Unmanned aerial vehicle route planning and pose calibration method and system

Abstract

The application provides a kind of unmanned aerial vehicle route planning and pose calibration method and system, belong to the unmanned aerial vehicle control technical field based on computer vision.In the process of unmanned aerial vehicle flight, every time a specified navigation point is reached, an orbit picture is obtained, the vertical distance of the location and the orbit line is identified and calculated, it is judged whether the unmanned aerial vehicle is within the error distance, that is, it is judged whether the unmanned aerial vehicle is yawed, if so, the unmanned aerial vehicle is corrected to the correct position, if not, it is judged whether the navigation point is the last navigation point, if not the last navigation point, the unmanned aerial vehicle continues to fly to the next navigation point, if the last navigation point, the unmanned aerial vehicle stops advancing and reaches the end point.The application optimizes orbit identification and improves orbit identification accuracy;By orbit line, the unmanned aerial vehicle route is drawn, and external factors such as map and train timetable are considered, the optimal unmanned aerial vehicle route is drawn under the condition of ensuring the minimum cost;The trajectory of the unmanned aerial vehicle in the air is automatically calibrated, thereby reducing the error in the flight process.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle control based on computer vision, in particular to an unmanned aerial vehicle route planning and pose calibration method and system. BACKGROUND

[0002] At present, the traditional railway inspection unmanned aerial vehicle is basically manually operated, and the worker needs to control the unmanned aerial vehicle to reach the positioning point, and the inspection personnel also need to follow the unmanned aerial vehicle into the tunnel in real time. This technology will consume a large amount of human resources and increase the learning cost of workers. At the same time, due to the judgment error of manual operation or the delay of information transmission, a series of safety hazards may be caused.

[0003] The unmanned aerial vehicle inspection can be applied to the inspection and maintenance of various facilities, such as power lines, bridges, wind power plants, and petrochemical facilities. These places usually require special requirements such as high-altitude perspective, high safety, and high data accuracy, and the unmanned aerial vehicle can meet these requirements. The traditional unmanned aerial vehicle line inspection flight cannot comprehensively consider the terrain and train and other external factors in the route planning, and only relies on the track line to achieve the flight path planning, which greatly increases the possibility of accidents during the flight of the unmanned aerial vehicle. Secondly, the unmanned aerial vehicle may deviate from the planned route during flight due to wind or external factors. In the aspect of unmanned aerial vehicle track inspection, the unmanned aerial vehicle flight personnel need to undergo various technical training and emergency training, which consumes a large amount of human resource training cost and has low work efficiency, and it is difficult to handle multiple tasks at the same time. In the aspect of unmanned aerial vehicle track inspection route design, the existing flight path planning method has low accuracy and cannot comprehensively consider the terrain, train and many other factors, so that the autonomous track inspection cannot be realized according to the plan.

[0004] An existing unmanned aerial vehicle railway track recognition method uses an unmanned aerial vehicle on-board camera to obtain a railway track video image and performs preprocessing. A pulse-coupled neural network is used to recognize the track line. A third-order Bezier curve fitting is used to implement flight calculation of local target points. This technology uses a flight calculation local target point method, which needs to calculate the line during the flight of the unmanned aerial vehicle, and calculates the subsequent flight route while flying. This method has a large amount of calculation, requires higher endurance of the unmanned aerial vehicle, and greatly increases the calculation cost of the unmanned aerial vehicle. This technology only plans local target waypoints during flight, which cannot guarantee that the unmanned aerial vehicle can effectively avoid obstacles during flight, thereby increasing the risk of accidents during unmanned aerial vehicle flight. This technology does not discuss how to select the target waypoint for the next flight of the unmanned aerial vehicle when encountering a track line intersection. When encountering a track line intersection, the unmanned aerial vehicle has difficulty in route selection. This technology has high real-time requirements, and it is difficult to balance real-time and accuracy. With the increasing amount of unmanned aerial vehicle image data, it is difficult to efficiently store and process these data.

[0005] Another existing unmanned aerial vehicle path planning method, device, storage medium and electronic equipment, first through k-means++, combined with the importance of each target point, the path planning is carried out, then based on the endurance of unmanned aerial vehicle and obstacle as constraint condition, complete coverage within the voyage limit, finally through DDQN algorithm for solving, realize path planning, improve the path planning effect. The technology sets different target points to calculate the local track, the calculation amount is large, it is difficult to balance with the endurance of unmanned aerial vehicle. The technology adopts path planning before unmanned aerial vehicle track, the calculation time is long, the applicable scene has certain limitation. The local calculation of the technology makes the image data of unmanned aerial vehicle storage larger, it is difficult to efficiently store and process these data. SUMMARY

[0006] The present application aims to provide an unmanned aerial vehicle route planning and pose calibration method and system to solve at least one of the technical problems in the background art.

[0007] In order to achieve the above-mentioned purpose, the present application adopts the following technical scheme:

[0008] In a first aspect, the present application provides an unmanned aerial vehicle route planning and pose calibration method, comprising:

[0009] Obtain the reverse track line image and perform visual recognition and line extraction to obtain the track line, and then judge the junction in the forward and reverse directions, and determine whether to implement multi-unmanned aerial vehicle scheduling according to the junction judgment result;

[0010] Obtain the map information in the observation range and perform visual recognition, mark all selectable reference objects in the map within the flight range of the unmanned aerial vehicle, and measure the distance between each marked point and the unmanned aerial vehicle release point; according to the selection priority order of the reference object, quantize the priority, and form a reference point selection function combining the reference point distance, select the optimal reference point and set it as the observation point of the unmanned aerial vehicle;

[0011] According to the track line and the determined observation point, an initial route is drawn; wherein, combining the map information and the flight parameters of the unmanned aerial vehicle, mark the area affecting the flight of the unmanned aerial vehicle, and in the straight line distance between the unmanned aerial vehicle and the observation point, if there is an obstacle affecting the flight of the unmanned aerial vehicle, the marked area will be avoided in the subsequent path planning as the final route;

[0012] In the process of UAV flight, a track picture is obtained every time a specified waypoint is reached, the vertical distance between the position of the UAV and the track line is identified and calculated, it is judged whether the UAV is within the error distance, that is, whether the UAV is off course, if so, the UAV is corrected to the correct position; if not, it is judged whether the waypoint is the last waypoint, if not, the UAV continues to fly to the next waypoint, if so, the UAV stops advancing and reaches the end point.

