Unmanned aerial vehicle path optimization method and device

Abstract

The invention discloses an unmanned aerial vehicle path optimization method and device, and relates to the technical field of path optimization. The method comprises the steps of determining at least one target monitoring point from at least two first monitoring points based on first information of a first area; the first information represents priorities of all first monitoring points in the first area; the priority of each first monitoring point is determined according to the risk coefficient of the geological disaster at the corresponding position; and configuring a surveying and mapping path of the unmanned aerial vehicle based on a genetic algorithm and the at least one target monitoring point to obtain a target surveying and mapping path. According to the technical scheme provided by the invention, the accuracy and rationality of the path planning result of the unmanned aerial vehicle are improved, so that the cruising efficiency of the unmanned aerial vehicle can be improved, and the surveying and mapping data processing amount is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of path optimization, and specifically to a method and device for optimizing the path of an unmanned aerial vehicle. Background Art

[0002] The drone-based surveying and mapping method has become one of the important surveying and mapping methods in modern terrain surveying and mapping. When a drone performs surveying and mapping operations, it needs to fly along a preset path and perform tasks such as image acquisition, temperature and humidity information acquisition at the monitoring points configured on the path, so as to analyze whether the terrain at each monitoring point is abnormal. In the prior art, the route of the drone patrol needs to be set according to the location of the monitoring point and the location of the obstacle, that is, to ensure that obstacles are avoided while passing through the monitoring point; however, different monitoring points present different risks in different surveying and mapping tasks and at different time periods, and their surveying and mapping requirements are also different. Each surveying and mapping path passes through all monitoring points, which will result in the need to pass through many relatively stable monitoring points in the surveying and mapping path planning, resulting in a longer drone patrol time and huge data processing pressure.

[0003] Therefore, how to overcome the above-mentioned technical problems and defects becomes a key issue that needs to be solved. Summary of the invention

[0004] The purpose of this application is to provide a UAV path optimization method and device to solve the problems raised in the above-mentioned background technology, improve the accuracy and rationality of the UAV path planning results, thereby improving the UAV cruising efficiency and reducing the amount of surveying and mapping data processing.

[0005] The technical solution of the embodiment of the present application is implemented as follows: The present application provides a method for optimizing a drone path, the method comprising: Based on the first information of the first area, at least one target monitoring point is determined from at least two first monitoring points; the first information represents the priority of all first monitoring points in the first area; the priority of each first monitoring point is determined according to the risk factor of geological disasters occurring at the corresponding location; Based on a genetic algorithm (GA) and the at least one target monitoring point, a mapping path of the UAV is configured to obtain a target mapping path.

[0006] In the above solution, the step of determining at least one target monitoring point from at least two first monitoring points based on the first information of the first area includes: Determining a priority of each of the at least two first monitoring points based on the first information of the first area; In order of priority, the first N first monitoring points are selected from the at least two first monitoring points as target monitoring points; N is an integer greater than half the number of the first monitoring points.

[0007] In the above solution, selecting the first N first monitoring points from the at least two first monitoring points as target monitoring points in order of priority includes: Based on the level of the current surveying and mapping task, a target monitoring strategy for the first area at the current moment is selected from at least one preset monitoring strategy; different monitoring strategies have different requirements for the risk coefficient of the monitoring point; Based on the target monitoring strategy, selecting at least two second monitoring points from the at least two first monitoring points; Based on the priority of each second monitoring point, the first N target monitoring points are selected from the at least two second monitoring points.

[0008] In the above solution, before determining at least one target monitoring point from at least two first monitoring points based on the first information of the first area, the method further includes: determining the first information of the first area.

[0009] In the above solution, the first information of the first area is determined, including: Based on the second information of each first monitoring point, the risk of geological disasters occurring at each first monitoring point is determined to obtain a second coefficient; the second information represents the change of sensor data at the first monitoring point; Determining a risk coefficient of each monitoring point based on a first coefficient and a corresponding second coefficient of each first monitoring point; the first coefficient is pre-configured according to the environmental complexity of the first monitoring point; All first monitoring points are sorted according to the risk coefficient, the priority of each first monitoring point is determined, and the first information of the first area is obtained.

[0010] In the above solution, configuring the mapping path of the UAV based on the GA algorithm and the at least one target monitoring point to obtain the target mapping path includes: Dividing the first area into grids to obtain at least one first grid; Selecting a grid corresponding to the position of each target monitoring point from the at least one first grid and marking it as a second grid; Selecting a grid corresponding to the obstacle point position from the at least one first grid and marking it as a third grid; Based on the second grid and the third grid, a target mapping path is configured using a GA algorithm.

