to the i Time point you, No. j The distance between the drone and the object when flying;

[0022] S3.2. The height fluctuation amplitude of the UAV is measured by the airborne inertial measurement unit. The flight distance of the drone from detecting obstacles to responding using the vehicle-mounted GNSS navigation system is The noise generated by the drone is measured using a sound level meter placed in the surrounding environment. ,in, , , Respectively refer to i Time point you, No. j The distance between the drone and the object during flight The measured height fluctuation amplitude, the distance the drone flies from the time an obstacle is detected to the time it reacts, and the noise generated by the drone;

[0023] S3.3. Based on , , , calculate the comprehensive evaluation index value of UAV safe flight , the calculation formula is:

[0024] ;

[0025] in, , , They are , , The corresponding weight value;

[0026] On this basis, the distance between the drone and the object during flight is calculated. Corresponding comprehensive evaluation index value of UAV safe flight ;

[0027] S3.4. Use the expert scoring method to set the threshold of comprehensive evaluation indicators for safe flight of drones , compare in turn and The size relationship of All greater than The comprehensive evaluation index value of the safe flight of the UAV is the condition, and then the obstacle corresponding to the minimum value that meets the condition is selected until the UAV flight distance when the reaction is made is the minimum safe flight distance of the UAV, which is recorded as ;

[0028] S3.5. The calculated time point is t 1, t 2…, you …, tn The corresponding minimum safe flight distance of the UAV is .

[0029] Furthermore, the specific implementation method of step S4 includes the following steps:

[0030] Furthermore, the specific implementation method of step S4 includes the following steps:

[0031] S4.1. A deep learning model is used to build a temperature-based prediction model for the minimum safe distance of drone flight. The specific structure is as follows:

[0032] Input layer: Set 8 input nodes, corresponding to the object type, object height, width, ambient wind speed, light intensity, drone length, drone width, and drone weight at each time point after data standardization;

[0033] Hidden Layer:

[0034] The first hidden layer: consists of 128 neurons and uses the ReLU activation function;

[0035] The second hidden layer consists of 64 neurons and uses the ReLU activation function.

[0036] The third hidden layer: consists of 32 neurons and uses the ReLU activation function;

[0037] The fourth hidden layer: consists of 16 neurons and uses the ReLU activation function;

[0038] Output layer: set 1 neuron corresponding to the minimum safe flight distance of the drone at each time point;

[0039] The object type, object height, width, ambient wind speed, light intensity, drone length, drone width, drone weight, and drone minimum safe flight distance corresponding to each time point are input into the temperature-based drone minimum safe flight distance prediction model for training, and the square of the error between the drone minimum safe flight distance predicted by the temperature-based drone minimum safe flight distance prediction model and the calculated drone minimum safe flight distance is used as the loss function. , the expression is:

[0040] ;

[0041] in, The minimum safe flight distance of a UAV predicted by the temperature-based prediction model for the minimum safe flight distance of a UAV;

[0042] The error back propagation method is used to optimize the parameters of the temperature-based UAV flight minimum safety distance prediction model, and the expression is:

[0043] :

[0044] in, are model parameters, is the learning rate, is the loss function with respect to the parameter The gradient of Represents the update of model parameters;

[0045] After optimization, the temperature-based prediction model for the minimum safe distance of UAV flight is obtained. P 1;

[0046] S4.2. A deep learning model is used to build a prediction model for the minimum safe distance of drone flight based on light intensity. The specific structure is as follows:

[0047] Input layer: Set 8 input nodes, corresponding to the object type, object height, width, ambient wind speed, temperature, drone length, drone width, and drone weight at each time point after data standardization;

[0048] Hidden Layer:

[0049] The first hidden layer: consists of 128 neurons and uses the ReLU activation function;

[0050] The second hidden layer consists of 64 neurons and uses the ReLU activation function.

[0051] The third hidden layer: consists of 32 neurons and uses the ReLU activation function;

[0052] The fourth hidden layer: consists of 16 neurons and uses the ReLU activation function;

[0053] Output layer: set 1 neuron corresponding to the minimum safe flight distance of the drone at each time point;

[0054] The object type, object height, width, ambient wind speed, temperature, drone length, drone width, drone weight, and drone minimum safe flight distance corresponding to each time point are input into the drone minimum safe flight distance prediction model based on light intensity for training, and the square of the error between the drone minimum safe flight distance predicted by the drone minimum safe flight distance prediction model based on light intensity and the calculated safety distance is used as the loss function. , the expression is:

[0055] ;

[0056] in, The minimum safe flight distance of a UAV predicted by the prediction model of the minimum safe flight distance of a UAV based on light intensity;

[0057] The error back propagation method is used to optimize the parameters of the model, and after optimization, the minimum safe distance prediction model for UAV flight based on light intensity is obtained. P 2;

[0058] S4.3. A deep learning model is used to build a prediction model for the minimum safe distance of UAV flight based on ambient wind speed. The specific structure is as follows:

[0059] Input layer: Set 8 input nodes, corresponding to the object type, object height, width, temperature, light intensity, drone length, drone width, and drone weight at each time point after data standardization;

[0060] Hidden Layer:

[0061] The first hidden layer: consists of 128 neurons and uses the ReLU activation function;

[0062] The second hidden layer consists of 64 neurons and uses the ReLU activation function.