[0013] Further, it is judged whether multiple UAVs need to be called for detection according to the obtained track line map; wherein it is judged whether the track is a one-way track or a two-way track according to the train timetable, if it is a one-way track, the turnout is judged; if it is a two-way track, it is judged whether the direction of the incoming train is one-way or two-way within the limited time according to the train timetable, if it is one-way, the turnout judgment module is performed, if it is two-way, multiple UAV scheduling is prepared.

[0014] Further, it is judged whether there is a crossing point in the map obtained from the railway line map, so as to judge whether there is a turnout on the track; if there is no turnout, the flight route is directly drawn; if there is a turnout, it is determined whether the turnout is within the target distance range of the UAV flight according to the known UAV flight distance parameter, if not, the useless turnout track is removed, if so, the train timetable module is combined to judge whether the turnout will have trains coming from different directions within the limited time, if not, the line without trains is considered as a useless turnout, if so, multiple UAV flight routes are drawn, UAV flight routes of the two track lines after the turnout are drawn respectively, and multiple UAV scheduling is started.

[0015] Further, the map information in the observation range is obtained and visual recognition is performed, all selectable reference objects in the UAV flight range are marked in the map, and the distance between each marked point and the UAV release point is measured; secondly, the priority of the reference object is quantified according to the selection priority order, and a reference point selection function is formed by combining the reference point distance to select the optimal reference point and set it as the UAV observation point.

[0016] Further, points are taken along the selected track line, the length S of the track line is calculated, the number n of waypoints needed is preliminarily determined according to the length S, and the interval S / n between the waypoints for the approximately straight line is calculated; it is judged whether there is a turning point in the process of taking points along the line, if there is a turning point, the turning point curvature is calculated to determine the number of waypoints needed for different curvatures of the turning point and then the points are taken; if there is no turning point, the points are taken according to the calculated interval.

[0017] Further, the judgment of the inflection point of the unmanned aerial vehicle is based on the calculation of the curvature K according to the curvature calculation formula, the curve not reaching the inflection point curvature threshold is regarded as an approximate straight line, and the curve exceeding the inflection point threshold needs to be further judged according to the curvature size; the number of required inflection points is determined according to the different curvature sizes, and then the points are taken.

[0018] In a second aspect, the present application provides an unmanned aerial vehicle route planning and pose calibration system, comprising:

[0019] The multi-unmanned aerial vehicle scheduling module is used for acquiring the reverse track line image, performing visual identification and line extraction, obtaining the track line, and then judging the turnout in the forward and reverse directions, and determining whether to implement multi-unmanned aerial vehicle scheduling according to the judgment result of the turnout.

[0020] The observation point selection module is used for acquiring the map information in the observation range and performing visual identification, marking all selectable reference objects in the map within the flight range of the unmanned aerial vehicle, and measuring the distance between each marked point and the unmanned aerial vehicle release point; according to the selection priority order of the reference objects, the priority is quantified, and a reference point selection function is formed in combination with the reference point distance to select the optimal reference point and set it as the observation point of the unmanned aerial vehicle.

[0021] The route drawing module is used for drawing an initial route according to the track line and the determined observation point; wherein, in combination with the map information and the flight parameters of the unmanned aerial vehicle, the regions affecting the flight of the unmanned aerial vehicle are marked, and if there are obstacles affecting the flight of the unmanned aerial vehicle in the straight-line distance between the unmanned aerial vehicle and the observation point, the marked region will be avoided in subsequent path planning as the final route.

[0022] The pose correction module is used for acquiring a track picture every time the unmanned aerial vehicle reaches a specified waypoint during flight, identifying and calculating the vertical distance between the position of the unmanned aerial vehicle and the track line, judging whether the unmanned aerial vehicle is within the error distance, that is, judging whether the unmanned aerial vehicle is off course, if yes, the unmanned aerial vehicle is corrected to the correct position; if no, it is judged whether the waypoint is the last waypoint, if not, the unmanned aerial vehicle continues to fly to the next waypoint, if yes, the unmanned aerial vehicle stops advancing and reaches the end point.

[0023] In a third aspect, the present application provides a non-transitory computer readable storage medium for storing computer instructions, which are executed by a processor to implement the unmanned aerial vehicle route planning and pose calibration method of the first aspect.

[0024] In a fourth aspect, the present application provides a computer device, comprising a memory and a processor, the processor and the memory being in communication with each other, the memory storing program instructions executable by the processor, and the processor invoking the program instructions to execute the method for route planning and pose calibration of a UAV as described in the first aspect.

[0025] In a fifth aspect, the present application provides an electronic device, comprising a processor, a memory and a computer program, wherein the processor is connected with the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to make the electronic device execute instructions for implementing the method for route planning and pose calibration of a UAV as described in the first aspect.

[0026] The present application has the advantages of optimizing track identification and improving track identification accuracy, drawing a UAV route through a track line, comprehensively considering external factors such as a map and a train timetable, and drawing an optimal UAV route under the condition of ensuring the minimum cost, and realizing automatic calibration of a UAV track in the air to reduce errors in the flight process.

[0027] The advantages of the additional aspects of the present application will be more apparent from the following description section or be understood through the practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0029] Figure 1 The flow chart of the method for track identification and route planning of a UAV based on computer vision according to the embodiments of the present application.

[0030] Figure 2 The process chart of image preprocessing and analysis according to the embodiments of the present application.

[0031] Figure 3 The flow chart of feature extraction according to the embodiments of the present application.

[0032] Figure 4 The flow chart of track direction judgment according to the embodiments of the present application.

[0033] Figure 5 The flow chart of junction judgment according to the embodiments of the present application.

[0034] Figure 6 The flow chart of a detection point of a UAV based on a map according to the embodiments of the present application.

[0035] Figure 7 A map-based UAV detection point priority quantification flowchart as described in embodiments of the present application.

[0036] Figure 8 A flight path drawing method flowchart as described in embodiments of the present application.

[0037] Figure 9 A map-based UAV flight path optimization method flowchart as described in embodiments of the present application.

[0038] Figure 10 A UAV pose calibration flowchart as described in embodiments of the present application. DETAILED DESCRIPTION

[0039] Embodiments of the present application are described below in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout the drawings denote the same or similar elements or elements having the same or similar functions. The embodiments described below with the accompanying drawings are exemplary and are only used to explain the present application, and cannot be interpreted as a limitation on the present application.

[0040] Those skilled in the art can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as generally understood by those skilled in the art in the field to which the present application belongs.

[0041] It should also be understood that terms such as those defined in a general dictionary should be understood to have meanings consistent with meanings in the context of the prior art, and should not be interpreted with idealized or overly formal meanings unless otherwise defined as such.

