Unmanned aerial vehicle real-time path planning system based on digital twinning

Abstract

The invention discloses an unmanned aerial vehicle real-time path planning system based on digital twinning, relates to the technical field of unmanned aerial vehicles, and aims to realize efficient and real-time path planning of an unmanned aerial vehicle in a campus traffic road environment. The core architecture of the system comprises a traffic flow monitoring end, a data analysis processing end, a digital twin model construction end and an unmanned aerial vehicle regulation and control end. A campus traffic digital twinborn model is constructed by using the traffic flow and the campus traffic road network three-dimensional diagram, the campus traffic flow in a future period of time is predicted in real time, and a prediction result is transmitted to the unmanned aerial vehicle regulation and control end; the number and flight paths of the unmanned aerial vehicles corresponding to the traffic roads are locked according to the predicted flow, accurate and dynamic path planning of the unmanned aerial vehicles is achieved, and the flight efficiency and adaptability of the unmanned aerial vehicles in the complex traffic environment of the campus are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicles (UAVs), and specifically relates to a real-time path planning system for UAVs based on digital twins. Background Art

[0002] With the continuous advancement of technology, the application scenarios of drones on university campuses are becoming increasingly diverse, including but not limited to logistics distribution, security monitoring, emergency rescue, etc.

[0003] University campuses are densely populated, and traffic roads often lack corresponding traffic lights and traffic control methods, so traffic accidents often occur; traditional drone path planning methods are difficult to adapt to the complex environment of university campuses and are often unable to timely and accurately monitor pedestrian and vehicle conflicts; when using drones to monitor campus traffic flow, the existing technology lacks effective solutions for allocating drone resources and planning their flight paths, and cannot fully tap the application potential of drones on campus. Based on the above problems, the present invention proposes a real-time drone path planning system based on digital twins. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides a real-time drone path planning system based on digital twins, which solves the problem that the existing technology lacks an effective method for dynamically allocating drone resources according to campus traffic flow.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A real-time path planning system for drones based on digital twins. This system includes the following: Traffic flow monitoring terminal, millimeter wave radars are arranged on various roads on campus to monitor campus traffic flow in real time and map the three-dimensional map of the campus traffic road network. The obtained campus traffic flow and the three-dimensional map of the campus traffic road network are transmitted to the data analysis and processing terminal through the bus; The data analysis and processing end receives and analyzes the campus traffic flow and the three-dimensional map of the campus traffic road network collected by the traffic flow monitoring end, processes the campus traffic flow into a spatiotemporal aligned campus traffic flow spatiotemporal dataset, and transmits it together with the three-dimensional map of the campus traffic road network to the digital twin model construction end; The digital twin model construction end receives the spatiotemporal dataset of campus traffic flow and the three-dimensional map of the campus traffic road network transmitted by the data analysis and processing end, builds a digital twin model of campus traffic, and makes real-time predictions of campus traffic flow in the future. The predicted campus traffic flow is then transmitted to the drone control end. The drone control terminal receives the campus traffic flow predicted by the campus traffic digital twin model, and locks the number of drones matching the corresponding traffic roads and the flight paths of the drones matching the corresponding traffic roads; Cloud storage terminal, used to store any data involved in any terminal of this solution; Used to store any calculation steps and methods described in any end of this solution.

[0006] As a further solution of the present invention, the specific method of the traffic flow monitoring terminal to monitor campus traffic flow in real time is: The campus traffic roads are divided into sections and each section is equipped with a millimeter wave radar; The total number of all campus traffic roads after segmentation is recorded as j, and the j segments of campus traffic roads are recorded as the campus traffic road sequence in the order of segmentation: ,in, for Any one of , and i is a counting index, ; Get the monitoring period T preset by the operator and determine the campus traffic roads within the monitoring period T The campus traffic flow rate is recorded as The duration of the monitoring period T is determined by the operator; Determine one day as a statistical period, and determine the total number of monitoring periods T within a statistical period, and record it as q; For campus traffic roads The campus traffic flow associated with q monitoring periods T within a statistical period is recorded as the campus traffic flow sequence: ; Process campus traffic roads according to the sequence of campus traffic roads The campus traffic flow sequence associated with each of the j sections of campus traffic roads is obtained by processing the data using the method of

