A heterogeneous unmanned swarm system based on virtual-real integration

Abstract

The present invention relates to the technical field of unmanned cluster simulation, and in particular, to a heterogeneous unmanned cluster system based on the combination of virtual and real. The system includes a motion prediction unit, which calculates the position of a drone at a certain moment in the future based on the Runge-Kutta method; a communication compensation unit, which adjusts the drone data through prediction based on the position of the drone at a certain moment in the future obtained by the motion prediction unit to align it with the current time point; a two-way clock synchronization unit, which selects the clock of a node with high precision in the network as a reference clock, and other nodes as sub-clocks, and performs clock synchronization with the reference clock; a master-slave time synchronization unit, which sets a fixed synchronization period based on the synchronization mechanism provided by the two-way clock synchronization unit, and each sub-simulator runs independently during this period, and pauses at the end of the period to exchange data.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned cluster simulation, and in particular to a heterogeneous unmanned cluster system based on the combination of virtual and real. Background Art

[0002] The present invention takes into account the experimental conditions of large-scale heterogeneous unmanned systems. In order to integrate the development experience of predecessors and successfully complete the difficult heterogeneous unmanned system cluster experiments, a heterogeneous unmanned system cluster test platform based on the combination of virtual and real is established, which integrates heterogeneous UAV intelligent bodies such as rotor / fixed wing, completes large-scale complex test verification, and achieves the purpose of shortening the development interval, improving safety and reducing experimental costs.

[0003] At present, due to the limitation of space and experimental cost, large-scale cluster experiments are difficult to carry out in domestic research institutions and experiments. Many research institutions and the military at home and abroad have developed and applied experimental platforms based on virtual-real integration. Among them, the first problem to be solved is to design a virtual-real experimental framework based on virtual-real integration that can run efficiently. After analyzing the performance requirements of digital twin models at different stages, a set of UAV digital twin model construction methods were established from the six aspects of "construction-group-integration-verification-calibration-management". Based on the proposed digital twin model construction criteria and methods, a set of virtual-real experimental platform design frameworks suitable for intelligent UAV clusters and virtual-real experimental platform frameworks for UAV management were designed. The models were described from the aspects of geometry, physics, behavior and rules, and the interactive mapping between physical entities and virtual entities was realized. After having a complete system architecture, the construction of the virtual-real platform was gradually realized. The simulation lacks visualization and the simulation granularity is low. The data-driven training environment modeling has certain differences in consistency with the real environment data, and there is no participation of physical nodes. The number of nodes in the intelligent hardware-in-the-loop simulation experiment is limited, and there is a lack of experimental capabilities to establish large-scale UAV nodes. There are few collaborative control studies currently conducted for the application scenarios of virtual-real collaborative tasks. Therefore, a heterogeneous unmanned swarm system based on virtual-real integration is designed. Summary of the invention

[0004] The purpose of the present invention is to provide a heterogeneous unmanned swarm system based on the combination of virtual and real, so as to solve the problems that the simulation proposed in the above background technology lacks visualization and the simulation granularity is low. The data-driven training environment modeling has certain differences in consistency with the real environment data, and there is no participation of physical nodes. The number of nodes in the intelligent hardware-in-the-loop simulation experiment is limited, and there is a lack of experimental capabilities to establish large-scale drone nodes. For collaborative task application scenarios that combine virtual and real, there are few collaborative control studies currently conducted.

[0005] To achieve the above object, the present invention aims to provide a heterogeneous unmanned cluster system based on virtual-real combination, including a motion prediction unit, which calculates the position of the unmanned aerial vehicle at a certain moment in the future based on the Runge-Kutta method;

[0006] A communication compensation unit, wherein the communication compensation unit adjusts the drone data by prediction based on the position of the drone at a certain moment in the future obtained by the motion prediction unit so as to align the drone data with the current time point;

[0007] A bidirectional clock synchronization unit, wherein the bidirectional clock synchronization unit selects the clock of a node with high accuracy in the network as a reference clock and other nodes as sub-clocks, and performs clock synchronization with the reference clock;

[0008] The master-slave time synchronization unit sets a fixed synchronization period based on the synchronization mechanism provided by the bidirectional clock synchronization unit. During this period, each sub-simulator runs independently and pauses at the end of the period to exchange data.