[0063] The third hidden layer: consists of 32 neurons and uses the ReLU activation function;

[0064] The fourth hidden layer: consists of 16 neurons and uses the ReLU activation function;

[0065] Output layer: set 1 neuron corresponding to the minimum safe flight distance of the drone at each time point;

[0066] The object type, object height, width, ambient wind speed, temperature, drone length, drone width, drone weight, and drone minimum safe flight distance corresponding to each time point are input into the drone minimum safe flight distance prediction model based on ambient wind speed for training, and the square of the error between the drone minimum safe flight distance predicted by the drone minimum safe flight distance prediction model based on ambient wind speed and the calculated safety distance is used as the loss function , the expression is:

[0067] ;

[0068] in, The minimum safe flight distance of a UAV predicted by the prediction model of the minimum safe flight distance of a UAV based on the ambient wind speed;

[0069] The error back propagation method is used to optimize the parameters of the model, and after optimization, the minimum safe distance prediction model for UAV flight based on ambient wind speed is obtained. P 3;

[0070] S4.4. Based on P 1. P 2. P 3. Predict the minimum safe distance for drone flight;

[0071] By inputting the drone flight time, the corresponding object type, object height, width, ambient wind speed, light intensity, drone length, drone width, and drone weight data are P 1, and then predict P 1 The minimum safe predicted distance for drone flight ;

[0072] By inputting the drone flight time, the corresponding object type, object height, width, ambient wind speed, temperature, drone length, drone width, and drone weight data are P 2, and then predict P 2. The minimum safe predicted distance for drone flight ;

[0073] By inputting the drone flight time, the corresponding object type, object height, width, light intensity, temperature, drone length, drone width, and drone weight data into P3, the prediction is obtained. P 3. The minimum safe predicted distance for drone flight ;

[0074] S4.5. Calculate the minimum safe predicted distance for UAV flight , the expression is:

[0075] ;

[0076] in, , , They are , , The weight coefficient of .

[0077] Furthermore, the specific implementation method of step S5 includes the following steps:

[0078] S5.1. Using airborne laser radar, the drone collects the three-dimensional coordinates of surrounding objects while flying. , For surrounding objects x Axis coordinates, For surrounding objects y Axis coordinates, For surrounding objects z Axis coordinates;

[0079] S5.2. Use the drone-mounted GNSS system to collect the drone’s three-dimensional coordinates , Space for drones x Axis coordinates, Space for drones y Axis coordinates, Space for drones z Axis coordinates;

[0080] S5.3. From the three-dimensional coordinates of surrounding objects , filtering and the spatial 3D coordinates of the drone Highly similar point cloud data, get the retained point cloud data The specific screening method is to traverse all three-dimensional coordinates ,Compare and Relationship:

[0081] when When , the corresponding point cloud data is retained;

[0082] when , delete the corresponding point cloud data;

[0083] in, is the height deviation threshold;

[0084] S5.4. From the retained point cloud data, arbitrarily select a point in the retained point cloud data Then, by calculating and comparing the points around this point with The distance between the points, select the points adjacent to the T Nearest Neighbor ;

[0085] Point-based and its adjacent T nearest neighbor points, fitting a plane PM, the corresponding plane equation is:

[0086] ;

[0087] in, , , refers to the component of the plane normal vector, , , Refers to the coordinates of any point in the retained point cloud data of the surrounding objects. is a constant;

[0088] based on and its adjacent T Nearest Neighbor , using the least squares method, we get , , , , and then determine the plane equation;

[0089] S5.5. Calculate the three-dimensional coordinates of the drone The distance Ds from the plane PM is calculated as:

[0090] ;

[0091] S5.6. Construct collision warning conditions and compare Ds with the minimum safe predicted distance of the UAV flight obtained in step S5 The relationship between:

[0092] when When , it indicates that the distance between the drone and the surrounding objects exceeds the comprehensive minimum safety prediction distance, and no warning is required;

[0093] when , it indicates that the distance between the drone and surrounding objects does not exceed the comprehensive minimum safety prediction distance, and an early warning is required.

[0094] Furthermore, the specific implementation method of step S6 includes the following steps:

[0095] S6.1. When the warning information is obtained based on step S5, a vertical line is drawn through the three-dimensional coordinates of the UAV to the fitting plane PM, and the intersection point of the vertical line and the plane PM is set as , the expression is as follows:

[0096] ;

[0097] ;

[0098] ;

[0099] in, is the intersection of the vertical line and plane PM x Axis coordinates, is the intersection of the vertical line and plane PM y Axis coordinates, is the intersection of the vertical line and plane PM z Axis coordinates, calculated The specific location of the coordinates;

[0100] S6.2. Construction arrive Vector , the expression is:

[0101] ;

[0102] Set the adjusted drone space position coordinates to ,based on , , If three points are on the same straight line, then The expression of the position coordinates is:

[0103] ;

[0104] in, is the scaling factor;

[0105] get The coordinates of ;

[0106] S6.3. After setting the drone and The distance between them is Dsn, and the relationship equation is constructed to obtain the expression:

[0107] ;

[0108] ;

[0109] Solving the equation based on the given Dsn, we get , then get Coordinates ;

[0110] S6.4. Based on the adjusted spatial position coordinates of the UAV, N points to adjust, N When selecting points, the flight speed of the drone is considered, and the expression is:

[0111] ;

[0112] in, is the flight speed of the drone; Adjust the flight time of the segment for the drone at the location. To adjust the distance between adjacent points;

[0113] S6.5. Based on the UAV's heading obtained in step S6.4 N points to adjust, among which the Then the first me The coordinates of the points are , after adjustment me The coordinates of the points are , me for N Any one of;

[0114] Based on arrive Vector , establish the adjusted me The coordinate relationship of the points is expressed as:

[0115] ;

[0116] ;

[0117] in, is the weight function, To adjust the scope of influence of the control; N After adjusting each point, the position optimization of the points near the UAV warning position is completed, and the UAV flight route is optimized.