[0042] Those skilled in the art can understand that, unless otherwise stated, the singular forms "one", "a", "said" and "the" used herein also include the plural forms. It should be further understood that the phrase "comprising" used in the specification of the present application means that the features, integers, steps, operations, elements and / or groups exist, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements and / or groups thereof.

[0043] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. The person skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples, without contradiction.

[0044] In order to facilitate the understanding of the present application, the present application will be further explained and described in specific embodiments in conjunction with the accompanying drawings, and the specific embodiments do not constitute a limitation on the embodiments of the present application.

[0045] The person skilled in the art should understand that the drawings are only schematic diagrams of the embodiments, and the components in the drawings are not necessarily necessary for the implementation of the present application.

[0046] The present application aims to solve the problems of large manual operation difficulty, low autonomous cruise precision and large error of current unmanned aerial vehicle cruise. A computer vision-based unmanned aerial vehicle route planning and pose calibration method is designed. The autonomous cruise flight of the unmanned aerial vehicle is realized by using computer vision recognition technology and route drawing. The flight error is reduced by autonomous pose calibration in the air. The present application optimizes the track recognition technology and improves the track recognition accuracy. The unmanned aerial vehicle route is drawn through the track line. The external factors such as map and train timetable are considered to draw the optimal unmanned aerial vehicle route under the condition of ensuring the minimum cost. The track of the unmanned aerial vehicle in the air is automatically calibrated, and the error in the flight process is reduced. The method of the present application is widely applied. It can not only be applied to unmanned aerial vehicle cruise, but also be applied to railway inspection, fixed-point survey and detection, unmanned aerial vehicle high-altitude recording and other aspects.

[0047] The main steps of the present application to realize track recognition and route planning are as follows: track image data is obtained by using the camera of the unmanned aerial vehicle, and the track line is recognized by computer vision; the initial unmanned aerial vehicle route is drawn in combination with the map information and train timetable of the region; the drawn unmanned aerial vehicle route is optimized according to the map obstacle information; the line is sent to the flying unmanned aerial vehicle for flying along the line; real-time pose calibration is performed during the flight of the unmanned aerial vehicle.

[0048] Embodiment 1

[0049] In this embodiment 1, first provide a kind of unmanned aerial vehicle route planning and pose calibration system, comprising: multi-unmanned aerial vehicle scheduling module, for obtaining reverse track line image, and carries out visual identification and line extraction, obtains track line, then carries out the judgment of switch in two directions of forward and reverse, according to the judgment result of switch, determine whether to implement multi-unmanned aerial vehicle scheduling.Observation point selection module, for obtaining map information in observation range and carrying out visual identification, mark all selectable reference objects in the map in the flight range of unmanned aerial vehicle, and measure the distance between each mark point and unmanned aerial vehicle release point;According to the priority order of reference object selection, the priority is quantified, and reference point distance is combined to form reference point selection function, select optimal reference point and set it as unmanned aerial vehicle observation point.Route drawing module, for obtaining initial route according to track line and determined observation point;Wherein, in the straight line distance between unmanned aerial vehicle and observation point, if there is obstacle affecting the flight of unmanned aerial vehicle, then avoid the marked area in subsequent path planning, as the final route.Pose correction module, for obtaining a track picture in the process of unmanned aerial vehicle flight every time a specified waypoint is reached, identify the vertical distance between the position of unmanned aerial vehicle and track line, judge whether unmanned aerial vehicle is within error distance, that is, judge whether unmanned aerial vehicle is yaw, if yes, then unmanned aerial vehicle is corrected to correct position;If not, then judge whether the waypoint is the last waypoint, if not the last waypoint, then unmanned aerial vehicle continues to fly to the next waypoint, if the last waypoint, then unmanned aerial vehicle stops advancing, reaches the end point.

[0050] In the embodiment, the unmanned aerial vehicle route planning and pose calibration method is realized by using the system, including: acquiring the reverse track line image, and performing visual recognition and line extraction to obtain the track line, then performing the switch judgment in the forward and reverse directions, determining whether to implement multi-unmanned aerial vehicle scheduling according to the switch judgment result; acquiring the map information in the observation range and performing visual recognition, marking all selectable reference objects in the map in the unmanned aerial vehicle flight range, and measuring the distance between each marked point and the unmanned aerial vehicle release point; according to the selection priority order of the reference object, quantifying the priority, and combining the reference point distance to form a reference point selection function, selecting the optimal reference point and setting it as the unmanned aerial vehicle observation point; according to the track line and the determined observation point, the initial route is drawn; wherein, the map information and the unmanned aerial vehicle flight parameters are combined to mark the area affecting the unmanned aerial vehicle flight, and in the straight line distance between the unmanned aerial vehicle and the observation point, if there is an obstacle affecting the unmanned aerial vehicle flight, the marked area will be avoided in the subsequent route planning as the final route; in the unmanned aerial vehicle flight process, an orbit picture is acquired every time a specified waypoint is reached, the vertical distance between the unmanned aerial vehicle and the track line is identified and calculated, it is judged whether the unmanned aerial vehicle is within the error distance, that is, whether the unmanned aerial vehicle is yawed, if so, the unmanned aerial vehicle is corrected to the correct position; if not, it is judged whether the waypoint is the last waypoint, if not, the unmanned aerial vehicle continues to fly to the next waypoint, if so, the unmanned aerial vehicle stops advancing and reaches the terminal point.

[0051] As shown in Figure 1 The method of the embodiment is a computer vision-based unmanned aerial vehicle track recognition and route planning method.

[0052] Specifically, the track is recognized by using computer vision technology, including:

[0053] As shown in Figure 2As shown, the image preprocessing includes: first, adjusting the image to a uniform size for subsequent processing and model input. Using filters (such as Gaussian filter, mean filter, median filter) to remove noise in the image. Using histogram equalization to improve the contrast of the image, making the track more clearly visible. Using a sharpening filter (such as Laplace filter) to enhance the edge features of the track, making it easier to identify in subsequent processing steps. Utilize the gray scale conversion for color space conversion of the picture, in order to analyze according to the picture features. Among them, the influence of environmental light is eliminated by using white balance; according to the recognition result of the previous frame, the region of interest is extracted; the region of interest is converted into a single channel brightness image. The white balance processing includes: using the gray world method to eliminate the influence of environmental light, and after transformation, the average values of R, G and B components in the image tend to the same gray value; the extraction of the region of interest includes: if the image is the first frame or the last frame image has no recognition result, take the lower half of the image region; if the last frame image has a recognition result, take the trapezoidal region near the recognition result by extending outward from the two lines with the vanishing point coordinate as the upper limit and the image bottom as the lower limit; Specifically, assuming h is the image height, w is the image width, the vanishing point coordinate is (u0, v0), and the coordinates of the end points with larger vertical coordinates in the left and right recognition results are (x0, h) and (x1, h) respectively, then the coordinates of the four vertices of the trapezoidal region are (min(x0, x0-60), h), (5u0 / 6, v0), (7u0 / 6, v0), (min(w, x1+60), h). Converting the image into a single channel brightness image includes: converting the image from BGR mode to HLS mode, taking the L component which represents the brightness information of the image; quantizing the brightness value to 4 values as a level, and obtaining the final single channel brightness image.