[0009] As a further improvement of the technical solution, in the motion prediction unit, the specific steps of using the Runge-Kutta method to calculate the position of the drone at a certain moment in the future are:

[0010] S2.1. Each drone node Provide current location and speed ;

[0011] S2.2, use the Runge-Kutta method to predict future positions;

[0012] S2.3. Calculate the drone's future time location.

[0013] As a further improvement of the technical solution, in S2.2, the specific expression for predicting the future position using the Runge-Kutta method is:

[0014] ;

[0015] Where f represents the acceleration caused by control input and other factors; Indicates the current speed; represents the predicted position change based on the current velocity and half the time step; based on and the change in position predicted by half the time step; based on and the predicted position change for the full time step.

[0016] As a further improvement of this technical solution, in S2.3, the calculation of the UAV at the future time The specific expression of the position is:

[0017] ;

[0018] in, Represents a drone node In time The predicted position of Represents the time increment of the forecast.

[0019] As a further improvement of the technical solution, in the communication compensation unit, the specific steps of adjusting the drone data by prediction to align it with the current time point are:

[0020] S5.1. The ground station receives data including the location from each drone node. and speed Status information;

[0021] S5.2, find the state information closest to the current time t as a benchmark;

[0022] S5.3. Use motion prediction to predict its position at the current time t, and adjust the data to align it with the current time point.

[0023] As a further improvement of the technical solution, in S5.3, the specific expression for using motion prediction to predict its position at the current time t and adjusting the data to align it with the current time point is:

[0024] ;

[0025] in, represents the predicted position of drone node j at time t; Represents the time at which the drone node j location; Data delay time of drone node j; Represents the time at which the drone node j speed.

[0026] As a further improvement of the technical solution, in the bidirectional clock synchronization unit, the specific steps of performing clock synchronization are:

[0027] S7.1. When the time synchronization cycle arrives, the slave clock sends a time slot synchronization query message to the reference clock and records the timestamp when the slave clock sends it. ;

[0028] S7.2. When the reference clock receives the time slot synchronization inquiry message, the reference clock records the timestamp of the local clock at this time. ;

[0029] S7.3, in the idle time slot, the reference clock replies to the slave clock with a synchronization response message, which carries a timestamp and ;

[0030] S7.4. When the slave clock receives the synchronization response message sent by the reference clock, the slave clock records the local clock timestamp at this time. ;

[0031] S7.5. Calculate clock offset based on timestamp ;

[0032] S7.6, sub-clock according to the calculated clock offset Adjust its own clock to synchronize with the reference clock.

[0033] As a further improvement of the technical solution, in S7.5, the clock offset is calculated The specific expression is:

[0034] ;

[0035] in, Indicates clock skew.

[0036] As a further improvement of the technical solution, in the master-slave time synchronization unit, by setting a fixed synchronization period, each sub-simulator runs independently during this period, and pauses at the end of the period to exchange data. The specific steps are:

[0037] S9.1. Set a fixed synchronization period T;

[0038] S9.2, based on the synchronization mechanism provided by the bidirectional clock synchronization unit, used to ensure that the time base of all nodes is consistent;

[0039] S9.3, in each synchronization cycle, each sub-simulator operates independently;

[0040] S9.4, at the end of each cycle T, the sub-simulators pause and exchange data;

[0041] S9.5. Check whether a critical event occurs at each pause. If a critical event is detected, adjust the step size. And sync immediately.

[0042] As a further improvement of the technical solution, in S9.5, when a critical event is detected, the step length is adjusted The specific expression for immediate synchronization is:

[0043] ;

[0044] in, Indicates the adjusted synchronization period; Indicates the time point when a key event occurs.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. In this heterogeneous unmanned swarm system based on the combination of virtual and real, the Runge-Kutta method is used for numerical integration with high accuracy, which can better predict the position of the UAV at a certain moment in the future. By using the future position prediction provided by the motion prediction unit, the communication compensation unit can reduce the data asynchrony problem caused by communication delay and improve the real-time performance of the system.