[0011] In the above solution, configuring the target mapping path based on the second grid and the third grid using the GA algorithm includes: Based on the second grid and the third grid, an initialization population is generated by using a GA algorithm; the initialization population includes a plurality of individuals, and different individuals correspond to different mapping paths; Determine a first sub-function and a second sub-function of the fitness function of each individual in the initialized population to obtain the fitness function of each individual; the first sub-function is determined according to the path length of the individual, and the second sub-function is determined according to the smoothness of the individual; Based on the fitness function of each individual, the initialization population is sequentially subjected to selection, crossover and mutation processing to obtain a target population; Based on the target population, a target mapping path is determined.

[0012] In the above scheme, the method further includes: Based on the grid distance difference between the third point and the first point in each of the three consecutive points, the second sub-function of each individual fitness function is determined; the three consecutive points represent three consecutive grids in the mapping path.

[0013] In the above solution, the second sub-function of each individual fitness function is determined based on the grid distance difference between the third point and the first point in each of the three consecutive points, including: Determine the angle type of each group of three consecutive points based on the grid distance difference between the third point and the first point in each group of three consecutive points; the angle type includes an acute angle, a right angle, an obtuse angle and a flat angle; The penalty mechanism and the angle type of each group of three consecutive points are used to determine the second sub-function of each individual fitness function.

[0014] The present application also provides a drone path optimization device, which includes: A first configuration unit is used to determine at least one target monitoring point from at least two first monitoring points based on first information of a first area; the first information represents the priority of all first monitoring points in the first area; the priority of each first monitoring point is determined according to the risk factor of geological disasters occurring at the corresponding location; The second configuration unit is used to configure the mapping path of the UAV based on the GA algorithm and the at least one target monitoring point to obtain the target mapping path.

[0015] The UAV path optimization method and device provided in the embodiment of the present application selects some monitoring points with higher geological disaster risks to perform surveying and mapping tasks according to the priority of the monitoring locations. This not only reduces the amount of data collected and processed during the surveying and mapping process while ensuring the accuracy of surveying and mapping, but also improves the accuracy and scientificity of the surveying and mapping path planning results, and reduces the consumption of surveying and mapping resources while ensuring the integrity of the surveying and mapping data. Furthermore, since the number of surveying and mapping points is reduced, the surveying and mapping distance can be reduced, thereby improving the surveying and mapping efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a method for optimizing a drone path provided in an embodiment of the present application; Figure 2 This is a schematic diagram of landslide statistical data in the drone path optimization method of the embodiment of the present application; Figure 3 A schematic diagram of a process of determining at least one target monitoring point from at least two first monitoring points in a method for optimizing a drone path according to an embodiment of the present application; Figure 4 This is a schematic diagram of a first information flow for determining a first area in the drone path optimization method according to an embodiment of the present application; Figure 5 A schematic diagram of a process for obtaining a target mapping path in the UAV path optimization method of an embodiment of the present application; Figure 6 This is a schematic diagram of a grid surface after the first area is gridded in the drone path optimization method of an embodiment of the present application; Figure 7 This is a schematic diagram of the target mapping path flow based on the second grid, the third grid and the GA algorithm in the drone path optimization method of the embodiment of the present application; Figure 8 A schematic diagram of a random path in the drone path optimization method of an embodiment of the present application; Fig. 9 This is another schematic diagram of a random path in the drone path optimization method of an embodiment of the present application; Fig.10 This is a schematic diagram of the statistical results of the surveying and mapping path after the GA algorithm optimization iteration in the UAV path optimization method of the embodiment of the present application; Fig.11 A schematic diagram of the structure of a drone path optimization device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] The present application embodiment provides a method for optimizing the path of a drone, which is applied to electronic devices, and can be specifically applied to ground control terminals, portable terminals, vehicle-mounted terminals, cloud service platforms, etc. Figure 1As shown, the method may include S101 and S102. S101 and S102 are described in detail below in conjunction with specific embodiments.

[0019] S101: Based on first information of a first area, determine at least one target monitoring point from at least two first monitoring points; the first information represents the priority of all first monitoring points in the first area; the priority of each first monitoring point is determined according to the risk factor of geological disasters occurring at the corresponding location.

[0020] In actual application, the first area can be understood as the area where surveying and mapping tasks need to be performed in the current patrol cycle, and can also be called a surveying and mapping area.