[0054] As shown, Figure 3 As shown, the feature extraction includes: using the method of pulse coupled neural network to identify the track line in the video image, using the simplified pulse coupled neural network to iteratively process the single channel brightness image, obtaining the binary image of the pixel excited, and then using the image entropy to judge the stopping moment of the iterative processing. Using Canny algorithm to detect the edge of the track, which is suitable for extracting clear boundary. The steps include: 1. using a Gaussian filter to smooth the image to remove noise. 2. Calculate the gradient of the image to determine the edge intensity and direction. 3. Apply non-maximum suppression to retain local maximum. 4. Use double threshold processing to determine the intensity of the edge, and form the final edge map by connecting the edges. Using the method of third-order Bezier curve fitting to obtain the straight line segment or curve segment where the track line is located; calculating the local target point of the unmanned aerial vehicle flight according to the straight line segment or curve segment where the track line is located.

[0055] Model training, including: first, a large number of images containing train tracks are obtained, including images of different angles, lighting conditions, weather conditions and scene backgrounds to increase the generalization ability of the model. The data set is divided into 70%-80% training set, 10%-15% validation set and 10%-15% test set to evaluate the performance of the model. Through the deep learning method of convolutional neural network, the image data is processed, and the existing architecture YOLO is selected as the basis. The pre-trained model on a large-scale data set is used to adapt to the specific task through fine-tuning. The training process: build the model using the deep learning framework and load the data. The input image is passed into the model, and the output is calculated. The difference between the model output and the true label is calculated through the loss function. Update the model parameters to minimize the loss. Repeat the above process until the specified number of iterations is reached.

[0056] In this embodiment, the track line selection is based on the train timetable and the multi-drone scheduling module. In the train timetable module, the timetable should include train number, departure time, arrival time, starting station, terminal station, passing station, train running speed, and track number, etc. Through the above information, unnecessary track lines can be deleted, thereby reducing unnecessary drone line patrol and flight. First, unmanned aerial vehicle base stations need to be established at appropriate locations along the railway. These base stations not only serve as takeoff and landing points for unmanned aerial vehicles, but also include charging facilities to ensure that unmanned aerial vehicles can continue to perform tasks. The distance between base stations depends on the flight radius of the unmanned aerial vehicle, and in general, base stations will be distributed at key nodes along the railway, such as near stations. In specific application scenarios, such as during railway maintenance, the unmanned aerial vehicle can perform tasks according to the pre-set flight path while receiving updates to the train timetable from the railway system. Once a train is detected to be approaching, the unmanned aerial vehicle will automatically send an early warning to ensure the safety of maintenance personnel. The data collected by the unmanned aerial vehicle can be transmitted in real time to the ground control center through a wireless network, and maintenance personnel can view the images taken by the unmanned aerial vehicle through the control center to timely understand the situation on the track.

[0057] As shown in Figure 4 The multi-drone scheduling module aims to determine whether multiple unmanned aerial vehicles need to be called for detection based on the obtained track line map, which mainly includes two modules: track direction judgment and junction judgment. The track direction judgment module will determine whether the track is a one-way or two-way track based on the train timetable module. If it is a one-way track, it will enter the junction judgment module; if it is a two-way track, it will determine whether the train direction is one-way or two-way within the limited time based on the train timetable. If it is one-way, it will enter the junction judgment module, and if it is two-way, it will prepare for multi-drone scheduling. First, the reverse track line map needs to be taken and visual recognition and line extraction need to be performed to obtain the track line, and both direction line maps need to be judged for junctions, and finally multi-drone scheduling is implemented.

[0058] As shown in Figure 5 , the judgment of the intersection module will determine whether there is a crossing point in the obtained railway line map, so as to determine whether there is a fork in the track. If there is no fork, the flight path is directly drawn; if there is a fork, it is determined whether the fork is within the target distance of the unmanned aerial vehicle flight according to the known unmanned aerial vehicle flight distance parameter (that is, whether the unmanned aerial vehicle needs to pass through the fork to reach the target waypoint), if not, the useless fork track is removed, if yes, the train timetable module is combined to determine whether the fork will have different direction trains within the limit time, if not, the line without train is considered as useless fork, and the flight path is drawn according to the original plan, if yes, the multi-unmanned aerial vehicle flight path is determined, the unmanned aerial vehicle flight path of the two track lines after the fork is drawn respectively, and the multi-unmanned aerial vehicle scheduling is started.

[0059] As shown in Figure 6 , Figure 7 , the map-based unmanned aerial vehicle observation point selection module aims to select appropriate observation points in complex environment, and more stably and efficiently obtain train arrival information. This module will determine according to the priority of reference object. First, the map information in the observation range is obtained and visual recognition is performed, all selectable reference objects in the unmanned aerial vehicle flight range are marked in the map, and the distance m between each marked point and the unmanned aerial vehicle release point is measured; secondly, according to the selection priority order of the reference object, the priority is quantified, and the reference point distance is combined to form the reference point selection function M(m) 。 Let a be the optimal distance between the detection point and the unmanned aerial vehicle release point (the optimal distance can be set by oneself). Let P~N(a, a / 100) be the reference point distance priority function, and the distribution of m satisfies the normal distribution. Let the selection priority of m point be N (the selection method and specific value of N are shown in Figure 7 ), then the reference point selection function of m point is M(m) = 100*P 2 (m)*N. The reference point with the maximum selection function value is selected as the optimal reference point and is set as the unmanned aerial vehicle observation point; finally, the unmanned aerial vehicle observation point is transmitted to the unmanned aerial vehicle flight path mapping module.

[0060] After the above track line and observation point selection modules are completed, the initial unmanned aerial vehicle flight path is drawn.