[0007] As a further solution of the present invention, the specific method of mapping the three-dimensional map of the campus traffic road network by the traffic flow monitoring terminal is: Based on any section of campus traffic road in the determined campus traffic road sequence , get The millimeter-wave radar is equipped with Three-dimensional map of the area, record ; Repeat the above steps to determine the three-dimensional graphs corresponding to each of the j campus traffic roads, and sort the j three-dimensional graphs according to the order of the campus traffic road sequence, which is recorded as the three-dimensional graph sequence ; For three-dimensional graph sequence After splicing and fitting, the three-dimensional graph G of the campus traffic road network is obtained.

[0008] As a further solution of the present invention, the data analysis and processing end processes campus traffic flow into a spatiotemporal aligned campus traffic flow spatiotemporal dataset in the following specific manner: Determine the start and end timestamps corresponding to each of the q monitoring periods T within a statistical period, and sort them by time, recording them as a timestamp sequence ,in Indicates the start timestamp and end timestamp of the first monitoring period T; Extract any section of campus traffic road , build Associated monitoring period T - campus traffic flow matrix ; The monitoring period T - campus traffic flow matrix Expressed as: ; express The relationship between campus traffic flow and time; Repeat the above steps to The same process is performed on each section of campus traffic road in the statistical cycle to obtain the monitoring period T of each section of campus traffic road in a statistical cycle - campus traffic flow matrix, and the sequence of the campus traffic road sequence is recorded as a matrix sequence ; Statistical periods that are different from the current statistical period are processed according to the above steps.

[0009] As a further solution of the present invention, the data analysis and processing end processes campus traffic flow into a spatiotemporal aligned campus traffic flow spatiotemporal dataset in a specific manner further comprising: Extract any section of campus traffic road from G The three-dimensional coordinate set of the center line , : ; in, express The three-dimensional coordinates of the center line, n is The total number of three-dimensional coordinates of the center line is n; based on The three-dimensional coordinate set of the center line ,calculate The spatial eigenvector of ,in Campus traffic roads The mean of the three-dimensional coordinates of the center line, Campus traffic roads length; Associated and Associated Perform tensor splicing to generate The spatiotemporal data of campus traffic flow aligned with time and space is recorded as ; statistics The corresponding spatiotemporal data of campus traffic flow aligned with each other are recorded as campus traffic flow spatiotemporal dataset, which is expressed as .

[0010] As a further solution of the present invention, the digital twin model construction end receives the campus traffic flow spatiotemporal dataset and the three-dimensional graph G of the campus traffic road network to build a digital twin model of campus traffic; Campus traffic digital twin model predicts campus traffic road sequence The campus traffic flow in a future monitoring period T is recorded as the predicted traffic flow sequence and expressed as: ; The predicted traffic flow sequence obtained by the campus traffic digital twin model is transmitted to the drone control end through the bus.

[0011] As a further solution of the present invention, the specific method of the drone control terminal locking the number of drones matching the corresponding traffic road is: Based on the determined predicted traffic volume: , using the traffic flow threshold preset by the operator: Group the j sections of campus traffic roads; Extract any section of campus traffic road from j sections of campus traffic road and its corresponding predicted traffic flow ,like , then Classified into groups ,like , then Classified into groups ,like , then Classified into groups ,like , then Classified into groups ; Repeat the above steps for the jth section of campus traffic roads to complete the grouping; Get the corresponding groups preset by the operator Number of drones: , and lock the number of drones corresponding to each of the j sections of campus traffic roads.

