Multi-unmanned aerial vehicle control method based on digital twinning

Abstract

According to the multi-unmanned-aerial-vehicle control method based on digital twinning, radar threats, striker threats and terrain threats are considered when the flight paths of the unmanned aerial vehicles are generated, interference of the radar threats, the striker threats and the terrain threats on movement of the unmanned aerial vehicles is reduced, and the success rate of task completion of the unmanned aerial vehicles is increased. Relevant functions of radar threats, striker threats, terrain threats and energy consumption cost are determined through a multi-attribute utility theory MAUT, comprehensive evaluation is performed on the energy consumption cost and the related threats, so that a proper flight path is screened out, and when a target function is determined, the target function is determined by comparing utility scores of different flight paths. And optimal flight paths for executing different tasks are screened out. The flight path not only can avoid threats to the greatest extent, but also can ensure that the unmanned aerial vehicle arrives at a task area within specified time and effectively executes a task.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) path planning, and in particular to a multi-UAV control method based on digital twins. Background Art

[0002] The widespread use of drones can be traced back to many fields. When drones perform tasks, they involve multiple considerations such as path planning, safety, and efficiency. The transmission drone system faces problems such as insufficient information and slow response. It is difficult to adjust the action under the rapidly changing ring and cannot achieve efficient path planning. With the advantage of virtual-real interconnection, digital twin technology is gradually being combined with drones. By building a virtual drone model to simulate the behavior of physical drones throughout the entire cycle, monitor their status, optimize their mission processes, and improve the autonomous mission capabilities of drones. However, the flight paths generated between existing virtual and real drones do not take external threats into account, and the safety of the flight paths is low, which in turn leads to a low success rate for drones to complete their missions. Summary of the invention

[0003] In view of this, the purpose of the present invention is to provide a multi-UAV control method based on digital twins, which takes into account radar threats, strike object threats and terrain threats when generating the flight path of the UAV, reduces the interference of radar threats, strike object threats and terrain threats on the movement of the UAV, and improves the success rate of the UAV in completing the mission.

[0004] In order to solve the above technical problems, the technical solution used in the present invention is:

[0005] The multi-UAV control method based on digital twin described in the present invention comprises the following steps:

[0006] S1. Preset the UAV path planning standard and generate more than one flight path according to different tasks to be performed.

[0007] S2. Compare the utility scores of the flight paths corresponding to the same task to be executed, and set the flight path with the highest utility score as the preliminary flight path of the current task to be executed.

[0008] S3. Establish a digital twin platform for drones, construct twin data layer, virtual layer, user layer and service layer. The twin data layer collects historical flight data of drones; one virtual model in the virtual layer corresponds to one drone.

[0009] S4. The virtual layer optimizes the flight path and transmits the optimized flight path to the service layer so that the UAV moves along the optimized flight path.

[0010] S5. Synchronize the current flight data of the UAV in the digital twin platform and monitor the flight status of the UAV in real time through the virtual layer.

[0011] Preferably, S1 specifically includes the following models:

[0012] S1.1. Preset radar threat probability, strike threat probability and terrain threat model;

[0013] S1.2. Preset an energy consumption cost function, and determine the radar threat cost, strike threat cost and terrain threat cost in the flight path through the energy consumption cost function;

[0014] S1.3. Calculate the flight path of the drone through the objective function. The objective function is v(x) = ω R cost R +ω M cost M +ω E cost E +ω L cost L ; where ω R +ω M +ω E +ω L =1; cost R is the radar threat cost, cost M To combat the threat cost, cost E is the terrain threat cost, cost L is the energy consumption cost; R is the weight of radar threat cost, ω M is the weight of the threat cost of the strike object, ω E is the weight of terrain threat cost, ω L is the weight of energy consumption cost.

[0015] Preferably, the energy consumption cost of the preset drone is positively correlated with the flight length; the energy consumption cost function is: Among them l i is the length of segment i.

[0016] Preferably, the radar threat probability

[0017]

[0018] R is the one-way propagation distance of electromagnetic waves; R h is the maximum unidirectional propagation distance of electromagnetic waves at height h, R 4 It is the fourth power of the one-way propagation distance of electromagnetic waves; It is the fourth power of the maximum one-way propagation distance of electromagnetic waves at height h.