[0047] 2. In this heterogeneous unmanned cluster system based on the combination of virtual and real, a high-precision node in the network is selected as the reference clock to ensure the consistency of the time reference and improve the synchronization accuracy of the overall system. The master-slave time synchronization mechanism is easy to expand and is suitable for complex systems composed of multiple sub-simulators. It can effectively manage the synchronization requirements of large-scale systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is the overall flow chart of the present invention;

[0049] Figure 2 Set up the interface map for LinkTrack;

[0050] Figure 3 Set the line diagram for the LinkTrack interface;

[0051] Figure 4 This is the main interface diagram of flight control;

[0052] Figure 5 Set up the map for the main flight control interface;

[0053] Figure 6 This is a diagram of the ground control station;

[0054] Figure 7 Modified diagram for the ground control station:

[0055] Figure 8 This is the UAV-1 information display diagram;

[0056] Fig. 9 This is the installation mode diagram;

[0057] Fig.10 Run the algorithm graph for the button;

[0058] Fig.11 Planning map for the trip path;

[0059] Fig.12 It is a UE three-dimensional graph;

[0060] The meaning of each number in the figure is:

[0061] 1. Motion prediction unit; 2. Communication compensation unit; 3. Bidirectional clock synchronization unit; 4. Master-slave time synchronization unit. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0063] Example 1: See Figure 1 As shown, a heterogeneous unmanned cluster system based on the combination of virtual and real is provided, including a motion prediction unit 1, which calculates the position of the unmanned aerial vehicle at a certain moment in the future based on the Runge-Kutta method;

[0064] In this example, the specific steps of using the Runge-Kutta method to calculate the position of the drone at a certain moment in the future are:

[0065] S2.1. Each drone node Provide current location and speed ;

[0066] S2.2, use the Runge-Kutta method to predict future positions;

[0067] In this example, the specific expression for predicting future positions using the Runge-Kutta method is:

[0068] ;

[0069] Where f represents the acceleration caused by control input and other factors; Indicates the current speed; represents the predicted position change based on the current velocity and half the time step; based on and the change in position predicted by half the time step; based on and the predicted position change for the full time step.

[0070] S2.3. Calculate the drone's future time location.

[0071] In this example, the drone is calculated at a future time. The specific expression of the position is:

[0072] ;

[0073] in, Represents a drone node In time The predicted position of Represents the time increment of the forecast.

[0074] The heterogeneous unmanned cluster system based on the combination of virtual and real also includes a communication compensation unit 2, which adjusts the drone data through prediction based on the position of the drone at a certain moment in the future obtained by the motion prediction unit 1, so as to align it with the current time point;

[0075] In this example, the specific steps to adjust the drone data through prediction to align it with the current time point are:

[0076] S5.1. The ground station receives data including the location from each drone node. and speed Status information;

[0077] S5.2, find the state information closest to the current time t as a benchmark;

[0078] S5.3. Use motion prediction to predict its position at the current time t, and adjust the data to align it with the current time point.

[0079] In this example, motion prediction is used to predict its position at the current time t, and the specific expression for adjusting the data to align it with the current time point is:

[0080] ;

[0081] in, represents the predicted position of drone node j at time t; Represents the time at which the drone node j location; Data delay time of drone node j; Represents the time at which the drone node j speed.

[0082] Specifically, during the simulation process, the aircraft status received by the ground station is delayed. In order to achieve a more accurate evaluation of the algorithm through 3D, the data sent to the ground station by different aircraft needs to be processed synchronously.