[0012] As a further solution of the present invention, the specific method for the drone control terminal to lock the flight path of the drone matching the corresponding traffic road is: extract and the corresponding number of drones ,in The value of Any one of; Get from G Area ,right Perform grid division, the number of divisions is , respectively denoted as ; for Any grid in , take the grid Coordinates of the center point of the plane ,in is a counting index, and ; Get the operator's preset Flight altitude within , joint coordinates Composition of three-dimensional coordinates , and As the initial position of the drone within the grid; Repeat the above steps to calculate within The three-dimensional coordinates associated with each grid are calculated according to The adjacent relationships between the grids form a cycle, and the three-dimensional coordinates are sorted according to this cycle order. The sorting result is: ,in, and For adjacent grids, and For adjacent grids, and so on; In three-dimensional coordinates as well as The line between two adjacent grids and The flight path of the UAV between the two grids is obtained by processing the subsequent adjacent grids in this way. The flight path of the UAV within the future monitoring period T can be obtained by analogy. The flight paths of UAVs within other campus traffic roads can be obtained, where the flight speed of the UAV is preset by the operator.

[0013] Beneficial effects of the present invention: (1) The present invention can obtain traffic flow information continuously around the clock by rationally arranging millimeter-wave radars on various roads on campus. It can maintain stable monitoring performance even in adverse weather conditions, ensuring the continuity and reliability of data. Based on the monitored traffic flow information, a digital twin model is established and traffic flow information in the future is predicted, providing a more forward-looking and accurate decision-making basis for the drone control end, facilitating the real-time or advance planning of the drone's flight path, and effectively improving the drone's mission execution efficiency. (2) The present invention constructs a complete three-dimensional map of the campus traffic road network by using millimeter-wave radar to map the three-dimensional maps of each section of the road, and splices and fits them through image fitting technology; it provides detailed and accurate spatial information for drone path planning, so that drones can more accurately perceive the geographical and spatial characteristics of campus traffic roads; this three-dimensional map that integrates traffic data and spatial information can help the system better understand the complexity of campus traffic, thereby comprehensively considering factors such as the direction, distance, and width of the road when planning the flight path, and formulating a more reasonable and efficient flight route for the drone; (3) The present invention combines traffic flow data with timestamps to construct a monitoring period T - campus traffic flow matrix, which can clearly show the relationship between campus traffic flow and time. By observing the traffic flow matrix, the peaks and troughs of traffic flow in different time periods can be found, and the flight time of the drone can be planned in advance. Secondly, the present invention introduces spatial feature vectors to combine traffic flow data with information such as the spatial position and length of the road to generate spatiotemporal traffic flow data that is aligned in time and space. This method not only retains the temporal characteristics of traffic flow data, but also adds information in the spatial dimension, providing a more comprehensive reference basis for drone path planning. (4) The present invention innovatively introduces the digital twin model into the prediction of traffic flow on university campuses, predicts the traffic flow in the next monitoring period in real time, and transmits the prediction results to the drone control end; this process realizes a seamless connection from data collection to prediction, thereby providing forward-looking and accurate guidance for the real-time path planning of drones based on the dynamic changes in traffic flow. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present invention will be further described below with reference to the accompanying drawings.

[0015] Figure 1 It is a structural diagram of the system of the present invention; Figure 2 Schematic diagram of the process of the method described in Example 2 of the present invention; Figure 3 Schematic diagram of the process of the method described in Example 3 of the present invention. DETAILED DESCRIPTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0017] Example 1

[0018] A real-time path planning system for UAVs based on digital twins, such as Figure 1 As shown, specifically including the following: The terminals (modules) in this system mainly include the following: The traffic flow monitoring terminal and this terminal are mainly used to collect basic data, including campus traffic flow and three-dimensional map of campus traffic road network; the implementation method is to segment the roads in the campus and equip each section of the road with a millimeter-wave radar to achieve full coverage of the roads in the university campus. The millimeter-wave radar can monitor the campus traffic flow in the corresponding section of the road in real time, and can map the three-dimensional map of the corresponding section of the road. Then, by summarizing the three-dimensional maps mapped by all millimeter-wave radars and performing fitting processing, the three-dimensional map of the campus traffic road network can be obtained. The traffic flow monitoring terminal interacts with the data analysis and processing terminal in real time through the bus, and transmits the monitored campus traffic flow and the three-dimensional map of the campus traffic road network to the data analysis and processing terminal through the bus.