[0061] As shown in Figure 8As shown, first, the selected track is taken along the line, the length S of the track is calculated, the number of required waypoints n is preliminarily determined according to the length S, and the waypoint interval S / n of the approximately straight line is calculated. It is judged whether there is a turning point in the process of taking along the line. If there is a turning point, the turning point curvature is calculated to determine the number of waypoints required for different curvatures according to the parameters, and then the points are taken. If there is no turning point, the points are taken according to the calculated interval. The last waypoint is the selected observation point. After the points are taken, the given flight parameters are input, including the distance m at which the unmanned aerial vehicle needs to fly on the left or right side of the track, the flight height h of the unmanned aerial vehicle, the flight speed of the unmanned aerial vehicle, etc.

[0062] In the formula, the unmanned aerial vehicle turning point is determined according to the curvature calculation formula to calculate the curvature K. The curve that does not reach the turning point curvature threshold is regarded as an approximately straight line, and the curve that exceeds the turning point threshold needs to be further judged for the curvature size. The number of required turning points is determined according to the different curvature sizes, and then the points are taken.

[0063] As shown in Figure 9 The map-based unmanned aerial vehicle obstacle area marking module aims to mark the area in the actual flight environment that may affect the flight of the unmanned aerial vehicle according to the actual environment and reference map information, so that the unmanned aerial vehicle avoids obstacles in flight and optimizes the planning of the track. In the planning of the route, this module can combine map information and unmanned aerial vehicle flight parameters (such as height) to mark the area that may affect the flight of the unmanned aerial vehicle. In the straight-line distance between the unmanned aerial vehicle and the observation point, if there is no obstacle that may affect the flight of the unmanned aerial vehicle, the unmanned aerial vehicle route drawing module is directly run; otherwise, the marked area is transmitted to the unmanned aerial vehicle route drawing module, and the subsequent track planning will avoid the area.

[0064] As shown in Figure 10 The unmanned aerial vehicle pose calibration aims to prevent the unmanned aerial vehicle from deviating from the planned route during flight, and is a checking mechanism to ensure that the unmanned aerial vehicle can correctly fly along the line within an error range. During the flight of the unmanned aerial vehicle, a track picture is taken every time a specified waypoint is reached, and the vertical distance between the position of the unmanned aerial vehicle and the track line is calculated according to the visual recognition technology to judge whether the unmanned aerial vehicle is within the error distance, that is, whether the unmanned aerial vehicle is off course. If yes, the unmanned aerial vehicle is corrected to the correct position; if no, it is judged whether the waypoint is the last waypoint (i.e. the terminal position). If not, the unmanned aerial vehicle continues to fly to the next waypoint; if yes, the unmanned aerial vehicle stops advancing and reaches the terminal.

[0065] The unmanned aerial vehicle route planning method of the embodiment can be combined with the train timetable to monitor railway construction. When the railway workers need to repair the railway in a certain area, this scheme can be selected to determine whether there is a train passing through during the repair time, so as to remind the repair personnel to evacuate quickly to avoid track accidents. First, the track line map is photographed by the unmanned aerial vehicle, and the track line is identified by computer vision to obtain the track line map. Then, the train timetable of the area where the track is located is input as shown in Table 1:

[0066] Table 1

[0067]

[0068] Through formula calculation, the following Table 2 can be obtained:

[0069] Table 2

[0070]

[0071] According to the judgment module of the fork, it can be judged that the station 2 is a fork, and the number is track 3. According to the timetable, it is calculated that the train does not pass through the track 3 within the reserved time, so the track line where the track 3 is located can be selected to be cleared. The track line where the track 1 and 2 are located is reserved, and the route planning is carried out. Through this method, the track line can be more accurately selected, and the route can be optimized in combination with the map, train timetable and other modules. When multiple lines pass through the train, the unmanned aerial vehicle scheduling module is also reserved. Thus, more accurate and efficient route planning is achieved.

[0072] The unmanned aerial vehicle route planning method described in this embodiment can also be applied to railway line environment monitoring. After natural disasters such as floods and earthquakes occur, due to the complex environmental conditions in the disaster area, it is difficult for detection personnel to enter and conduct on-site exploration. The present application can use an unmanned aerial vehicle to quickly assess the damage to the railway line and provide a repair plan. When the unmanned aerial vehicle patrols the disaster area, it can accurately identify the location of the tracks, and through simple operation by the operator, the unmanned aerial vehicle can automatically find the location of the tracks and plan a flight route along the direction of the tracks, avoiding the safety hazards of personnel entering the disaster area in person. Since the flight route is planned before flight, the camera viewing angle can be freely switched during the patrol, and by shooting multi-angle, different distance railway line environment monitoring videos, picture data, and transmitting them to the ground base station, personnel can further analyze the extent of the track damage. The unmanned aerial vehicle can take multiple angle shots during the track patrol and send the shooting data to the ground control center to determine whether there is any abnormality in the railway, and to develop a repair plan. The system sends maintenance instructions to the unmanned aerial vehicle, which then performs the corresponding operations according to the instructions, such as more detailed inspection of a specific area, recording problem points, or performing simple repair tasks. After the unmanned aerial vehicle completes the task, it will feed back the execution situation to the track monitoring system, which will adjust the maintenance plan according to the feedback information to ensure the safe operation of the track system. During the track patrol of the unmanned aerial vehicle, the unmanned aerial vehicle track patrol system can directly send information to the track monitoring system, which contains the location, type and severity of track abnormalities. This transfers the analysis and decision-making functions from the ground control center to the unmanned aerial vehicle itself, enabling the unmanned aerial vehicle to autonomously decide whether further inspection or simple repair tasks are needed based on real-time detected abnormalities. For example, when the unmanned aerial vehicle detects a slight crack on the track, it can directly send a request to the track monitoring system, which decides whether to dispatch maintenance personnel for on-site inspection or whether to immediately conduct more detailed unmanned aerial vehicle inspection based on the feedback information from the unmanned aerial vehicle and the current track load situation. This decentralization of decision-making power to the unmanned aerial vehicle can significantly reduce the processing burden of the ground control center, improve maintenance efficiency and response speed. At the same time, the autonomous decision-making capability of the unmanned aerial vehicle also makes track maintenance more flexible and efficient, able to adapt to different situations and ensure the safe and stable operation of the track system.