[0083] The heterogeneous unmanned cluster system based on virtual-real combination also includes a bidirectional clock synchronization unit 3, which selects the clock of a node with high precision in the network as a reference clock and other nodes as sub-clocks, and performs clock synchronization with the reference clock;

[0084] In this example, the specific steps for clock synchronization are:

[0085] S7.1. When the time synchronization cycle arrives, the slave clock sends a time slot synchronization query message to the reference clock and records the timestamp when the slave clock sends it. ;

[0086] S7.2. When the reference clock receives the time slot synchronization inquiry message, the reference clock records the timestamp of the local clock at this time. ;

[0087] S7.3, in the idle time slot, the reference clock replies to the slave clock with a synchronization response message, which carries a timestamp and ;

[0088] S7.4. When the slave clock receives the synchronization response message sent by the reference clock, the slave clock records the local clock timestamp at this time. ;

[0089] S7.5. Calculate clock offset based on timestamp ;

[0090] In this example, the clock offset is calculated The specific expression is:

[0091] ;

[0092] in, Indicates clock skew.

[0093] S7.6, sub-clock according to the calculated clock offset Adjust its own clock to synchronize with the reference clock.

[0094] Specifically, the two-way clock synchronization algorithm: select the clock of a high-precision node in the network as the reference clock, and the inquiring party as the sub-clock, and the sub-clock needs to be synchronized with the master clock. This algorithm does not require the reference unit to reply within a time slot, will not affect the transmission of high-priority services, and ensures the service quality of the service and the real-time nature of message transmission.

[0095] This heterogeneous unmanned cluster system based on the combination of virtual and real also includes a master-slave time synchronization unit 4. The master-slave time synchronization unit 4 sets a fixed synchronization period based on the synchronization mechanism provided by the bidirectional clock synchronization unit 3. During this period, each sub-simulator runs independently and pauses at the end of the period to exchange data.

[0096] In this example, a fixed synchronization period is set, during which each sub-simulator runs independently and pauses at the end of the period to exchange data. The specific steps are as follows:

[0097] S9.1. Set a fixed synchronization period T;

[0098] S9.2, based on the synchronization mechanism provided by the bidirectional clock synchronization unit 3, to ensure that the time base of all nodes is consistent;

[0099] S9.3, in each synchronization cycle, each sub-simulator operates independently;

[0100] S9.4, at the end of each cycle T, the sub-simulators pause and exchange data;

[0101] S9.5. Check whether a critical event occurs at each pause. If a critical event is detected, adjust the step size. And sync immediately.

[0102] In this example, when a critical event is detected, the step size is adjusted The specific expression for immediate synchronization is:

[0103] ;

[0104] in, Indicates the adjusted synchronization period; Indicates the time point when a key event occurs.

[0105] Specifically, although the master-slave synchronization also has a synchronization cycle, each simulator no longer runs independently within each cycle step. Instead, the network simulator first performs a cycle of simulation and caches the network events that occur within this cycle. When the network simulator pauses at the cycle event time point, it sends its event simulation results to the drone simulation end, and reproduces these network events that affect the drone simulation during the process of advancing the drone simulation. After advancing to the synchronization time point, the two simulators exchange information with each other for time synchronization. This synchronization method realizes the response of the drone end to network events during simulation, so it will not accumulate a large amount of synchronization errors like the time-stepping synchronization method.

[0106] The key to the test system for combining virtual and real is to add the simulation function of the networked communication system to the UAV simulation. Its time synchronization algorithm must ensure stable and orderly data interaction on the basis of minimizing the impact on the system. In order to enable the UAV simulation to respond to non-periodic asynchronous network events in a timely manner, a variable-step synchronization method is improved on the basis of combining bidirectional clock synchronization and master-slave synchronization.