[0021] In the prior art, the boundaries of plots and obstacles in the surveying area are generally confirmed by the longitude and latitude of the map itself, and then the specific surveying location, i.e., the monitoring point, is determined. After the surveying points in the surveying area are determined, each time the surveying task is performed, the drone will pass through all the monitoring points to perform corresponding operations such as image acquisition. However, for the same monitoring point, its geological disaster risk factor may be different in different seasons. During the period when the terrain morphology of the monitoring point is relatively stable, the necessity of its monitoring will be greatly reduced. For example, monitoring point A is a high-risk area for landslides determined based on continuous remote sensing data analysis. Figure 2 The landslide statistics of monitoring point A in the previous year are shown. Figure 2 It can be seen that landslides mostly occur in the rainy season, while in winter, the probability of landslides is almost close to 0; therefore, if the monitoring point A is still surveyed and mapped at the original frequency in winter, it will cause a waste of surveying and mapping resources and increase the processing volume of surveying and mapping data.

[0022] Therefore, in order to reduce the waste of surveying and mapping resources and the amount of data processing while ensuring the reliability of surveying and mapping results, the corresponding surveying and mapping locations can be flexibly configured according to the terrain change characteristics of each monitoring point in different time periods.

[0023] Based on this, in one embodiment, if Figure 3 As shown, the step of determining at least one target monitoring point from at least two first monitoring points based on the first information of the first area may include S301 and S302. S301 and S302 are further described in detail below in conjunction with specific embodiments.

[0024] S301: Determine a priority of each first monitoring point among at least two first monitoring points based on first information of a first area.

[0025] In actual application, the first monitoring point can be understood as a pre-set original monitoring point, which can be determined based on historical remote sensing data or calibrated manually; the target monitoring point can be understood as a monitoring point that needs to be passed in an actual surveying and mapping task.

[0026] In actual application, the risk factor of geological disasters such as landslides at each monitoring point in the current time period can be determined based on the sensor detection data at each monitoring point, and the priority of the monitoring point can be defined based on the risk factor.

[0027] Based on this, in one embodiment, before determining at least one target monitoring point from at least two first monitoring points based on the first information of the first area, the method may further include determining the first information of the first area.

[0028] In one embodiment, if Figure 4 As shown, the first information for determining the first area may include S401 to S403. S401 to S403 are described in detail below.

[0029] S401: Based on the second information of each first monitoring point, determine the risk of geological disasters at each first monitoring point to obtain a second coefficient; the second information represents the change of sensor data at the first monitoring point.

[0030] In actual application, the second coefficient may also be called a second risk coefficient, which represents an actual risk coefficient calculated according to the sensor.

[0031] In actual application, the second information can be determined based on the sensor data within a preset time period, for example, the sensor data of the previous three months; here, the sensor data can be in the form of inclination sensors, displacement sensors, etc.; if there is a long-term slow displacement, it may be a sign of an early stage of a landslide, and if there is a sudden accelerated displacement, it may mean that a landslide is about to occur. By analyzing the displacement time series of the monitoring points, including changes in displacement, displacement velocity and acceleration, it is possible to determine whether there is a displacement trend and displacement velocity at the corresponding position, thereby quantifying the risk factor of a landslide at each monitoring point based on the sensor data of all monitoring points.

[0032] Exemplarily, five monitoring points A1 to A5 are set to collect data on rainfall, soil moisture, terrain slope, and vegetation coverage for the previous three months; The statistical results are shown in Table 1:

[0033] According to the collected data, the change rate of each parameter is calculated, and each change rate is weighted to obtain the risk value; the risk value calculation formula is as follows: ; in, represents the rainfall change rate, represents the soil moisture change rate, represents the vegetation cover change rate, Indicates the terrain slope; to represents weight; Then each risk value is normalized to obtain the actual risk coefficient of each monitoring point, that is, the second coefficient.

[0034] S402: Determine a risk coefficient of each monitoring point based on a first coefficient and a corresponding second coefficient of each first monitoring point; the first coefficient is pre-configured according to the environmental complexity of the first monitoring point.

[0035] In actual application, the first coefficient can also be called the configuration risk coefficient, which can be understood as a pre-configured fixed risk coefficient; the first coefficient can be defined by technical personnel according to the environmental complexity of the monitoring point, that is, the more complex the environment, the greater the risk of geological disasters at the monitoring point. At the same time, the degree of influence of the monitoring point on the monitoring results is higher, that is, the higher its importance to the overall monitoring results; the value of the first coefficient can be configured to be greater than 0 and less than 1.