[0019] The data analysis and processing end and this end receive the campus traffic flow and the three-dimensional map of the campus traffic road network transmitted from the traffic flow monitoring end through the bus, process the campus traffic flow through a fixed monitoring period, and combine the processed campus traffic flow with the three-dimensional map of the campus traffic road network to convert it into a spatiotemporal aligned campus traffic flow spatiotemporal dataset, making it more organized and relevant in the time and space dimensions, which is convenient for subsequent digital twin model modeling and analysis. At the same time, the data analysis and processing end transmits the processed campus traffic flow spatiotemporal dataset and the three-dimensional map of the campus traffic road network to the digital twin model construction end through the bus.

[0020] The digital twin model construction end and the receiving data analysis and processing end transmit the spatiotemporal data set of campus traffic flow and the three-dimensional map of the campus traffic road network; First, a corresponding digital twin model is constructed in virtual space based on the three-dimensional map of the campus traffic road network, and recorded as the campus traffic digital twin model. The campus traffic digital twin model is trained and corrected using the spatiotemporal dataset of campus traffic flow. The campus traffic digital twin model predicts the campus traffic flow in the next monitoring cycle in real time, and transmits the predicted campus traffic flow to the drone control end through the bus.

[0021] The drone control end and the local end include a computing unit for real-time processing of campus traffic flow within a future monitoring cycle predicted by the campus traffic digital twin model; and also include several drones to be dispatched; Drones are planned and dispatched by analyzing campus traffic flow predicted by the digital twin model.

[0022] Cloud storage, including several databases; This terminal interacts with each terminal in the system in real time and stores any data and any operation steps involved in each terminal.

[0023] This system aims to achieve real-time monitoring, analysis, and prediction of campus traffic flow through the coordinated cooperation of various terminals, as well as intelligent control of drones based on the prediction results. With the help of technologies such as digital twins, a more intelligent and efficient system framework for campus traffic management and related drone applications is built, which helps to improve the operational efficiency of campus traffic, optimize resource allocation (that is, the rational use of drones), and provide strong support for traffic management on campus.

[0024] Example 2

[0025] This embodiment further discloses a method for constructing a spatiotemporal dataset of campus traffic flow and a digital twin model of campus traffic based on embodiment 1. Figure 2 As shown, the specific steps include: First, all campus traffic roads within the current university campus are obtained, including roads for pedestrians and roads for vehicles. Before segmenting the campus traffic roads, it is necessary to clarify the detection diameter of the millimeter-wave radar used. The detection diameter represents the monitoring range of the millimeter-wave radar. It is necessary to ensure that each segmented campus traffic road can be completely monitored by the millimeter-wave radar used; The total number of segmented campus traffic roads is recorded as j, and according to the order of segmentation of the j-segment campus traffic road, the j-segment campus traffic road is recorded as the campus traffic road sequence, which is expressed as ,in, for Any one of them, and i is a counting index, starting from 1 and the maximum value is j.

[0026] Determine a monitoring period T, the duration of which is set by the operator based on actual conditions and needs; Extract campus traffic road sequence within a monitoring period T Any campus traffic road , and further obtain campus traffic roads The millimeter-wave radar equipped on campus traffic roads The campus traffic flow monitored in real time within the monitoring period T is recorded as , represents the campus traffic flow monitored in the first monitoring period T, that is, .

[0027] Then determine the time of one day (i.e., 0:00 to 24:00 every day) as a statistical period, and determine the total number of monitoring periods T within a statistical period based on the length of the monitoring period T set according to the operation combined with the actual situation and needs, and record it as q. The q is not a fixed value. If the time of the monitoring period T changes during the implementation of this system, then q must also change accordingly.