[0118] Beneficial effects of the present invention:

[0119] The invention discloses a method for optimizing the flight path of a UAV, proposes a method for comprehensively predicting the safe flight distance of a UAV, considers surrounding object information, environmental factors, UAV information, etc., establishes a UAV flight safe distance prediction model, and realizes the prediction of the minimum safe distance of a UAV flying in different environments;

[0120] The invention discloses a method for optimizing the flight path of an unmanned aerial vehicle (UAV), and proposes a collision warning method based on a comprehensive minimum safe predicted distance for UAV flight. By comparing the measured distance between the UAV and surrounding objects with the predicted minimum safe distance, the collision warning for UAV flight is realized.

[0121] The present invention discloses a method for optimizing the flight path of a UAV, which dynamically adjusts the flight path of the UAV to maintain a safe distance, thereby directly reducing the risk of collision with surrounding objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0122] Figure 1 The present invention is a flowchart of a method for optimizing the flight path of a UAV. DETAILED DESCRIPTION

[0123] In order to make the purpose, technical solution and advantages of the present invention more clear, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the specific embodiments described are only part of the embodiments of the present invention, rather than all of the specific embodiments. The components of the specific embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations, and the present invention can also have other embodiments.

[0124] Therefore, the following detailed description of the specific embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents the selected specific embodiments of the present invention. Based on the specific embodiments of the present invention, all other specific embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0125] In order to further understand the content, features and effects of the present invention, the following specific implementation methods are given as examples, and the attached Figure 1 The detailed instructions are as follows:

[0126] Embodiment 1:

[0127] A method for optimizing a UAV flight path comprises the following steps:

[0128] S1. Collecting information data of surrounding objects, environmental factors, and drone information during the operation of the drone. The surrounding object information data includes object type, object height, and width. The environmental factors data include ambient wind speed, light intensity, and temperature. The drone information data includes drone length, drone width, and drone weight.

[0129] Furthermore, in step S1, data on the height, width, ambient wind speed, light intensity, temperature, drone length, drone width, and drone weight of the object at the same time point are collected, and the selected time point is set as t 1, t 2…, you …, tn ,in, you For the i A point in time, n is the total number of time points.

[0130] Furthermore, data such as object type, object height, and width are obtained from traffic management departments and traffic infrastructure inspection departments, environmental wind speed, light intensity, and temperature are obtained from meteorological management departments, and data such as drone length, drone width, and drone weight are obtained from drone manufacturers and supplier websites. Among them, the object types mainly consider four categories: buildings, vegetation, traffic infrastructure, and traffic ancillary facilities. The corresponding numbers for buildings, vegetation, traffic infrastructure, and traffic ancillary facilities are 0.25, 0.50, 0.75, and 1.00, respectively.

[0131] S2. The data collected in step S1 is standardized, and the standardized data is divided into three groups of factor data sets to obtain a data set for temperature analysis, a data set for light intensity analysis, and a data set for ambient wind speed analysis;

[0132] Furthermore, in step S2, the standardized data are divided into three groups of factor data sets, to obtain a data set for temperature analysis, a data set for light intensity analysis, and a data set for ambient wind speed analysis;

[0133] The factors in the data set for temperature analysis include object type, object height, object width, ambient wind speed, light intensity, drone length, drone width, and drone weight;

[0134] The factors in the data set for analyzing light intensity include object type, object height, object width, ambient wind speed, temperature, drone length, drone width, and drone weight;

[0135] The factors in the data set for environmental wind speed analysis include object type, object height, object width, light intensity, temperature, drone length, drone width, and drone weight.

[0136] S3. Construct a method to calculate the minimum safe flight distance of UAVs, and calculate the minimum safe flight distance of UAVs at different time points;

[0137] Furthermore, the specific implementation method of step S3 includes the following steps:

[0138] S3.1. According to i Time point you , the corresponding object type, object height, width, ambient wind speed, light intensity, temperature, drone length, drone width, and drone weight data are used to establish a simulated drone flight field. In the simulated drone flight field, the distance between the drone and the object is set to , m Refers to the total number of distances between the drone and the object when it is set to fly. Point to the i Time point you, No. j The distance between the drone and the object when flying;

[0139] S3.2. The height fluctuation amplitude of the UAV is measured by the airborne inertial measurement unit. The flight distance of the drone from detecting obstacles to responding using the vehicle-mounted GNSS navigation system is The noise generated by the drone is measured using a sound level meter placed in the surrounding environment. ,in, , , Respectively refer to i Time point you, No. j The distance between the drone and the object during flight The measured height fluctuation amplitude, the distance the drone flies from the time an obstacle is detected to the time it reacts, and the noise generated by the drone;