[0036] In actual application, the environmental characteristics of each monitoring point (such as whether geological disasters have occurred, terrain complexity, etc.) can be used as evaluation conditions to evaluate its environmental complexity, and then the first coefficient can be quantified according to the environmental complexity; specifically, an environmental complexity evaluation system can be established, and each evaluation condition can be quantitatively assigned a value, and then the final environmental complexity can be calculated based on the score of each evaluation condition.

[0037] For example, the evaluation conditions of the environmental complexity of monitoring point A3 are shown in Table 2:

[0038] In actual application, after determining the environmental complexity, the configuration risk coefficient, that is, the first coefficient, can be determined based on the ratio of the environmental complexity score to the total score.

[0039] For example, the calculation formula of the configuration risk coefficient is expressed as: ; in, represents the configuration risk coefficient, i.e. the first coefficient, represents the environmental complexity score, Represents the total score of environmental complexity; according to the calculation formula and the evaluation results in Table 2, it can be calculated that the configuration risk coefficient of monitoring point A3 is 0.24.

[0040] In actual application, after determining the first coefficient and the second coefficient, the two risk coefficients can be weighted to obtain the risk coefficient of the monitoring point; wherein the sum of the weight coefficients of the first coefficient and the second coefficient is 1; here, the first coefficient is a static score, and the second coefficient is a dynamic score. When weighting, weights can be allocated according to the relative importance of geological attributes and real-time sensor data in risk assessment; specifically, when geological attributes are considered to be more important than real-time data, the weight of the first coefficient can be set greater than the weight of the second coefficient. Correspondingly, when real-time sensor data is considered to be more important than geological attributes, the weight of the second coefficient can be set greater than the weight of the first coefficient. Here, how to allocate weights can be determined based on expert knowledge or historical data analysis.

[0041] For example, the calculation formula of the risk factor can be expressed as: ; in, represents the risk factor, and is the weight; S403: Sort all first monitoring points according to risk coefficients, determine the priority of each first monitoring point, and obtain first information of the first area.

[0042] In actual application, after calculating the risk factor, you can sort it from high to low, and then define the priority according to the sorting order. The higher the score, the higher the priority.

[0043] After determining the priority of each monitoring point, the target monitoring point may be determined, ie, S302 is executed.

[0044] S302: Selecting the first N first monitoring points from the at least two first monitoring points as target monitoring points in order of priority; N is an integer greater than half the number of the first monitoring points.

[0045] Here, after determining the priority, the number of surveying and mapping points to be selected for this surveying and mapping task, that is, the number of target monitoring points, can be determined based on the number of monitoring points preset in the first area (i.e., the first monitoring points). For example, if the surveying and mapping area to be performed for the surveying and mapping task is 217 square kilometers and the preset number of surveying and mapping points is 137, then the number of surveying and mapping points actually selected cannot be less than 69.

[0046] In actual application, a coefficient threshold can be set in advance, and monitoring points with risk coefficients greater than the threshold can be selected as surveying points for this task; the coefficient threshold can be configured based on historical data or expert opinions.

[0047] In actual application, the configured coefficient threshold may include a lower threshold and an upper threshold. These two thresholds divide the monitoring points into three parts: low-risk monitoring points, medium-risk monitoring points and high-risk monitoring points. The monitoring points with risk coefficients lower than the lower threshold are low-risk monitoring points, the monitoring points with risk coefficients between the lower threshold and the upper threshold are medium-risk monitoring points, and the monitoring points with risk coefficients greater than the upper threshold are high-risk monitoring points. The level of the current surveying and mapping task can be determined according to the specific task content, and different monitoring point screening strategies can be customized for different levels, thereby reducing the amount of surveying and mapping tasks and data processing while ensuring that the surveying and mapping point selection results are complete and reliable and the surveying and mapping path planning results are accurate and reliable.

[0048] In actual application, the levels of surveying and mapping tasks performed are different, and the requirements for risk factors will also be different; therefore, different screening strategies can be customized according to different surveying and mapping tasks, so that target monitoring points can be selected according to the screening strategies to ensure the accuracy of the target monitoring point selection results, and then ensure the accuracy and reliability of the surveying and mapping results.

[0049] Based on this, in one embodiment, selecting the first N first monitoring points from the at least two first monitoring points as target monitoring points in order of priority may include: Based on the level of the current surveying and mapping task, a target monitoring strategy for the first area at the current moment is selected from at least one preset monitoring strategy; different monitoring strategies have different requirements for the risk coefficient of the monitoring point; Based on the target monitoring strategy, selecting at least two second monitoring points from the at least two first monitoring points; Based on the priority of each second monitoring point, the first N target monitoring points are selected from the at least two second monitoring points.