[0028] Repeated monitoring of campus traffic roads Campus traffic flow during the first monitoring period T Methods for campus traffic roads The campus traffic flow associated with q monitoring periods T within a statistical period is obtained and recorded as a campus traffic flow sequence in the order of acquisition, which is expressed as ; The above steps are based on campus traffic roads As an example, we get the campus traffic road In a statistical cycle, that is, the campus traffic flow sequence associated with q monitoring cycles T, similarly, the campus traffic road sequence is calculated according to this method. By performing the same process on any one of them, we can obtain the campus traffic flow sequence associated with any section of campus traffic road within a statistical period.

[0029] In Example 1, it is mentioned that the millimeter wave radar is used to monitor campus traffic flow in real time and is also used to map the three-dimensional map of campus traffic roads. Therefore, based on the campus traffic road sequence determined in the above method, , extract any one of the campus traffic roads again Perform sample processing.

[0030] Get campus transportation routes The millimeter-wave radar equipped and the campus traffic roads it mapped The three-dimensional map in the area is recorded as ; Similarly, for the campus traffic road sequence , we can get j three-dimensional graphs, and then sort the three-dimensional graphs corresponding to the j campus traffic roads in the order of the campus traffic road sequence, and record them as the three-dimensional graph sequence: ; Use image fitting technology (this technology is an existing technology and will not be described in detail here) to perform the following on the three-dimensional image sequence: After splicing and fitting, the final three-dimensional map of the campus traffic road network corresponding to the current university campus is obtained and recorded as G.

[0031] Determine a statistical period and the q monitoring periods T associated with the statistical period, further obtain the start timestamp and end timestamp corresponding to each of the q monitoring periods T, and sort the start timestamps and end timestamps corresponding to each of the q monitoring periods T in chronological order, where the end timestamp of one monitoring period T is also the start timestamp of the next adjacent monitoring period T; The sorted results are recorded as a timestamp sequence: , where the first timestamp and the second timestamp constitute the timestamp interval: , the timestamp interval represents the duration, start timestamp, and end timestamp of the first monitoring period T within the statistical period, and the start timestamp and end timestamp of subsequent monitoring periods T correspond to the timestamps in the timestamp sequence in sequence; Road sequence for campus traffic Any section of campus traffic road , further build campus traffic roads Associated monitoring period T - campus traffic flow matrix , the monitoring period T——campus traffic flow matrix Expressed as: , the matrix records the campus traffic roads corresponding to different monitoring periods T within a statistical period By analyzing the data in the matrix, we can observe the changing trend of campus traffic flow and identify peak and off-peak periods.

[0032] Based on campus traffic roads The processing steps are then used to process the campus traffic road sequence The same process is performed on each section of campus traffic road in the statistical cycle, and finally the monitoring cycle T of each section of campus traffic road in a statistical cycle can be obtained - the campus traffic flow matrix, and recorded as a matrix sequence according to the order of the campus traffic road sequence, which is expressed as ; Subsequent statistical periods that are different from the current statistical period are processed according to the above steps to obtain the matrix sequence corresponding to the campus traffic road sequence associated with any statistical period.

[0033] Then obtain the campus traffic road sequence from the determined campus traffic road network three-dimensional graph G Any section of campus traffic road , extract campus traffic roads The three-dimensional coordinate set of the center line , Display as: in, to Represents campus traffic roads The three-dimensional coordinates of the center line, n is the campus traffic road The total number of three-dimensional coordinates of the center line, a total of n, the three-dimensional coordinate set Provides campus transportation routes The precise location in three-dimensional space helps to accurately determine the location and direction of each road. These three-dimensional coordinates can be used for more accurate path planning.

[0034] Based on the determined campus traffic roads The three-dimensional coordinate set of the center line , and further calculate the campus traffic roads The spatial eigenvector of ; in, Campus traffic roads The mean of the abscissas in the three-dimensional coordinates of the center line, Campus traffic roads The mean of the vertical coordinates in the three-dimensional coordinates of the center line, Campus traffic roads The mean value of the Z-axis coordinate in the three-dimensional coordinates of the center line, the coordinate Used to indicate campus traffic roads Position in the current space, Campus traffic roads length.