[0019] The formula for calculating the radar threat cost is:

[0020] m is the evenly divided flight segment, R ka For radar threats, L i is the flight segment, ka is the radar threat center, d ka is the distance between each segmentation point and ka, n R is the total number of radar threat points, and U is the number of different UAVs.

[0021] Preferably, the threat probability of the object

[0022]

[0023] RM h =RM max -kb|h-0.5H|, Kb=(RM max +RM min ) / 0.5H;RM max is the effective attack range of the object, RM min is the minimum attack radius, H is the attack height, RM h is the attack radius of the striking object at height h, RM is the horizontal attack distance, eM -1 is the attack probability of the strike object.

[0024] The formula for calculating the threat cost of a strike object is:

[0025]

[0026] m is the evenly divided flight segment, M kb To combat physical threats, L i is the flight segment, kb is the threat center of the object, d kb is the distance between each segmentation point and kb, n Rm is the total number of threat points of the strike object; U is different UAVs.

[0027] Preferably, authentication and secure communication are performed between a virtual model and a drone via a unique device key.

[0028] Preferably, MQTT is used for communication between a virtual model and a drone.

[0029] Preferably, S5 also includes S6; S6, if the physical environment changes, the flight path is regenerated, and then after obtaining the current flight data of the UAV on the current flight path, the current flight path is optimized through the virtual layer; the UAV moves along the optimized flight path.

[0030] Compared with the prior art, the beneficial effects of the handle connection structure described in the present invention are mainly reflected in: when driving the drone to perform a task, N drones take off from different areas and go to different mission areas to perform tasks, and then plan different flight paths according to different mission needs. After determining the target, radar threats, strike threats and terrain threats are considered when generating the flight path of the drone, reducing the interference of radar threats, strike threats and terrain threats on the movement of the drone, and improving the success rate of the drone in completing the task. The related functions of radar threats, strike threats, terrain threats and energy consumption costs are determined by the multi-attribute utility theory MAUT, and the energy consumption costs and related threat data are comprehensively evaluated to screen out a suitable flight path. When determining the objective function, the best flight path for performing different tasks is screened out by comparing the utility scores of different flight paths. This flight path can not only avoid threats to the greatest extent, but also ensure that the drone reaches the mission area within the specified time and effectively performs the task.

[0031] Through the digital twin platform, the corresponding flight paths of the planning department for different tasks are calculated, and the radar threat probability and the strike threat probability are determined to enable the UAV to avoid radar and strike objects; the threat boundary is determined through the terrain threat model, so that the UAV can avoid terrain obstacles. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The above and other purposes, features and advantages of the present invention will become more clear by more specific description of the preferred embodiments of the present invention shown in the accompanying drawings. The same reference numerals indicate the same parts in all the accompanying drawings, and the accompanying drawings are not deliberately scaled to the actual size, and the focus is on illustrating the main purpose of the present invention.

[0033] Figure 1 It is a schematic diagram of the control system in the present invention.

[0034] Figure 2 It is a flow chart of the control method in the present invention.

[0035] Figure 3 Schematic diagram of the flight path in the present invention. DETAILED DESCRIPTION

[0036] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments, so that those skilled in the art can better understand the present invention and implement it. However, the embodiments cited are not intended to limit the present invention. In the present embodiment, it should be understood that the orientations or positional relationships indicated by terms such as "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", and "outside" are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention.

[0037] It should be noted that when an element is considered to be "connected" to another element, it may be directly connected to the other element and integrated therewith, or there may be an intermediate element at the same time. The terms "installed", "one end", "the other end" and similar expressions used in the present invention are for illustrative purposes only.

[0038] like Figure 1-3 As shown, a multi-UAV control system based on digital twins includes a digital twin platform, which includes a virtual layer, a physical layer, a server layer, a user layer and a twin data layer connected by signals, wherein the physical layer is the UAV and the physical environment, and the virtual layer is the virtual model and the virtual environment; the twin data layer is provided for collecting historical flight data of the physical layer, the virtual layer is used to analyze the historical flight data and optimize the flight path, and the physical layer moves along the optimized flight path; when the physical layer moves along the optimized flight path, the twin data layer synchronizes the current flight data of the physical layer to the virtual layer, and the user layer detects the dynamics of the physical layer through the virtual layer; if it is necessary to actively control the flight of the UAV, the control signal is output through the user layer to actively control the flight of the UAV.