[0140] S3.3. Based on , , , calculate the comprehensive evaluation index value of UAV safe flight , the calculation formula is:

[0141] ;

[0142] in, , , They are , , The corresponding weight value;

[0143] On this basis, the distance between the drone and the object during flight is calculated. Corresponding comprehensive evaluation index value of UAV safe flight ;

[0144] S3.4. Use the expert scoring method to set the threshold of comprehensive evaluation indicators for safe flight of drones , compare in turn and The size relationship of All greater than The comprehensive evaluation index value of the safe flight of the UAV is the condition, and then the obstacle corresponding to the minimum value that meets the condition is selected until the UAV flight distance when the reaction is made is the minimum safe flight distance of the UAV, which is recorded as ;

[0145] S3.5. The calculated time point is t 1, t 2…, you …, tn The corresponding minimum safe flight distance of the UAV is .

[0146] S4. Construct a prediction model for the minimum safe distance of UAV flight based on different single environmental factors, calculate the minimum safe distance of UAV flight under different single environmental factors, and then calculate the comprehensive minimum safe prediction distance of UAV flight;

[0147] Furthermore, the specific implementation method of step S4 includes the following steps:

[0148] S4.1. A deep learning model is used to build a temperature-based prediction model for the minimum safe distance of drone flight. The specific structure is as follows:

[0149] Input layer: Set 8 input nodes, corresponding to the object type, object height, width, ambient wind speed, light intensity, drone length, drone width, and drone weight at each time point after data standardization;

[0150] Hidden Layer:

[0151] The first hidden layer: consists of 128 neurons and uses the ReLU activation function;

[0152] The second hidden layer consists of 64 neurons and uses the ReLU activation function.

[0153] The third hidden layer: consists of 32 neurons and uses the ReLU activation function;

[0154] The fourth hidden layer: consists of 16 neurons and uses the ReLU activation function;

[0155] Output layer: set 1 neuron corresponding to the minimum safe flight distance of the drone at each time point;

[0156] The object type, object height, width, ambient wind speed, light intensity, drone length, drone width, drone weight, and drone minimum safe flight distance corresponding to each time point are input into the temperature-based drone minimum safe flight distance prediction model for training, and the square of the error between the drone minimum safe flight distance predicted by the temperature-based drone minimum safe flight distance prediction model and the calculated drone minimum safe flight distance is used as the loss function. , the expression is:

[0157] ;

[0158] in, The minimum safe flight distance of a UAV predicted by the temperature-based prediction model for the minimum safe flight distance of a UAV;

[0159] The error back propagation method is used to optimize the parameters of the temperature-based UAV flight minimum safety distance prediction model, and the expression is:

[0160] :

[0161] in, are model parameters, is the learning rate, is the loss function with respect to the parameter The gradient of Represents the update of model parameters;

[0162] After optimization, the temperature-based prediction model for the minimum safe distance of UAV flight is obtained. P 1;

[0163] S4.2. A deep learning model is used to build a prediction model for the minimum safe distance of drone flight based on light intensity. The specific structure is as follows:

[0164] Input layer: Set 8 input nodes, corresponding to the object type, object height, width, ambient wind speed, temperature, drone length, drone width, and drone weight at each time point after data standardization;

[0165] Hidden Layer:

[0166] The first hidden layer: consists of 128 neurons and uses the ReLU activation function;

[0167] The second hidden layer consists of 64 neurons and uses the ReLU activation function.

[0168] The third hidden layer: consists of 32 neurons and uses the ReLU activation function;

[0169] The fourth hidden layer: consists of 16 neurons and uses the ReLU activation function;

[0170] Output layer: set 1 neuron corresponding to the minimum safe flight distance of the drone at each time point;

[0171] The object type, object height, width, ambient wind speed, temperature, drone length, drone width, drone weight, and drone minimum safe flight distance corresponding to each time point are input into the drone minimum safe flight distance prediction model based on light intensity for training, and the square of the error between the drone minimum safe flight distance predicted by the drone minimum safe flight distance prediction model based on light intensity and the calculated safety distance is used as the loss function. , the expression is:

[0172] ;

[0173] in, The minimum safe flight distance of a UAV predicted by the prediction model of the minimum safe flight distance of a UAV based on light intensity;

[0174] The error back propagation method is used to optimize the parameters of the model, and after optimization, the minimum safe distance prediction model for UAV flight based on light intensity is obtained. P 2;

[0175] S4.3. A deep learning model is used to build a prediction model for the minimum safe distance of UAV flight based on ambient wind speed. The specific structure is as follows:

[0176] Input layer: Set 8 input nodes, corresponding to the object type, object height, width, temperature, light intensity, drone length, drone width, and drone weight at each time point after data standardization;

[0177] Hidden Layer:

[0178] The first hidden layer: consists of 128 neurons and uses the ReLU activation function;

[0179] The second hidden layer consists of 64 neurons and uses the ReLU activation function.