[0035] Campus traffic roads Associated monitoring period T - campus traffic flow matrix Traffic roads to campus The associated spatial eigenvector Perform tensor splicing to finally generate campus traffic roads The associated spatiotemporal aligned campus traffic flow spatiotemporal data is recorded as ; Summarize campus traffic road sequence The spatiotemporal data of campus traffic flow associated with each section of campus traffic road in the dataset are recorded as the spatiotemporal dataset of campus traffic flow and sorted according to the sequence of campus traffic road, which is expressed as .

[0036] The above steps are all completed in the data analysis and processing end, which then transmits the campus traffic flow spatiotemporal dataset to the data analysis and processing end through the bus. The three-dimensional graph G of the campus traffic road network is transmitted to the digital twin model construction end for constructing a corresponding campus traffic digital twin model (digital twin technology is a well-known existing technology and will not be described in detail here); The campus traffic digital twin model simulates and predicts campus traffic road sequences in real time The campus traffic flow in a future monitoring period T is recorded as the predicted traffic flow sequence and expressed as: The predicted time range is preset by the operator based on the actual situation, and the minimum limit (the shortest prediction time) is the time of one monitoring cycle T; The digital twin model construction end then predicts the traffic flow sequence through the bus Transmitted to the drone control end.

[0037] Example 3 This embodiment further discloses a UAV flight path planning method applicable to campus traffic roads based on embodiment 1 and embodiment 2. Figure 3 As shown, the specific steps include: This embodiment is completed in the drone control terminal, which receives the predicted traffic flow sequence transmitted by the digital twin model construction terminal. First, it is necessary to group all the predicted traffic flows in the predicted traffic flow sequence and obtain the traffic flow thresholds preset by the operator: , extract campus traffic road sequence Any campus traffic road , get campus traffic roads Associated predicted traffic flow ; like Less than or equal to , then the campus traffic roads Classified into groups ; like Less than or equal to , then the campus traffic roads Classified into groups ; like Less than or equal to , then the campus traffic roads Classified into groups ; like Greater than , then the campus traffic roads Classified into groups ; Completing the above steps can determine the campus traffic roads Associated predicted traffic flow Group you belong to; Traffic road sequence on campus Repeat the above process for all campus traffic roads in Steps to complete the road sequence for campus transportation Perform grouping operations; Based on the identified groups , again get the number of drones corresponding to each group preset by the operator: , and lock the campus traffic road sequence The number of drones corresponding to each campus traffic road.

[0038] Determine campus traffic routes The corresponding number of drones ,in The value of Any one of; Then obtain the campus traffic road from the campus traffic road network three-dimensional graph G The area of ​​​​ , using campus transportation roads The corresponding number of drones Traffic roads on campus Area Perform grid division and obtain campus traffic roads after grid division Associated grids, respectively ; For campus traffic roads Associated Grids: Any grid in , get the grid The plane where the campus is located (campus traffic road The coordinates of the center point of the plane where the ; Then obtain the campus traffic road by the operator Preset flight altitude of the drone , and combined with the determined center point coordinates Combined into a three-dimensional coordinate, also recorded as And the original two-dimensional coordinates Continue forest cover; The three-dimensional coordinates As a grid The initial position of the drone; Repeat the above processing grid Steps, for Grid The same process is performed on each grid in The adjacent relationship between the grids forms a circular route, and the The three-dimensional coordinates associated with each grid are sorted, and the sorted result is: ; in, For adjacent grids, For adjacent grids, ..., are adjacent grids, and and is the adjacent grid; Then, the three-dimensional coordinates associated with any two adjacent grids are used as the flight path of the drone between the two adjacent grids. For example, for the adjacent grids and , get and Associated 3D coordinates as well as , connect the three-dimensional coordinates by straight lines as well as Get the drone in the adjacent network and The flight path between Repeat the above steps and perform the same processing on each group of adjacent grids, and finally you can get the campus traffic roads. The flight paths of the drones in the future monitoring period T can be obtained by analogy. The flight paths of the drones in other campus traffic roads can also be obtained. Based on the above method, the campus traffic roads are obtained The flight path of the UAV in the next monitoring period T, and similarly, the campus traffic road sequence can be obtained Except campus traffic roads The flight path of the drone associated with other campus traffic roads other than the designated area; the flight speed of the drone is preset by the operator based on actual needs.