[0073] The unmanned aerial vehicle route planning method described in this embodiment can also be applied to railway tunnel detection. When detecting a railway tunnel, it is necessary to check whether the tunnel wall has cracks, water leakage, and whether the lighting and ventilation equipment in the tunnel are working normally. There are problems such as poor light conditions, tortuous space environment, and complex breathing air composition in the railway tunnel. Using the unmanned aerial vehicle carrying the present application can prevent detection personnel from entering, making the railway tunnel detection safer. The unmanned aerial vehicle flies according to the route planning of the present application for the track, and the flight distance of the unmanned aerial vehicle can be controlled by inputting the flight distance, avoiding operation errors of the operator in a relatively dark environment. Reasonable road planning can achieve obstacle avoidance flight of the unmanned aerial vehicle.

[0074] The unmanned aerial vehicle route planning method described in this embodiment can also be applied to security inspection work in a large city, aiming to improve the safety and emergency response efficiency of the city. The core of this system is multiple unmanned aerial vehicles equipped with high-definition cameras, infrared sensors, and real-time communication equipment. Through pre-set flight path planning algorithms, these unmanned aerial vehicles can regularly patrol key areas in the city, such as business districts, transportation hubs, and large event venues, within a specific time period. In order to ensure that the unmanned aerial vehicle can safely and efficiently perform tasks, flight path planning technology is used to create optimal flight routes. These paths not only consider the maximization of coverage, but also fully consider the existence of obstacles (such as high-rise buildings) and no-fly zones. In addition, the unmanned aerial vehicle system adopts a dynamic adjustment mechanism, which can adjust the flight trajectory in real time according to changes in ground traffic conditions and crowd density, to avoid densely populated areas or emergency sites, while maintaining effective monitoring of important targets. The data collected by the unmanned aerial vehicle, including video streams and image information, is transmitted in real time to the command center through high-speed wireless networks. The command center uses artificial intelligence analysis software to automatically detect abnormal behavior or potential threats and sends alerts to the nearest patrol team. For example, if the unmanned aerial vehicle captures suspicious behavior or finds fire smoke, the system will immediately notify the relevant departments to take quick action. This instant feedback mechanism greatly shortens the time interval from discovery to processing, enhancing the city's ability to respond to emergencies. In addition to regular patrols, unmanned aerial vehicles also play an important role during large public events. They can hover above the event site, providing a bird's eye view to help the police better understand the situation and coordinate security forces. In emergency situations, unmanned aerial vehicles can also serve as temporary communication relay stations to ensure smooth communication between rescue teams. Overall, the application of this technology has significantly improved the city's public security level, providing a safer living environment for citizens.

[0075] In summary, the unmanned aerial vehicle route planning and pose calibration method of the embodiment improves the accuracy of computer vision in identifying track lines through model training. When drawing the initial unmanned aerial vehicle route, the factors such as turnout and two-way trains are judged by combining the train timetable, thereby reducing unnecessary route drawing and unmanned aerial vehicle use; by combining track map information, selecting appropriate markers as reference points, the route is drawn more efficiently. By combining track map obstacle information, the initial unmanned aerial vehicle route is optimized to avoid visible obstacles, thereby reducing the risk of unmanned aerial vehicle flight affected by obstacles. Real-time pose calibration is performed during the unmanned aerial vehicle autonomous track patrol flight to ensure that the unmanned aerial vehicle accurately flies along the line and reduces the unmanned aerial vehicle flight error caused by external factors such as wind.

[0076] On the basis of existing image processing and visual recognition, the embodiment trains a large amount of data model, adds track pictures of different angles, different directions, different colors, different times, etc. to the database for continuous picture processing and visual recognition training, thereby improving the accuracy and real-time performance of the technology. When drawing the initial unmanned aerial vehicle route, the factors such as turnout and two-way trains are judged by combining the train timetable, thereby reducing unnecessary route drawing and unmanned aerial vehicle use. In the train timetable module, the track and train timetable information are input into a predetermined table, including train number, departure time, arrival time, starting station, terminal station, passing station, train speed, etc. After the turnout judgment module, each track line is numbered, and whether each track line has a train passing, the train passing situation within the reserved time, etc. are calculated by formula, and the useless track line (i.e. the track line that does not need to be patrolled by the unmanned aerial vehicle) is removed through the required parameters. At the same time, according to the multi-unmanned aerial vehicle scheduling module, the single and double track and turnout are judged by combining the train timetable module to determine whether multiple unmanned aerial vehicles are needed for patrol, which can reduce resource waste, improve efficiency, and also improve accuracy. By combining track map information, appropriate markers are selected as reference points to draw the route more efficiently. The embodiment combines track map information to select observation points (i.e. the selection of patrol end points) by judging markers such as bridges, tunnels, base stations, signboards, tall trees, etc. This method selects the optimal observation point by importing the map information of the area, modeling the map, and judging the markers. By combining track map obstacle information, it can be determined whether there are obstacles that affect the autonomous flight of the unmanned aerial vehicle near the planned route, including tall trees, tunnels, bridges, mountains, power poles, no-fly zones, etc. The obstacle optimization route module improves the risk of unmanned aerial vehicle flight. To ensure the accuracy of the pose during the unmanned aerial vehicle flight, while reducing the computer calculation cost and data transmission cost, the invention uses intermittent track picture shooting for pose calibration, i.e. shooting track pictures of the specified direction after reaching the predetermined navigation point for distance analysis, and then adjusting the position of the unmanned aerial vehicle.

[0077] The computer vision-based unmanned aerial vehicle track recognition and flight path planning method provided in the embodiment, through computer vision recognition method for picture, video material pretreatment, and then using gray scale conversion for color space conversion of the picture, effectively extracts the track information in the image, inputs the starting point, end point, waypoint and flight height, and draws the unmanned aerial vehicle flight path accordingly. Some method steps are replaced, that is, the unmanned aerial vehicle flight path is not drawn, the photographed picture information is processed in real time during the unmanned aerial vehicle flight, and the flight path is changed according to the actual situation, so that the unmanned aerial vehicle is controlled during flight. In the specific implementation process, some other schemes may exist, which are not listed here.