[0180] The third hidden layer: consists of 32 neurons and uses the ReLU activation function;

[0181] The fourth hidden layer: consists of 16 neurons and uses the ReLU activation function;

[0182] Output layer: set 1 neuron corresponding to the minimum safe flight distance of the drone at each time point;

[0183] The object type, object height, width, ambient wind speed, temperature, drone length, drone width, drone weight, and drone minimum safe flight distance corresponding to each time point are input into the drone minimum safe flight distance prediction model based on ambient wind speed for training, and the square of the error between the drone minimum safe flight distance predicted by the drone minimum safe flight distance prediction model based on ambient wind speed and the calculated safety distance is used as the loss function , the expression is:

[0184] ;

[0185] in, The minimum safe flight distance of a UAV predicted by the prediction model of the minimum safe flight distance of a UAV based on the ambient wind speed;

[0186] The error back propagation method is used to optimize the parameters of the model, and after optimization, the minimum safe distance prediction model for UAV flight based on ambient wind speed is obtained. P 3;

[0187] S4.4. Based on P 1. P 2. P 3. Predict the minimum safe distance for drone flight;

[0188] By inputting the drone flight time, the corresponding object type, object height, width, ambient wind speed, light intensity, drone length, drone width, and drone weight data are P 1, and then predict P 1 The minimum safe predicted distance for drone flight ;

[0189] By inputting the drone flight time, the corresponding object type, object height, width, ambient wind speed, temperature, drone length, drone width, and drone weight data are P 2, and then predict P 2. The minimum safe predicted distance for drone flight ;

[0190] By inputting the drone flight time, the corresponding object type, object height, width, light intensity, temperature, drone length, drone width, and drone weight data into P3, the prediction is obtained. P 3. The minimum safe predicted distance for drone flight ;

[0191] S4.5. Calculate the minimum safe predicted distance for UAV flight , the expression is:

[0192] ;

[0193] in, , , They are , , The weight coefficient of .

[0194] S5. Construct a collision warning method based on the comprehensive minimum safe predicted distance of UAV flight, and issue a warning when the distance between the UAV and surrounding objects does not exceed the comprehensive minimum safe predicted distance;

[0195] Furthermore, the specific implementation method of step S5 includes the following steps:

[0196] S5.1. Using airborne laser radar, the drone collects the three-dimensional coordinates of surrounding objects while flying. , For surrounding objects x Axis coordinates, For surrounding objects y Axis coordinates, For surrounding objects z Axis coordinates;

[0197] S5.2. Use the drone-mounted GNSS system to collect the drone’s three-dimensional coordinates , Space for drones x Axis coordinates, Space for drones y Axis coordinates, Space for drones z Axis coordinates;

[0198] S5.3. From the three-dimensional coordinates of surrounding objects , filter and the spatial 3D coordinates of the drone Highly similar point cloud data, get the retained point cloud data The specific screening method is to traverse all three-dimensional coordinates ,Compare and Relationship:

[0199] when When , the corresponding point cloud data is retained;

[0200] when , delete the corresponding point cloud data;

[0201] in, is the height deviation threshold;

[0202] S5.4. From the retained point cloud data, arbitrarily select a point in the retained point cloud data Then, by calculating and comparing the points around this point with The distance between the points, select the points adjacent to the T Nearest Neighbor ;

[0203] Point-based and its adjacent T nearest neighbor points, fitting a plane PM, the corresponding plane equation is:

[0204] ;

[0205] in, , , refers to the component of the plane normal vector, , , Refers to the coordinates of any point in the retained point cloud data of the surrounding objects. is a constant;

[0206] based on and its adjacent T Nearest Neighbor , using the least squares method, we get , , , , and then determine the plane equation;

[0207] S5.5. Calculate the three-dimensional coordinates of the drone The distance Ds from the plane PM is calculated as:

[0208] ;

[0209] S5.6. Construct collision warning conditions and compare Ds with the minimum safe predicted distance of the UAV flight obtained in step S5 The relationship between:

[0210] when When , it indicates that the distance between the drone and the surrounding objects exceeds the comprehensive minimum safety prediction distance, and no warning is required;

[0211] when , it indicates that the distance between the drone and surrounding objects does not exceed the comprehensive minimum safety prediction distance, and an early warning is required.

[0212] S6. Based on the warning information obtained in step S5, the flight path of the UAV is optimized.

[0213] Furthermore, the specific implementation method of step S6 includes the following steps:

[0214] S6.1. When the warning information is obtained based on step S5, a vertical line is drawn through the three-dimensional coordinates of the UAV to the fitting plane PM, and the intersection point of the vertical line and the plane PM is set as , the expression is as follows:

[0215] ;

[0216] ;

[0217] ;

[0218] in, is the intersection of the vertical line and plane PM x Axis coordinates, is the intersection of the vertical line and plane PM y Axis coordinates, is the intersection of the vertical line and plane PM z Axis coordinates, calculated The specific location of the coordinates;

[0219] S6.2. Construction arrive Vector , the expression is:

[0220] ;

[0221] Set the adjusted drone space position coordinates to ,based on , , If three points are on the same straight line, then The expression of the position coordinates is:

[0222] ;

[0223] in, is the scaling factor;

[0224] get The coordinates of ;

[0225] S6.3. After setting the drone and The distance between them is Dsn, and the relationship equation is constructed to obtain the expression:

[0226] ;

[0227] ;

[0228] Solving the equation based on the given Dsn, we get , then get Coordinates ;

[0229] S6.4. Based on the adjusted spatial position coordinates of the UAV, N points to adjust, N When selecting points, the flight speed of the drone is considered, and the expression is:

[0230] ;

[0231] in, is the flight speed of the drone; Adjust the flight time of the segment for the drone at the location. To adjust the distance between adjacent points;

[0232] S6.5. Based on the UAV's heading obtained in step S6.4 N points to adjust, among which the Then the first me The coordinates of the points are , after adjustment me The coordinates of the points are , me for N Any one of;

[0233] Based on arrive Vector , establish the adjusted me The coordinate relationship of the points is expressed as:

[0234] ;

[0235] ;

[0236] in, is the weight function, To adjust the scope of influence of the control; N After adjusting each point, the position optimization of the points near the UAV warning position is completed, and the UAV flight route is optimized.