[0039] Some of the data in the formulas described above are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0040] The above contents are merely examples and explanations of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in similar ways. As long as they do not deviate from the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

[0041] It is important to note that all user data collected in this application is collected with the user's consent and authorization. Furthermore, the use of user data is legal and compliant, and the use and processing of user data complies with the relevant laws, regulations, and standards of the relevant regions.

Claims

1. A real-time path planning system for UAVs based on digital twins, characterized by: This system includes the following: Traffic flow monitoring terminal, millimeter wave radars are arranged on various roads on campus to monitor campus traffic flow in real time and map the three-dimensional map of the campus traffic road network. The obtained campus traffic flow and the three-dimensional map of the campus traffic road network are transmitted to the data analysis and processing terminal through the bus; The data analysis and processing end receives and analyzes the campus traffic flow and the three-dimensional map of the campus traffic road network collected by the traffic flow monitoring end, processes the campus traffic flow into a spatiotemporal aligned campus traffic flow spatiotemporal dataset, and transmits it together with the three-dimensional map of the campus traffic road network to the digital twin model construction end; The digital twin model construction end receives the spatiotemporal dataset of campus traffic flow and the three-dimensional map of the campus traffic road network transmitted by the data analysis and processing end, builds a digital twin model of campus traffic, and makes real-time predictions of campus traffic flow in the future. The predicted campus traffic flow is then transmitted to the drone control end. The drone control terminal receives the campus traffic flow predicted by the campus traffic digital twin model, and locks the number of drones matching the corresponding traffic roads and the flight paths of the drones matching the corresponding traffic roads; Cloud storage terminal, used to store any data involved in any terminal of this solution; Used to store any calculation steps and methods described in any end of this solution.

2. A real-time path planning system for UAVs based on digital twins according to claim 1, characterized in that: The specific method of the traffic flow monitoring terminal to monitor campus traffic flow in real time is: The campus traffic roads are divided into sections and each section is equipped with a millimeter wave radar; The total number of all campus traffic roads after segmentation is recorded as j, and the j segments of campus traffic roads are recorded as the campus traffic road sequence in the order of segmentation: ,in, for Any one of , and i is a counting index, ; Get the monitoring period T preset by the operator and determine the campus traffic roads within the monitoring period T The campus traffic flow rate is recorded as The duration of the monitoring period T is determined by the operator; Determine one day as a statistical period, and determine the total number of monitoring periods T within a statistical period, and record it as q; For campus traffic roads The campus traffic flow associated with q monitoring periods T within a statistical period is recorded as the campus traffic flow sequence: ; Process campus traffic roads according to the sequence of campus traffic roads The campus traffic flow sequence associated with each of the j sections of campus traffic roads is obtained by processing the data using the method of 3. A real-time path planning system for UAVs based on digital twins according to claim 2, characterized in that: The specific method of mapping the three-dimensional map of the campus traffic road network by the traffic flow monitoring terminal is: Based on any section of campus traffic road in the determined campus traffic road sequence , get The millimeter-wave radar is equipped with Three-dimensional map of the area, record ; Repeat the above steps to determine the three-dimensional graphs corresponding to each of the j campus traffic roads, and sort the j three-dimensional graphs according to the order of the campus traffic road sequence, which is recorded as the three-dimensional graph sequence ; For three-dimensional graph sequence After splicing and fitting, the three-dimensional graph G of the campus traffic road network is obtained.