[0078] Embodiment 2

[0079] The embodiment 2 provides a non-transitory computer readable storage medium for storing computer instructions, when the computer instructions are executed by a processor, the unmanned aerial vehicle flight path planning and pose calibration method as described above is realized, and the method comprises the following steps:

[0080] An image of a reverse track line is obtained, visual recognition and line extraction are performed, a track line is obtained, a judgment of a fork in two directions of forward and reverse is performed, and whether to implement multi-unmanned aerial vehicle scheduling is determined according to the judgment result of the fork;

[0081] Map information in an observation range is obtained and visual recognition is performed, all selectable reference objects in the flight range of the unmanned aerial vehicle are marked on the map, and the distance between each marked point and the release point of the unmanned aerial vehicle is measured; according to the selection priority order of the reference objects, the priority is quantified, and a reference point selection function is formed in combination with the reference point distance to select the optimal reference point and set it as the observation point of the unmanned aerial vehicle;

[0082] An initial flight path is drawn according to the track line and the determined observation point; wherein, in combination with the map information and the flight parameters of the unmanned aerial vehicle, a region affecting the flight of the unmanned aerial vehicle is marked, and in the straight line distance between the unmanned aerial vehicle and the observation point, if there is an obstacle affecting the flight of the unmanned aerial vehicle, the marked region will be avoided in subsequent path planning as the final flight path;

[0083] During the flight of the unmanned aerial vehicle, an image of the track is obtained every time a specified waypoint is reached, the vertical distance between the position of the unmanned aerial vehicle and the track line is calculated, it is judged whether the unmanned aerial vehicle is within the error distance, that is, whether the unmanned aerial vehicle is off course, if yes, the unmanned aerial vehicle is corrected to the correct position; if no, it is judged whether the waypoint is the last waypoint, if not, the unmanned aerial vehicle continues to fly to the next waypoint, if yes, the unmanned aerial vehicle stops moving and reaches the end point.

[0084] Embodiment 3

[0085] The embodiment 3 provides a computer device, comprising a memory and a processor, the processor and the memory are in communication with each other, the memory stores program instructions which can be executed by the processor, and the processor calls the program instructions to execute the unmanned aerial vehicle route planning and pose calibration method as described above, the method comprises the following steps:

[0086] Obtaining the reverse track line image, performing visual identification and line extraction, obtaining the track line, then performing the junction judgment in the forward and reverse directions, and determining whether to implement multi-unmanned aerial vehicle scheduling according to the junction judgment result;

[0087] Obtaining the map information in the observation range and performing visual identification, marking all selectable reference objects in the map in the flight range of the unmanned aerial vehicle, and measuring the distance between each marked point and the unmanned aerial vehicle release point; according to the selection priority order of the reference objects, quantifying the priority, and combining the reference point distance to form a reference point selection function, selecting the optimal reference point and setting it as the observation point of the unmanned aerial vehicle;

[0088] According to the track line and the determined observation point, an initial route is drawn; wherein, in combination with the map information and the unmanned aerial vehicle flight parameters, the region affecting the unmanned aerial vehicle flight is marked, and in the straight line distance between the unmanned aerial vehicle and the observation point, if there is an obstacle affecting the unmanned aerial vehicle flight, the marked region will be avoided in the subsequent route planning as the final route;

[0089] In the process of unmanned aerial vehicle flight, an orbit picture is obtained every time a specified waypoint is reached, the vertical distance between the position of the unmanned aerial vehicle and the track line is identified and calculated, whether the unmanned aerial vehicle is within the error distance is judged, that is, whether the unmanned aerial vehicle is yawed is judged, if yes, the unmanned aerial vehicle is corrected to the correct position; if no, whether the waypoint is the last waypoint is judged, if not, the unmanned aerial vehicle continues to fly to the next waypoint, if yes, the unmanned aerial vehicle stops advancing and reaches the terminal point.

[0090] Embodiment 4

[0091] The embodiment 4 provides an electronic device, comprising a processor, a memory and a computer program; wherein the processor is connected with the memory, and the computer program is stored in the memory; when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes the instructions for realizing the unmanned aerial vehicle route planning and pose calibration method as described above, and the method comprises the following steps:

[0092] Obtaining the reverse track line image, performing visual identification and line extraction, obtaining the track line, then performing the junction judgment in the forward and reverse directions, and determining whether to implement multi-unmanned aerial vehicle scheduling according to the junction judgment result;

[0093] Obtain the map information in the observation range and perform visual recognition, mark all the selectable reference objects in the map within the flight range of the unmanned aerial vehicle, and measure the distance between each marked point and the unmanned aerial vehicle launching point; according to the selection priority order of the reference objects, quantize the priority, combine the reference point distance to form a reference point selection function, select the optimal reference point and set it as the observation point of the unmanned aerial vehicle;

[0094] According to the track line and the determined observation point, an initial flight path is drawn; wherein, in combination with the map information and the flight parameters of the unmanned aerial vehicle, the areas affecting the flight of the unmanned aerial vehicle are marked, and in the straight line distance between the unmanned aerial vehicle and the observation point, if there is an obstacle affecting the flight of the unmanned aerial vehicle, the marked area will be avoided in the subsequent path planning as the final flight path.

[0095] In the flight process of the unmanned aerial vehicle, an orbit picture is obtained every time a specified waypoint is reached, the vertical distance between the position of the unmanned aerial vehicle and the track line is identified and calculated, it is judged whether the unmanned aerial vehicle is within the error distance, that is, whether the unmanned aerial vehicle is yawed, if so, the unmanned aerial vehicle is corrected to the correct position; if not, it is judged whether the waypoint is the last waypoint, if not, the unmanned aerial vehicle continues to fly to the next waypoint, if so, the unmanned aerial vehicle stops advancing and reaches the terminal point.

[0096] In summary, the unmanned aerial vehicle flight path planning and pose calibration method described in the embodiments of the present application aims to realize autonomous orbit flying of the unmanned aerial vehicle with high precision, high efficiency, low risk and low error. It improves the accuracy of computer vision recognition of the track line by training a new model. When drawing the initial flight path of the unmanned aerial vehicle, the factors such as the turnout and the two-way train are judged by combining the train timetable, thereby reducing unnecessary flight path drawing and unmanned aerial vehicle use; by combining the track map information, suitable markers are selected as reference points to more efficiently draw the flight path. By combining the track map obstacle information, the initial unmanned aerial vehicle flight path is optimized to avoid visible obstacles, thereby reducing the risk of being affected by obstacles during the flight of the unmanned aerial vehicle. Real-time pose calibration is performed during the autonomous orbit flying of the unmanned aerial vehicle to ensure that the unmanned aerial vehicle flies accurately along the line and reduce the flight error of the unmanned aerial vehicle caused by external factors such as wind.

[0097] Those skilled in the art will appreciate that embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0098] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart or flows and / or block diagram block or blocks. Figure 1 one or more flow or flows and / or block diagram block or blocks. Figure 1 one or more flow or flows and / or block diagram block or blocks.

[0099] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart or flows and / or block diagram block or blocks. Figure 1 one or more flow or flows and / or block diagram block or blocks. Figure 1 one or more flow or flows and / or block diagram block or blocks.