[0237] The key points and intended protection points of the present invention are:

[0238] (1) A comprehensive method for predicting the safe flight distance of UAVs.

[0239] (2) A collision warning method based on the comprehensive minimum safe prediction distance of UAV flight.

[0240] (3) A method for optimizing UAV flight paths.

[0241] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0242] Although the present application has been described above with reference to specific embodiments, various modifications may be made thereto and parts thereof may be replaced with equivalents without departing from the scope of the present application. In particular, as long as there is no structural conflict, the various features in the specific embodiments disclosed in the present application may be used in combination with each other in any manner, and the fact that these combinations are not exhaustively described in this specification is only for the sake of omitting space and saving resources. Therefore, the present application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for optimizing the flight path of an unmanned aerial vehicle, characterized in that: The steps include: S1. Collecting surrounding object information data, environmental factor data, and drone information data during the operation of the drone, the surrounding object information data includes object type, object height, and width, the environmental factor data includes ambient wind speed, light intensity, and temperature, and the drone information data includes drone length, drone width, and drone weight; S2. Standardize the data collected in step S1, and divide the standardized data into three groups of factor data sets to obtain a data set for temperature analysis, a data set for light intensity analysis, and a data set for ambient wind speed analysis; S3. Construct a method for calculating the minimum safe flight distance of a UAV and calculate the minimum safe flight distance of a UAV at different time points; The specific implementation method of step S3 includes the following steps: S3.

1. According to the object type, object height, width, ambient wind speed, light intensity, temperature, drone length, drone width, and drone weight data corresponding to the i-th time point ti, a simulated drone flight field is established. In the simulated drone flight field, the distance between the drone and the object during flight is set to m refers to the total distance between the drone and the object when it is flying. It refers to the distance between the jth UAV and the object at the i-th time point ti; S3.

2. The height fluctuation amplitude of the UAV is measured by the airborne inertial measurement unit. The flight distance of the drone from detecting obstacles to responding using the vehicle-mounted GNSS navigation system is The noise generated by the drone is measured using a sound level meter placed in the surrounding environment. in, They refer to the distance between the i-th time point ti and the j-th UAV and the object when flying. The measured height fluctuation amplitude, the distance the drone flies from the time an obstacle is detected to the time it reacts, and the noise generated by the drone; S3.

3. Based on Calculate the comprehensive evaluation index value of UAV safe flight The calculation formula is: Among them, α1, α2, and α3 are The corresponding weight value; On this basis, the distance between the drone and the object during flight is calculated. Corresponding comprehensive evaluation index value of UAV safe flight S3.

4. Use expert scoring method to set the threshold of comprehensive evaluation index for safe flight of drones Compare and The size relationship of All greater than The comprehensive evaluation index value of the safe flight of the UAV is the condition, and then the obstacle corresponding to the minimum value that meets the condition is selected until the UAV flight distance when the reaction is made is the minimum safe flight distance of the UAV, which is recorded as S3.

5. Calculate the minimum safe flight distance of the drone corresponding to the time points t1, t2…, ti…, tn: S4. Construct a prediction model for the minimum safe distance of UAV flight based on different single environmental factors, calculate the minimum safe distance of UAV flight under different single environmental factors, and then calculate the comprehensive minimum safe predicted distance of UAV flight; S5. Construct a collision warning method based on the comprehensive minimum safe predicted distance of UAV flight, and issue a warning when the distance between the UAV and surrounding objects does not exceed the comprehensive minimum safe predicted distance; S6. Based on the warning information obtained in step S5, the flight path of the UAV is optimized.

2. The method for optimizing the flight path of an unmanned aerial vehicle according to claim 1, characterized in that: In step S1, data on object height, width, ambient wind speed, light intensity, temperature, drone length, drone width, and drone weight at the same time point are collected, and the selected time points are set as t1, t2…, ti…, tn, where ti is the i-th time point and n is the total number of time points.

3. The method for optimizing the flight path of an unmanned aerial vehicle according to claim 2, characterized in that: In step S2, the standardized data are divided into three groups of factor data sets, to obtain a data set for temperature analysis, a data set for light intensity analysis, and a data set for ambient wind speed analysis; The factors in the data set for temperature analysis include object type, object height, object width, ambient wind speed, light intensity, drone length, drone width, and drone weight; The factors in the data set for analyzing light intensity include object type, object height, object width, ambient wind speed, temperature, drone length, drone width, and drone weight; The factors in the data set for environmental wind speed analysis include object type, object height, object width, light intensity, temperature, drone length, drone width, and drone weight.