[0050] In actual application, the monitoring strategy may also be referred to as a screening strategy, or as a screening criterion, and the embodiments of the present application do not limit this, as long as its function can be achieved.

[0051] In actual application, surveying and mapping tasks can be divided into three levels: level one, level two and level three, and the surveying and mapping accuracy increases from level one to level three. Here, level one surveying and mapping tasks can also be called low-precision surveying and mapping tasks, level two surveying and mapping tasks can also be called medium-precision surveying and mapping tasks, and level three surveying and mapping tasks can also be called high-precision surveying and mapping tasks.

[0052] For example, the classification standards for surveying and mapping tasks are shown in Table 3:

[0053] In actual application, when the surveying and mapping task level is level one (low precision), data processing is relatively simple. At this time, the monitoring strategy is configured to use all first monitoring points as second monitoring points; when the surveying and mapping task level is level two, in order to balance the data processing volume and the reliability of the monitoring results, the monitoring strategy can be configured to use all high-risk monitoring points and medium-risk monitoring points among the first monitoring points as second monitoring points; when the surveying and mapping task level is level three, in order to reduce data processing pressure, all high-risk monitoring points among the first monitoring points can be used as second monitoring points.

[0054] S102: Based on the GA algorithm and the at least one target monitoring point, a mapping path of the UAV is configured to obtain a target mapping path.

[0055] In one embodiment, if Figure 5 As shown, configuring the mapping path of the UAV based on the GA algorithm and the at least one target monitoring point to obtain the target mapping path may include S501 to S504, and S501 to S504 are described in detail below.

[0056] S501: Divide the first area into grids to obtain at least one first grid.

[0057] In actual application, after the first area is divided into a plurality of first grids, all the first grids may be labeled with numbers.

[0058] S502: Select a grid corresponding to the position of each target monitoring point from the at least one first grid, and mark it as a second grid.

[0059] In actual application, an association table between the first grid and the coordinate position of the first area can be established first, and then according to the coordinates of the monitoring point, the grid corresponding to the monitoring point is determined in all grids and marked as the second grid. Figure 6 The mesh surface after meshing the first area is shown. Figure 6 As shown in the grids numbered 18, 36, 61 and 65; the second grid can also be called a surveying and mapping grid, or a patrol grid, which is not limited in the embodiment of the present application as long as its function can be achieved.

[0060] S503: Select a grid corresponding to the obstacle point position from the at least one first grid, and mark it as a third grid.

[0061] In actual application, in order to ensure the safety of the drone during the surveying process, it is necessary to avoid obstacles such as trees and mountains when specifying the surveying path. Therefore, in the path planning, it is necessary to mark the obstacle position, that is, the third grid, such as Figure 6As shown in the grids numbered 13, 14, 23, 24, 34, 37, 38, 47, 73, 76, 77 and 83; the marking method of the third grid refers to the marking method of the second grid; here, the third grid can also be called an obstacle grid, which is not limited in the embodiment of the present application as long as its function can be achieved.

[0062] S504: Based on the second grid and the third grid, configure a target mapping path using a GA algorithm.

[0063] The GA algorithm is a search optimization algorithm that simulates natural selection and genetic mechanisms. It finds the optimal solution or approximate optimal solution to a problem by simulating operations such as selection, crossover (recombination), and mutation in the biological evolution process. In the field of path planning, the GA algorithm can be used to optimize model parameters to better plan surveying and mapping paths.

[0064] In one embodiment, if Figure 7 As shown, configuring the target mapping path based on the second grid and the third grid using the GA algorithm may include S701 to S704. S701 to S704 are described in detail below.

[0065] S701: Based on the second grid and the third grid, generate an initialization population using a GA algorithm; the initialization population includes a plurality of individuals, and different individuals correspond to different mapping paths.

[0066] In practical applications, the GA algorithm is used to generate an initialization population, which can also be called the GA algorithm initialization process. During the initialization process, multiple feasible paths will be randomly generated, and each feasible path needs to pass through all second grids and avoid all third grids. An individual in the initialization population represents a feasible mapping path.

[0067] S702: Determine the first sub-function and the second sub-function of the fitness function of each individual in the initialized population to obtain the fitness function of each individual; the first sub-function is determined according to the path length of the individual, and the second sub-function is determined according to the smoothness of the individual.