4. A real-time path planning system for UAVs based on digital twins according to claim 3, characterized in that: The specific method for the data analysis and processing end to process the campus traffic flow into a spatiotemporal aligned campus traffic flow spatiotemporal dataset is as follows: Determine the start and end timestamps corresponding to each of the q monitoring periods T within a statistical period, and sort them by time, recording them as a timestamp sequence ,in Indicates the start timestamp and end timestamp of the first monitoring period T; Extract any section of campus traffic road , build Associated monitoring period T - campus traffic flow matrix ; The monitoring period T - campus traffic flow matrix Expressed as: ; express The relationship between campus traffic flow and time; Repeat the above steps to The same process is performed on each section of campus traffic road in the statistical cycle to obtain the monitoring period T of each section of campus traffic road in a statistical cycle - campus traffic flow matrix, and the sequence of the campus traffic road sequence is recorded as a matrix sequence ; Statistical periods that are different from the current statistical period are processed according to the above steps.

5. The real-time path planning system for UAV based on digital twin according to claim 4 is characterized in that: The specific method of processing the campus traffic flow into a spatiotemporal aligned campus traffic flow spatiotemporal dataset by the data analysis and processing end further includes: Extract any section of campus traffic road from G The three-dimensional coordinate set of the center line , : ; in, express The three-dimensional coordinates of the center line, n is The total number of three-dimensional coordinates of the center line is n; based on The three-dimensional coordinate set of the center line ,calculate The spatial eigenvector of ,in Campus traffic roads The mean of the three-dimensional coordinates of the center line, Campus traffic roads length; Associated and Associated Perform tensor splicing to generate The spatiotemporal data of campus traffic flow aligned with time and space is recorded as ; statistics The corresponding spatiotemporal data of campus traffic flow aligned with each other are recorded as campus traffic flow spatiotemporal dataset, which is expressed as .

6. A real-time path planning system for UAVs based on digital twins according to claim 5, characterized in that: The digital twin model construction end receives the campus traffic flow spatiotemporal dataset and the three-dimensional graph G of the campus traffic road network to build a digital twin model of campus traffic; Campus traffic digital twin model predicts campus traffic road sequence The campus traffic flow in a future monitoring period T is recorded as the predicted traffic flow sequence and expressed as: ; The predicted traffic flow sequence obtained by the campus traffic digital twin model is transmitted to the drone control end through the bus.

7. The real-time path planning system for UAV based on digital twin according to claim 6, characterized in that: The specific method for the drone control terminal to lock the number of drones matching the corresponding traffic road is: Based on the determined predicted traffic volume: , using the traffic flow threshold preset by the operator: Group the j sections of campus traffic roads; Extract any section of campus traffic road from j sections of campus traffic road and its corresponding predicted traffic flow ,like , then Classified into groups ,like , then Classified into groups ,like , then Classified into groups ,like , then Classified into groups ; Repeat the above steps for the jth section of campus traffic roads to complete the grouping; Get the corresponding groups preset by the operator Number of drones: , and lock the number of drones corresponding to each of the j sections of campus traffic roads.

8. The real-time path planning system for UAV based on digital twin according to claim 7, characterized in that: The specific method for the drone control terminal to lock the flight path of the drone that matches the corresponding traffic road is as follows: extract and the corresponding number of drones ,in The value of Any one of; Get from G Area ,right Perform grid division, the number of divisions is , respectively denoted as ; for Any grid in , take the grid Coordinates of the center point of the plane ,in is a counting index, and ; Get the operator's preset Flight altitude within , joint coordinates Composition of three-dimensional coordinates , and As the initial position of the drone within the grid; Repeat the above steps to calculate within The three-dimensional coordinates associated with each grid are calculated according to The adjacent relationships between the grids form a cycle, and the three-dimensional coordinates are sorted according to this cycle order. The sorting result is: ,in, and For adjacent grids, and For adjacent grids, and so on; In three-dimensional coordinates as well as The line between two adjacent grids and The flight path of the UAV between the two grids is obtained by processing the subsequent adjacent grids in this way. The flight path of the UAV within the future monitoring period T can be obtained by analogy. The flight paths of UAVs within other campus traffic roads can be obtained, where the flight speed of the UAV is preset by the operator.

Images