[0107] Among the asynchronous events processed by network simulation, not all events require the system to respond immediately, such as the transmission of data information for observation and debugging at the simulation end. Such events will not modify the flight status of the drone like mission instructions, so for such common events, the synchronization does not need to be very strict. Therefore, in the improved algorithm, the system will judge the current synchronization status. If the triggered event is not a mission event that will change the state of the drone system, the position and attitude will still be updated according to the set initial synchronization cycle; if the triggered event is a mission event or a control command event, a new synchronization method will be executed. When the task layer of the drone simulation end issues mission instructions and other information that will change the state of the drone system, the entire collaborative simulation will be suspended, and the drone simulation end will transmit the relevant mission information to the network simulation end and insert the relevant mission events in its own event queue. Then, the network simulation end starts, advances the network simulation, executes the highest priority task event, and then pauses to send the simulation results to the drone simulation end, and uses the execution time of the previous task event as the new synchronization cycle to advance the drone end simulation, and then continues to interact with the task information and position attitude information at the synchronization point, and continues to simulate the next task event until the task event is processed. The collaborative simulation synchronization cycle will be restored to the synchronization cycle for processing ordinary events. This method does not accumulate system errors and enables the drone system to respond to certain non-periodic asynchronous network events in a timely manner. This is because this method adds asynchronous event time points on the basis of periodic event time points, allowing the drone system to respond to emergency network events in a timely manner, and when there are no emergency tasks or commands, it can perform master-slave synchronization, which can effectively reduce the number of synchronous interactions and improve simulation efficiency.

[0108] Example 2: Please refer to Figure 2-12 As shown, a heterogeneous unmanned cluster system based on the combination of virtual and real is provided.

[0109] The heterogeneous unmanned cluster system positioning system based on virtual and real combination adopts UWB positioning system and is connected to the simulation system through a centralized ground station. The virtual and real combination scenario is 8 meters In the 8-meter field, a U-shaped channel is constructed. Two 480 rotor drones and two unmanned vehicles form a formation of two to cross the U-shaped channel to search, identify and attack. During the formation search, obstacles are automatically identified and avoided. The specific process is as follows:

[0110] 1) Site calibration

[0111] UWB is a carrier-free communication technology that uses non-sinusoidal narrow pulses from nanoseconds to microseconds to transmit data. UWB has the advantages of high time resolution, strong penetration, low power consumption, good anti-multipath effect, and high security, so it is often used in the fields of communication and positioning.

[0112] ANCHOR is generally used as a reference position point and is generally installed at a fixed reference point; TAG is generally used as a point to be positioned and is generally installed on a carrier to be positioned; CONSOLE is generally used to monitor the operating status of the system and issue instructions to other nodes (ANCHOR, TAG), and is generally connected to a Terminal, such as a computer, tablet computer, etc.

[0113] During calibration, A0 and A3 are the base stations that must work, and A0 is automatically set as the origin of the coordinate system (0, 0, 0). The A0-A3 direction is the X-axis direction of the coordinate system. The default X-axis direction is north. The A1 and A2 coordinates are considered to be in the positive direction of the Y-axis. If four base stations are required for calibration, A1 and A2 are required to be on the same side of the X-axis. If the base station layout is set in a square, A0-A1-A2-A3 are placed in a counterclockwise order, with A0-A1 facing east and A0-A3 facing north.

[0114] The nose of the aircraft is facing north and the power-on position of the aircraft is the tail.

[0115] Double-click to open the NAssistant software and select LinkTrack settings. Figure 2 and 3 As shown;

[0116] Click one-button calibration and observe the changes in X and Y of A0, A1, A2, and A3 base stations. When the values ​​remain stable, the positioning system calibration is completed. (Note: During the calibration process, do not place aircraft and other objects within the rectangular range of A0, A1, A2, and A3 base stations, and do not power on the aircraft, because when the base station is calibrated, the signal sent encounters objects within the rectangular range, and the signal will be fed back to the base station, which may affect the calibration)

[0117] After the calibration is completed, do not close the LinkTrack setting interface, place the aircraft in the calibration area, select line in the main interface, and observe the aircraft's position information. (Note: UWB generates the aircraft's X and Y position information, and the Z value data is generated by laser height determination)

[0118] When the aircraft's position information is roughly consistent with its actual position, it indicates that the positioning system is normal.

[0119] 2) Cluster control ground station

[0120] like Figure 4 This is an introduction to the cluster control ground station page. The flight control main interface is as follows: Figure 4 and 5 As shown:

[0121] Set up the main interface:

[0122] Function: Supports adding and deleting multiple aircraft, and setting the added aircraft, including MATLAB communication port, steel body, actual flight and simulation type, communication protocol, device port, etc., which is easy to operate.