[0068] In practical applications, the fitness function is generally divided into two parts, namely the first sub-function and the second sub-function; the first sub-function is used to determine the length of the path, and the second sub-function is used to determine the smoothness of the path; the first sub-function can also be called a length function, and the second sub-function can also be called a smoothness function.

[0069] In practical application, the goal of the first sub-function is to have the shortest path; the first sub-function can be determined according to the total length of each path; illustratively, the first sub-function can be expressed as: ; in, represents the first subfunction, Indicates i The total length of the paths.

[0070] In practical applications, the goal of the second sub-function is to achieve the highest smoothness; the smoother the path, the larger the angle formed by three consecutive grids, and the larger the angle, the larger the distance between the first and third points in the three consecutive points; therefore, the second part of the fitness function can be calculated based on the grid distance difference from the third point to the first point in all three consecutive points in the path.

[0071] Based on this, in one embodiment, the method may further include: determining a second sub-function of each individual fitness function based on a grid distance difference from the third point to the first point in each of the three consecutive points; the three consecutive points represent three consecutive grids in the mapping path.

[0072] In practical applications, the grid distance difference of each group of three consecutive points may correspond to different angles. For example, Figure 8 and Fig. 9 Two random paths are shown, Figure 8 The three grids numbered 26, 16 and 17 are a set of three consecutive points with a right angle, and the grids numbered 45, 36 and 26 are a set of three consecutive points with an obtuse angle. Fig. 9 The grids numbered 28, 18, and 29 are a set of three consecutive points with an acute angle. After calculation and comparison, Fig. 9 The distance from the third point to the first point among three consecutive points with an acute angle is the smallest; based on the goal of smoothness, a penalty can be imposed on non-ideal angle conditions, so that solutions with fewer non-ideal conditions can obtain higher fitness values, making them more likely to be retained in the selection process, thereby improving the accuracy of the optimization results.

[0073] Based on this, in one embodiment, determining the second sub-function of each individual fitness function based on the grid distance difference between the third point and the first point in each of the three consecutive points may include: Determine the angle type of each group of three consecutive points based on the grid distance difference between the third point and the first point in each group of three consecutive points; the angle type includes an acute angle, a right angle, an obtuse angle and a flat angle; The penalty mechanism and the angle type of each group of three consecutive points are used to determine the second sub-function of each individual fitness function.

[0074] In practical applications, when using the penalty mechanism and the angle type of each group of three consecutive points to determine the second sub-function of each individual fitness function, no penalty is given for the right angle case, a penalty of 5 is given for the obtuse angle case, a penalty of 30 is given for the right angle case, and an infinite penalty is given for the acute angle case, and the inverse of the sum of the penalties is used as the second part of the fitness function.

[0075] After determining the first part and the second part of the fitness function, the first sub-function and the second sub-function may be weighted to obtain the fitness function of the corresponding individual.

[0076] Exemplarily, the fitness function can be expressed as: ; in, Indicates i The fitness function of an individual is and Respectively represent i The first and second subfunctions of the individuals, and Represents weight.

[0077] S703: Based on the fitness function of each individual, the initialization population is sequentially subjected to selection, crossover and mutation processing to obtain a target population.

[0078] Here, the selection operation is performed first; after determining the fitness function, the sum of all individual fitness functions is calculated first, then the proportion of each individual is calculated, and then the parent individual is selected in a roulette manner based on the proportion of each individual.

[0079] Then, a crossover operation is performed; two individuals are randomly selected from the selected parent generation for crossover to generate new offspring; a new path, i.e., offspring individuals, can be obtained by exchanging the paths before and after the crossover point.

[0080] Then, a mutation operation is performed; specifically, the offspring are randomly mutated to introduce new changes and generate new individuals, which will replace the original individuals.

[0081] The new individuals after selection, crossover and mutation operations are combined with the old individuals to form a new population. At this point, one iteration process is completed; repeat the above iteration process until the predetermined number of iterations is reached or the fitness is no longer significantly improved, and the final generation population, that is, the target population, is obtained.

[0082] S704: Determine a target mapping path based on the target population.

[0083] In practical applications, the individual with the highest fitness can be selected from the target population as the optimal solution, and the path corresponding to this individual is the target mapping path.

[0084] For example, in the surveying task for area A with an area of ​​217 square kilometers, it is necessary to perform surveying tasks on 69 surveying points. The statistical results of the surveying path after the GA algorithm optimization iteration are as follows: Fig.10 shown; from Fig.10It can be seen that after 30 iterations, the path length is reduced from the initial 50 kilometers to 29.34 kilometers, effectively shortening the path length.