4. The method for optimizing the flight path of an unmanned aerial vehicle according to claim 3, characterized in that: The specific implementation method of step S4 includes the following steps: S4.

1. A deep learning model is used to build a temperature-based prediction model for the minimum safe distance of drone flight. The specific structure is as follows: Input layer: Set 8 input nodes, corresponding to the object type, object height, width, ambient wind speed, light intensity, drone length, drone width, and drone weight at each time point after data standardization; Hidden Layer: The first hidden layer: consists of 128 neurons and uses the ReLU activation function; The second hidden layer consists of 64 neurons and uses the ReLU activation function. The third hidden layer: consists of 32 neurons and uses the ReLU activation function; The fourth hidden layer: consists of 16 neurons and uses the ReLU activation function; Output layer: set 1 neuron corresponding to the minimum safe flight distance of the drone at each time point; The object type, object height, width, ambient wind speed, light intensity, drone length, drone width, drone weight, and drone minimum safe flight distance corresponding to each time point are input into the temperature-based drone minimum safe flight distance prediction model for training, and the square of the error between the drone minimum safe flight distance predicted by the temperature-based drone minimum safe flight distance prediction model and the calculated drone minimum safe flight distance is used as the loss function Q1, which is expressed as: in, The minimum safe flight distance of a UAV predicted by the temperature-based prediction model for the minimum safe flight distance of a UAV; The error back propagation method is used to optimize the parameters of the temperature-based UAV flight minimum safety distance prediction model, and the expression is: Among them, θ is the model parameter, η is the learning rate, is the gradient of the loss function with respect to the parameter θ, ← represents the update of the model parameters; After optimization, the temperature-based prediction model P1 for the minimum safe distance of UAV flight was obtained; S4.

2. A deep learning model is used to build a prediction model for the minimum safe distance of drone flight based on light intensity. The specific structure is as follows: Input layer: Set 8 input nodes, corresponding to the object type, object height, width, ambient wind speed, temperature, drone length, drone width, and drone weight at each time point after data standardization; Hidden Layer: The first hidden layer: consists of 128 neurons and uses the ReLU activation function; The second hidden layer consists of 64 neurons and uses the ReLU activation function. The third hidden layer: consists of 32 neurons and uses the ReLU activation function; The fourth hidden layer: consists of 16 neurons and uses the ReLU activation function; Output layer: set 1 neuron corresponding to the minimum safe flight distance of the drone at each time point; The object type, object height, width, ambient wind speed, temperature, drone length, drone width, drone weight, and drone minimum safe flight distance corresponding to each time point are input into the drone minimum safe flight distance prediction model based on light intensity for training, and the square of the error between the drone minimum safe flight distance predicted by the drone minimum safe flight distance prediction model based on light intensity and the calculated safety distance is used as the loss function Q2, which is expressed as: in, The minimum safe flight distance of a UAV predicted by the prediction model of the minimum safe flight distance of a UAV based on light intensity; The error back propagation method is used to optimize the parameters of the model, and after optimization, the minimum safe distance prediction model P2 for UAV flight based on light intensity is obtained; S4.

3. A deep learning model is used to build a prediction model for the minimum safe distance of UAV flight based on ambient wind speed. The specific structure is as follows: Input layer: Set 8 input nodes, corresponding to the object type, object height, width, temperature, light intensity, drone length, drone width, and drone weight at each time point after data standardization; Hidden Layer: The first hidden layer: consists of 128 neurons and uses the ReLU activation function; The second hidden layer consists of 64 neurons and uses the ReLU activation function. The third hidden layer: consists of 32 neurons and uses the ReLU activation function; The fourth hidden layer: consists of 16 neurons and uses the ReLU activation function; Output layer: set 1 neuron corresponding to the minimum safe flight distance of the drone at each time point; The object type, object height, width, temperature, light intensity, drone length, drone width, drone weight, and drone minimum safe flight distance corresponding to each time point are input into the drone minimum safe flight distance prediction model based on ambient wind speed for training, and the square of the error between the drone minimum safe flight distance predicted by the drone minimum safe flight distance prediction model based on ambient wind speed and the calculated safety distance is used as the loss function Q3, which is expressed as: in, The minimum safe flight distance of a UAV predicted by the prediction model of the minimum safe flight distance of a UAV based on the ambient wind speed; The error back propagation method is used to optimize the parameters of the model, and after optimization, the minimum safe distance prediction model P3 for UAV flight based on ambient wind speed is obtained; S4.

4. Based on P1, P2 and P3, predict the minimum safe distance for UAV flight; By inputting the drone flight time, the corresponding object type, object height, width, ambient wind speed, light intensity, drone length, drone width, and drone weight data into P1, the minimum safe predicted distance M of the drone flight of P1 is predicted. min-P1 ; By inputting the drone flight time, the corresponding object type, object height, width, ambient wind speed, temperature, drone length, drone width, and drone weight data into P2, the minimum safe predicted distance M of the drone flight of P2 is predicted. min-P2 ; By inputting the drone flight time, the corresponding object type, object height, width, light intensity, temperature, drone length, drone width, and drone weight data into P3, the minimum safe predicted distance M of the drone flight of P3 is predicted. min-P3 ; S4.

5. Calculate the minimum safe predicted distance M for UAV flight min-R , the expression is: M min-R =β1M min-P1 +β2M min-P2 +β3M min-P3 Among them, β1, β2, and β3 are M min-P1 、M min-P2 、M min-P3 The weight coefficient of .