[0123] 3) Pre-flight inspection

[0124] Before flying, make sure that the UWB module cable is connected to the adapter board and the GPS cable is not connected! The flight control parameters are set correctly, that is, use QGC to check the parameters: EKF2_AID_MASK is vision position fusion and EKF2_HGT_MODE is Range sensor. (You can refer to the pre-flight parameter setting document)

[0125] Open the "Cluster Control Ground Station" software, set relevant parameters, check the aircraft status, click "Take Off", and the aircraft hovers stably, indicating that the aircraft status is normal, and then program control can be performed.

[0126] Click Settings on the left, change the MATLAB port number on the left of the first line to 28005, the MATLAB local port number on the right to 28004, and the local port number on the last line to 15603. Click the up arrow button to connect to the No. 13 unmanned vehicle;

[0127] Click Settings on the left, change the MATLAB port number on the left of the first line to 28007, the MATLAB local port number on the right to 28006, and the local port number on the last line to 15604. Click the up arrow button to connect to the No. 14 unmanned vehicle.

[0128] Click Settings on the left, change the MATLAB port number on the left of the first line to 27005, the MATLAB local port number on the right to 27004, and the local port number on the last line to 15503. Click the up arrow button to connect to aircraft No. 3.

[0129] Click Settings on the left, change the MATLAB port number on the left of the first line to 27007, the MATLAB local port number on the right to 27006, and the local port number on the last line to 15504. Click the up arrow button to connect to aircraft No. 4.

[0130] 4) Simulation configuration and startup

[0131] Open the browser and enter the URL "192.168.151.10:18016" to enter the simulation platform;

[0132] On the top menu of the homepage: click Others -> Console -> Deployment Console -> Server List, select the Ground Control Station tab on the far right: Configure the ground station IP address 192.168.151.168; Figure 6 As shown;

[0133] Note: Just fill in the IP address of the machine where the Matlab cluster control ground station is located, and click Save after filling in.

[0134] Click Experiment Management, select the actual flight underground passage in the experimental scene in the experiment list, and after entering, select the specified equipment in the combat programming list on the left and click Edit; Figure 7 As shown;

[0135] Only two fields, "Is it a real equipment?" and "Real ID," need to be modified. Change the "Is it a real equipment?" field of the corresponding equipment to "Yes." Change the "Real ID" field to the ID of the corresponding physical aircraft on the Matlab cluster control ground station, such as Figure 8 As shown;

[0136] Click to start the installation mode: After the startup is completed, the connection information of the physical object is as follows, and you can check which equipment is successfully connected. Fig. 9 As shown;

[0137] Go back to the home page and click Algorithm Access->Path Planning Algorithm. Click the algorithm to modify it. After the modification is completed, click the main.ipynb file on the left and select the item to run on the right. Fig.10 As shown, click the Run button above to run the algorithm.

[0138] 5) Operation effect

[0139] When the algorithm is running, the aircraft takes off and flies to the first waypoint. The NX board on the aircraft automatically identifies the search target and performs path planning, such as Fig.11 As shown;

[0140] During the flight, the 2D situation displays the aircraft’s trajectory in real time and the UE 3D in the virtual scene. Fig.12 As shown;

[0141] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only preferred examples of the present invention, and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A heterogeneous unmanned cluster system based on the combination of virtual and real, characterized by: include A motion prediction unit (1), wherein the motion prediction unit (1) calculates the position of the drone at a certain moment in the future based on the Runge-Kutta method; A communication compensation unit (2), wherein the communication compensation unit (2) adjusts the drone data through prediction based on the position of the drone at a certain moment in the future obtained by the motion prediction unit (1) so as to align the data with the current time point; A bidirectional clock synchronization unit (3), wherein the bidirectional clock synchronization unit (3) selects a clock of a node with high accuracy in the network as a reference clock and other nodes as sub-clocks, and performs clock synchronization with the reference clock; A master-slave time synchronization unit (4) is provided. The master-slave time synchronization unit (4) sets a fixed synchronization cycle based on the synchronization mechanism provided by the bidirectional clock synchronization unit (3). During this cycle, each sub-simulator operates independently and pauses at the end of the cycle to exchange data.