[0085] In summary, by selecting some monitoring points with higher geological disaster risks to perform surveying and mapping tasks according to the priority of the monitoring locations, it is possible to reduce the amount of data collected and processed during the surveying and mapping process while ensuring the accuracy of the surveying and mapping, and to improve the accuracy and scientificity of the surveying and mapping path planning results. On the basis of ensuring the integrity of the surveying and mapping data, it reduces the consumption of surveying and mapping resources. Furthermore, since the number of surveying and mapping points is reduced, the surveying and mapping distance can be reduced, thereby improving the surveying and mapping efficiency.

[0086] In order to implement the drone path optimization method of the present application, the present application embodiment also provides a drone path optimization device, such as Fig.11 As shown, the device may include: The first configuration unit 1101 is used to determine at least one target monitoring point from at least two first monitoring points based on first information of the first area; the first information represents the priority of all first monitoring points in the first area; the priority of each first monitoring point is determined according to the risk factor of geological disasters occurring at the corresponding location; The second configuration unit 1102 is used to configure the mapping path of the UAV based on the GA algorithm and the at least one target monitoring point to obtain the target mapping path.

[0087] In one embodiment, the first configuration unit 1101 may be specifically configured to: Determining a priority of each of the at least two first monitoring points based on the first information of the first area; In order of priority, the first N first monitoring points are selected from the at least two first monitoring points as target monitoring points; N is an integer greater than half the number of the first monitoring points.

[0088] In one embodiment, the first configuration unit 1101 may be specifically configured to: Based on the level of the current surveying and mapping task, a target monitoring strategy for the first area at the current moment is selected from at least one preset monitoring strategy; different monitoring strategies have different requirements for the risk coefficient of the monitoring point; Based on the target monitoring strategy, selecting at least two second monitoring points from the at least two first monitoring points; Based on the priority of each second monitoring point, the first N target monitoring points are selected from the at least two second monitoring points.

[0089] In one embodiment, the device may further include a processing unit, and the processing unit is configured to: determine first information of the first area.

[0090] In one embodiment, the processing unit may be specifically used for: Based on the second information of each first monitoring point, the risk of geological disasters occurring at each first monitoring point is determined to obtain a second coefficient; the second information represents the change of sensor data at the first monitoring point; Determining a risk coefficient of each monitoring point based on a first coefficient and a corresponding second coefficient of each first monitoring point; the first coefficient is pre-configured according to the environmental complexity of the first monitoring point; All first monitoring points are sorted according to the risk coefficient, the priority of each first monitoring point is determined, and the first information of the first area is obtained.

[0091] In one embodiment, the second configuration unit 1102 may be specifically configured to: Dividing the first area into grids to obtain at least one first grid; Selecting a grid corresponding to the position of each target monitoring point from the at least one first grid and marking it as a second grid; Selecting a grid corresponding to the obstacle point position from the at least one first grid and marking it as a third grid; Based on the second grid and the third grid, a target mapping path is configured using a genetic algorithm.

[0092] In one embodiment, the second configuration unit 1102 may be specifically configured to: Based on the second grid and the third grid, an initialization population is generated by using a genetic algorithm; the initialization population includes a plurality of individuals, and different individuals correspond to different mapping paths; Determine a first sub-function and a second sub-function of the fitness function of each individual in the initialized population to obtain the fitness function of each individual; the first sub-function is determined according to the path length of the individual, and the second sub-function is determined according to the smoothness of the individual; Based on the fitness function of each individual, the initialization population is sequentially subjected to selection, crossover and mutation processing to obtain a target population; Based on the target population, a target mapping path is determined.

[0093] In one embodiment, the processing unit may also be used for: Based on the grid distance difference between the third point and the first point in each of the three consecutive points, the second sub-function of each individual fitness function is determined; the three consecutive points represent three consecutive grids in the mapping path.

[0094] In one embodiment, the processing unit may be specifically used for: Determine the angle type of each group of three consecutive points based on the grid distance difference between the third point and the first point in each group of three consecutive points; the angle type includes an acute angle, a right angle, an obtuse angle and a flat angle; The penalty mechanism and the angle type of each group of three consecutive points are used to determine the second sub-function of each individual fitness function.

[0095] It should be noted that: the drone path optimization device provided in the above embodiment only uses the division of the above program modules as an example when performing drone-based mapping. In actual applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device can be divided into different program modules to complete all or part of the processing described above. In addition, the drone path optimization device provided in the above embodiment and the drone path optimization method embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0096] It should be noted that: "first", "second", etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0097] In addition, the technical solutions described in the embodiments of the present application can be combined arbitrarily without conflict.