[0100] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart or flows and / or block diagram block or blocks. Figure 1 one or more flow or flows and / or block diagram block or blocks. Figure 1 one or more flow or flows and / or block diagram block or blocks.

[0101] The above description is only specific implementation of the present application, and is not intended to limit the scope of protection of the present application. Those skilled in the art should understand that various modifications or variations can be made on the basis of the disclosed technical solutions without inventive labor, and all these modifications or variations should be covered within the scope of protection of the present application.

Claims

1. A method for unmanned aerial vehicle (UAV) flight path planning and pose calibration, characterized in that, The method comprises the following steps: Obtain the reverse track line image, and perform visual recognition and line extraction to obtain the track line, and then perform forward and reverse direction turnout judgment, and determine whether to implement multi-UAV scheduling according to the turnout judgment result; Obtain the map information in the observation range and perform visual recognition, mark all selectable reference objects in the map within the UAV flight range, and measure the distance between each marked point and the UAV release point; according to the selection priority order of the reference objects, quantize the priority, and form a reference point selection function combining the reference point distance to select the optimal reference point and set it as the UAV observation point; According to the track line and the determined observation point, the initial flight path is drawn; wherein, combined with the map information and the UAV flight parameters, the areas affecting the UAV flight are marked, and if there are obstacles affecting the UAV flight in the straight-line distance between the UAV and the observation point, the marked area will be avoided in the subsequent flight path planning as the final flight path; During the UAV flight, a track picture is obtained every time a specified waypoint is reached, the vertical distance between the UAV position and the track line is calculated, it is judged whether the UAV is within the error distance, that is, whether the UAV is off course, if so, the UAV is corrected to the correct position; if not, it is judged whether the waypoint is the last waypoint, if not, the UAV continues to fly to the next waypoint, if so, the UAV stops advancing and reaches the end point. 2.The UAV route planning and pose calibration method of claim 1, wherein, According to the obtained track line map, it is judged whether multiple UAVs need to be called for detection; wherein, according to the train timetable, it is judged whether the track is a one-way track or a two-way track, if it is a one-way track, the turnout judgment is performed; if it is a two-way track, it is judged whether the direction of the incoming train is one-way or two-way within the limited time, if it is one-way, the turnout judgment module is performed, if it is two-way, the multi-UAV scheduling is prepared. 3.The UAV route planning and pose calibration method of claim 1, wherein, According to the obtained railway line map, it is judged whether there is a crossing point in the map to determine whether there is a turnout; if there is no turnout, the flight path is directly drawn; If there is a turnout, it is determined whether the turnout is within the target distance range of the UAV flight according to the known UAV flight distance parameters, if not, the useless turnout track line is removed, if so, the train timetable module is combined to judge whether the turnout will have trains coming from different directions within the limited time, if not, the line without trains is considered as a useless turnout, if so, the multi-UAV flight path is drawn, the UAV flight paths of the two track lines after the turnout are drawn respectively, and the multi-UAV scheduling is started. 4.The UAV route planning and pose calibration method of claim 1, wherein, Obtain the map information in the observation range and perform visual recognition, mark all selectable reference objects in the map within the UAV flight range, and measure the distance between each marked point and the UAV release point; secondly, according to the selection priority order of the reference objects, quantize the priority, and form a reference point selection function combining the reference point distance to select the optimal reference point and set it as the UAV observation point. 5.The UAV route planning and pose calibration method of claim 4, wherein, For the selected track line, points are taken along the line, the length S of the track line is calculated, the number n of required waypoints is preliminarily determined according to the length S, and the waypoint interval S / n for the approximately straight line is calculated; it is judged whether there is a turning point in the process of taking points along the line, if there is a turning point, the turning point curvature is calculated to determine the number of required waypoints for different curvatures according to the parameters, and then the points are taken; if there is no turning point, the points are taken according to the calculated interval. 6.The UAV route planning and pose calibration method of claim 5, wherein, The judgment of the turning point of the unmanned aerial vehicle is based on the curvature calculation formula to calculate the curvature K, and the curve that does not reach the turning point curvature threshold is regarded as an approximately straight line, while the curve that exceeds the turning point threshold needs to be further judged for the curvature size; the number of required turning points is determined according to the different curvature sizes, and then the points are taken.

7. An unmanned aerial vehicle route planning and pose calibration system, comprising: It comprises: A multi-unmanned aerial vehicle scheduling module is used to acquire a reverse track line image, and perform visual recognition and line extraction to obtain a track line, and then judge a fork in two directions, and determine whether to implement multi-unmanned aerial vehicle scheduling according to the judgment result of the fork; An observation point selection module is used to acquire map information in an observation range and perform visual recognition, mark all selectable reference objects in the map within the flight range of the unmanned aerial vehicle, and measure the distance between each marked point and the unmanned aerial vehicle release point; according to the selection priority order of the reference objects, the priority is quantified, and a reference point selection function is formed by combining the reference point distance to select the optimal reference point and set it as the unmanned aerial vehicle observation point; A flight path drawing module is used to draw an initial flight path according to the track line and the determined observation point; wherein, in combination with the map information and the unmanned aerial vehicle flight parameters, the areas affecting the flight of the unmanned aerial vehicle are marked, and if there are obstacles affecting the flight of the unmanned aerial vehicle in the straight line distance between the unmanned aerial vehicle and the observation point, the marked area will be avoided in subsequent path planning as the final flight path; A pose correction module is used to acquire a track picture every time the unmanned aerial vehicle reaches a specified waypoint during flight, identify and calculate the vertical distance between the position of the unmanned aerial vehicle and the track line, judge whether the unmanned aerial vehicle is within the error distance, that is, whether the unmanned aerial vehicle is off course, if so, the unmanned aerial vehicle is corrected to the correct position; if not, it is judged whether the waypoint is the last waypoint, if not, the unmanned aerial vehicle continues to fly to the next waypoint, if so, the unmanned aerial vehicle stops advancing and reaches the terminal point.

8. A non-transitory computer-readable storage medium, comprising: The non-transitory computer readable storage medium is used to store computer instructions, which are executed by the processor to implement the unmanned aerial vehicle flight path planning and pose calibration method of any one of claims 1-6.

9. A computer device, comprising: It comprises a memory and a processor, the processor and the memory communicate with each other, the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the unmanned aerial vehicle flight path planning and pose calibration method of any one of claims 1-6.

10. An electronic device, comprising: It comprises: A processor, a memory and a computer program; wherein the processor is connected with the memory, the computer program is stored in the memory, when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes the instructions for realizing the unmanned aerial vehicle route planning and pose calibration method according to any one of claims 1-6.

Images