2. The heterogeneous unmanned cluster system based on virtual-real integration according to claim 1 is characterized by: In the motion prediction unit (1), the specific steps of using the Runge-Kutta method to calculate the position of the drone at a certain moment in the future are: S2.

1. Each drone node Provide current location and speed ; S2.2, use the Runge-Kutta method to predict future positions; S2.

3. Calculate the drone's future time location.

3. The heterogeneous unmanned cluster system based on virtual-real integration according to claim 2 is characterized by: In S2.2, the specific expression for predicting the future position using the Runge-Kutta method is: ； Where f represents the acceleration caused by control input and other factors; Indicates the current speed; represents the predicted position change based on the current velocity and half the time step; based on and the change in position predicted by half the time step; based on and the predicted position change for the full time step.

4. The heterogeneous unmanned cluster system based on virtual-real integration according to claim 3 is characterized by: In S2.3, the calculation of the drone at the future time The specific expression of the position is: ； in, Represents a drone node In time The predicted position of Represents the time increment of the forecast.

5. The heterogeneous unmanned cluster system based on virtual-real integration according to claim 1 is characterized by: In the communication compensation unit (2), the specific steps of adjusting the drone data by prediction to align it with the current time point are: S5.

1. The ground station receives data including the location from each drone node. and speed Status information; S5.2, find the state information closest to the current time t as a benchmark; S5.

3. Use motion prediction to predict its position at the current time t, and adjust the data to align it with the current time point.

6. The heterogeneous unmanned cluster system based on virtual-real integration according to claim 5 is characterized by: In S5.3, the specific expression for using motion prediction to predict its position at the current time t and adjusting the data to align it with the current time point is: ； in, represents the predicted position of drone node j at time t; Represents the time at which the drone node j location; Data delay time of drone node j; Represents the time at which the drone node j speed.

7. The heterogeneous unmanned cluster system based on virtual-real integration according to claim 1 is characterized by: In the bidirectional clock synchronization unit (3), the specific steps of performing clock synchronization are: S7.

1. When the time synchronization cycle arrives, the slave clock sends a time slot synchronization query message to the reference clock and records the timestamp when the slave clock sends it. ; S7.

2. When the reference clock receives the time slot synchronization inquiry message, the reference clock records the timestamp of the local clock at this time. ; S7.3, in the idle time slot, the reference clock replies to the slave clock with a synchronization response message, which carries a timestamp and ; S7.

4. When the slave clock receives the synchronization response message sent by the reference clock, the slave clock records the local clock timestamp at this time. ; S7.

5. Calculate clock offset based on timestamp ; S7.6, sub-clock according to the calculated clock offset Adjust its own clock to synchronize with the reference clock.

8. The heterogeneous unmanned cluster system based on virtual-real integration according to claim 7 is characterized by: In S7.5, the clock offset is calculated The specific expression is: ； in, Indicates clock skew.

9. The heterogeneous unmanned cluster system based on virtual-real integration according to claim 1 is characterized by: In the master-slave time synchronization unit (4), a fixed synchronization period is set, during which each sub-simulator runs independently and pauses at the end of the period to exchange data. The specific steps are as follows: S9.

1. Set a fixed synchronization period T; S9.2, based on the synchronization mechanism provided by the bidirectional clock synchronization unit (3), used to ensure that the time base of all nodes is consistent; S9.3, in each synchronization cycle, each sub-simulator operates independently; S9.4, at the end of each cycle T, the sub-simulators pause and exchange data; S9.

5. Check whether a critical event occurs at each pause. If a critical event is detected, adjust the step size. And sync immediately.

10. The heterogeneous unmanned cluster system based on virtual-real integration according to claim 9 is characterized by: In S9.5, when a critical event is detected, the step length is adjusted The specific expression for immediate synchronization is: ； in, Indicates the adjusted synchronization period; Indicates the time point when a key event occurs.

Images