[0098] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application.

Claims

1. A UAV path optimization method, characterized in that: The method comprises: Based on the first information of the first area, at least one target monitoring point is determined from at least two first monitoring points; the first information represents the priority of all first monitoring points in the first area; the priority of each first monitoring point is determined according to the risk factor of geological disasters occurring at the corresponding location; Based on the genetic algorithm and the at least one target monitoring point, a mapping path of the UAV is configured to obtain a target mapping path.

2. The method according to claim 1, characterized in that The determining at least one target monitoring point from at least two first monitoring points based on the first information of the first area includes: Determining a priority of each of the at least two first monitoring points based on the first information of the first area; In order of priority, the first N first monitoring points are selected from the at least two first monitoring points as target monitoring points; N is an integer greater than half the number of the first monitoring points.

3. The method according to claim 2, characterized in that The selecting, in order of priority, the first N first monitoring points from the at least two first monitoring points as target monitoring points comprises: Based on the level of the current surveying and mapping task, a target monitoring strategy for the first area at the current moment is selected from at least one preset monitoring strategy; different monitoring strategies have different requirements for the risk coefficient of the monitoring point; Based on the target monitoring strategy, selecting at least two second monitoring points from the at least two first monitoring points; Based on the priority of each second monitoring point, the first N target monitoring points are selected from the at least two second monitoring points.

4. The method according to claim 2, characterized in that: Before determining at least one target monitoring point from at least two first monitoring points based on the first information of the first area, the method further includes: determining the first information of the first area.

5. The method according to claim 4, characterized in that The determining of the first information of the first area includes: Based on the second information of each first monitoring point, the risk of geological disasters occurring at each first monitoring point is determined to obtain a second coefficient; the second information represents the change of sensor data at the first monitoring point; Determining a risk coefficient of each monitoring point based on a first coefficient and a corresponding second coefficient of each first monitoring point; the first coefficient is pre-configured according to the environmental complexity of the first monitoring point; All first monitoring points are sorted according to the risk coefficient, the priority of each first monitoring point is determined, and the first information of the first area is obtained.

6. The method according to any one of claims 1 to 5, characterized in that The configuring the mapping path of the UAV based on the genetic algorithm and the at least one target monitoring point to obtain the target mapping path includes: Dividing the first area into grids to obtain at least one first grid; Selecting a grid corresponding to the position of each target monitoring point from the at least one first grid and marking it as a second grid; Selecting a grid corresponding to the obstacle point position from the at least one first grid and marking it as a third grid; Based on the second grid and the third grid, a target mapping path is configured using a genetic algorithm.

7. The method according to claim 6, characterized in that The configuring the target mapping path by using a genetic algorithm based on the second grid and the third grid includes: Based on the second grid and the third grid, an initialization population is generated by using a genetic algorithm; the initialization population includes a plurality of individuals, and different individuals correspond to different mapping paths; Determine a first sub-function and a second sub-function of the fitness function of each individual in the initialized population to obtain the fitness function of each individual; the first sub-function is determined according to the path length of the individual, and the second sub-function is determined according to the smoothness of the individual; Based on the fitness function of each individual, the initialization population is sequentially subjected to selection, crossover and mutation processing to obtain a target population; Based on the target population, a target mapping path is determined.

8. The method according to claim 7, characterized in that The method further comprises: Based on the grid distance difference between the third point and the first point in each of the three consecutive points, the second sub-function of each individual fitness function is determined; the three consecutive points represent three consecutive grids in the mapping path.

9. The method according to claim 8, characterized in that The determining of the second sub-function of each individual fitness function based on the grid distance difference between the third point and the first point in each of the three consecutive points comprises: Determine the angle type of each group of three consecutive points based on the grid distance difference between the third point and the first point in each group of three consecutive points; the angle type includes an acute angle, a right angle, an obtuse angle and a flat angle; The penalty mechanism and the angle type of each group of three consecutive points are used to determine the second sub-function of each individual fitness function.

10. The UAV path optimization device is characterized by: The device comprises: A first configuration unit is used to determine at least one target monitoring point from at least two first monitoring points based on first information of a first area; the first information represents the priority of all first monitoring points in the first area; the priority of each first monitoring point is determined according to the risk factor of geological disasters occurring at the corresponding location; The second configuration unit is used to configure the mapping path of the UAV based on the genetic algorithm and the at least one target monitoring point to obtain the target mapping path.

Images