5. The method for optimizing the flight path of an unmanned aerial vehicle according to claim 4, characterized in that: The specific implementation method of step S5 includes the following steps: S5.

1. Using airborne laser radar, the drone collects the three-dimensional coordinates of surrounding objects while flying. zb (x zb ,y zb ,z zb ), x zb is the x-axis coordinate of the surrounding object, y zb is the y-axis coordinate of the surrounding object, z zb is the z-axis coordinate of the surrounding objects; S5.

2. Use the drone-mounted GNSS system to collect the drone’s three-dimensional coordinates S wr (x wr ,y wr ,z wr ), x wr is the spatial x-axis coordinate of the drone, y wr is the spatial y-axis coordinate of the drone, z wr is the spatial z-axis coordinate of the drone; S5.

3. From the three-dimensional coordinates S of the surrounding objects zb (x zb ,y zb ,z zb ), filter the spatial three-dimensional coordinates S of the drone wr (x wr ,y wr ,z wr ) point cloud data with high similarity, and obtain the retained point cloud data S zb-b (x zb-b ,y zb-b ,z zb-b ), the specific screening method is to traverse all three-dimensional coordinates S zb (x zb ,y zb ,z zb ), compare z zb With z wr Relationship: When z zb ≤|z wr -z tr |, keep the corresponding point cloud data; When z zb >|z wr -z tr |, delete the corresponding point cloud data; Among them, z tr is the height deviation threshold; S5.

4. From the retained point cloud data, arbitrarily select a point of the retained point cloud data Then, by calculating and comparing the points around this point with The distance between them, select the T nearest neighbor points adjacent to this point Point-based and its adjacent T nearest neighbor points, fitting a plane PM, the corresponding plane equation is: a1x zb-b +b1y zb-b +c1z zb-b +d1=0 Among them, a1, b1, c1 refer to the components of the plane normal vector, x zb-b ,y zb-b 、z zb-b Refers to the coordinates of any point in the retained point cloud data of the surrounding objects, d1 is a constant; based on and its T nearest neighbors The least square method is used to fit a1, b1, c1, and d1, and then the plane equation is determined; S5.

5. Calculate the three-dimensional spatial coordinates S of the drone wr (x wr ,y wr ,z wr ) and the distance Ds between the plane PM, calculated as: S5.

6. Construct collision warning conditions and compare Ds with the minimum safe predicted distance M of the UAV flight obtained in step S5. min-R The relationship between: When DS>M min-R When , it indicates that the distance between the drone and the surrounding objects exceeds the comprehensive minimum safety prediction distance, and no warning is required; When Ds≤M min-R , it indicates that the distance between the drone and surrounding objects does not exceed the comprehensive minimum safety prediction distance, and an early warning is required.

6. A method for optimizing the flight path of an unmanned aerial vehicle according to claim 5, characterized in that: The specific implementation method of step S6 includes the following steps: S6.

1. When the warning information is obtained based on step S5, a vertical line is drawn through the three-dimensional coordinates of the UAV to the fitting plane PM, and the intersection of the vertical line and the plane PM is set as S jd (x jd ,y jd ,z jd ), the expression is as follows: Among them, x jd is the x-axis coordinate of the intersection of the vertical line and plane PM, y jd is the y-axis coordinate of the intersection of the vertical line and plane PM, z jd is the z-axis coordinate of the intersection of the vertical line and plane PM, and S is calculated. jd The specific location of the coordinates; S6.

2. Construction of S wr To S jd Vector The expression is: Set the adjusted UAV space position coordinates to S wr-n (x wr-n ,y wr-n ,z wr-n ), based on S wr-n , S wr , S jd If the three points are on the same straight line, then S wr-n The expression of the position coordinates is: Where γ is the scaling factor; Get S wr-n The coordinates of (x wr +γ(x wr -x jd ),y wr +γ(y wr -y jd ),z wr +γ(z wr -z jd )); S6.

3. Set the drone to adjust S wr-n With S jd The distance between them is Dsn, and the relationship equation is constructed to obtain the expression: Dsn>M min-R Solve the equation based on the given Dsn, calculate γ, and then get S wr-n The coordinates S wr-n (x wr-n ,y wr-n ,z wr-n ); S6.

4. Based on the adjusted spatial position coordinates of the UAV, adjust the N points along the UAV's forward direction. When selecting the N points, consider the flight speed of the UAV, and the expression is: Where, v is the flight speed of the drone; t N The time for the drone to adjust the flight section at the location, L N To adjust the distance between adjacent points; S6.

5. Adjust the N points in the UAV's forward direction obtained in step S6.4, where the first point in S wr Then the coordinates of the mi-th point along the direction of the drone are S wr+mi (x wr+mi ,y wr+mi ,z wr+mi ), the coordinates of the adjusted point mi are S wr+mi-n (x wr+mi-n ,y wr+mi-n ,z wr+mi-n ), mi is any one of N; Based on S wr To S jd Vector Establish the coordinate relationship of the adjusted mi-th point, the expression is: Among them, λ is the weight function, and σ is the influence range of the adjustment control; after N points are adjusted, the position optimization of the points near the UAV warning position is completed, and the UAV flight route is optimized.

Images