[0039] A multi-UAV control method based on digital twins includes the following steps:

[0040] S1. Preset the UAV path planning standard and generate more than one flight path according to different tasks to be performed; specifically including S1.1-S1.3.

[0041] S1.1. Preset radar threat probability, strike threat probability and terrain threat model.

[0042] S1.2. Preset an energy consumption cost function, and determine the radar threat cost, strike threat cost and terrain threat cost in the flight path through the energy consumption cost function.

[0043] S1.3. Calculate the flight path of the drone through the objective function. The objective function is v(x) = ω Rcost R +ω M cost M +ω E cost E +ω L cost L ; where ω R +ω M +ω E +ω L =1; cost R is the radar threat cost, cost M To combat the threat cost, cost E is the terrain threat cost, cost L is the energy consumption cost; R is the weight of radar threat cost, ω M is the weight of the threat cost of the strike object, ω E is the weight of terrain threat cost, ω L is the weight of the energy cost. In this embodiment, a threshold of the objective function v(x) is preset, and the path corresponding to the objective function v(x) being less than the threshold is set as the flight path, thereby reducing the impact of radar threat, strike threat, terrain threat and energy cost on the UAV. By presetting the threshold of the objective function v(x), multiple flight paths are generated for different tasks to be executed.

[0044] S2. Compare the utility scores of the flight paths corresponding to the same pending task, and set the flight path with the highest utility score as the preliminary flight path of the current pending task. In this embodiment, the utility scores of the flight paths corresponding to the same pending task are compared, specifically: compare the size of the objective function v(x) of the flight paths corresponding to the same pending task, and set the flight path with the smallest objective function v(x) as the preliminary flight path; the smaller v(x), the smaller the impact of radar threat, strike threat, and terrain threat on the UAV.

[0045] S3. Establish a digital twin platform for drones, collect historical flight data of drones, and construct a virtual layer, where a virtual model in the virtual layer corresponds to a drone. In a preferred embodiment, the flight data includes flight trajectory, flight speed, flight altitude, flight attitude, drone position, and environmental parameters; collecting flight data through a twin data layer is an existing technology.

[0046] S4. The virtual layer optimizes the preliminary flight path and transmits the optimized flight path to the UAV. In a preferred embodiment, the virtual layer optimizes the flight path through a model predictive control algorithm.

[0047] S5. A virtual model and a drone are authenticated and communicate securely through a unique device key; the current flight data of the drone is synchronized in the digital twin platform, and the flight status of the drone is monitored in real time through the virtual layer.

[0048] S6. If the physical environment changes, the flight path is regenerated, and then after obtaining the current flight data of the UAV on the current flight path, the current flight path is optimized through the virtual layer; the UAV moves along the optimized flight path.

[0049] The above method drives the UAV to perform tasks. N UAVs take off from different areas and go to different mission areas to perform tasks. Different flight paths are planned according to different missions. After determining the target, radar threats, strike threats and terrain threats are considered when generating the flight path of the UAV, reducing the interference of radar threats, strike threats and terrain threats on the movement of the UAV and improving the success rate of the UAV in completing the mission. The related functions of radar threats, strike threats, terrain threats and energy consumption costs are determined by multi-attribute utility theory MAUT, and the energy consumption costs and related threat data are comprehensively evaluated to screen out the appropriate flight path.

[0050] When determining the objective function, the utility scores of different flight paths are compared to select the best flight path for performing different tasks. This flight path can not only avoid threats to the greatest extent, but also ensure that the drone reaches the mission area within the specified time and effectively performs the mission.

[0051] Through the digital twin platform, the corresponding flight paths of the planning department for different tasks are calculated, and the radar threat probability and the strike threat probability are determined to enable the UAV to avoid radar and strike objects; the threat boundary is determined through the terrain threat model, so that the UAV can avoid terrain obstacles.

[0052] Energy consumption is related to the power consumption from the starting point to the end point. In a preferred embodiment, the altitude and speed of the drone are constant, and the energy consumption cost of the drone is positively correlated with the flight length; the energy consumption cost function is: Among them l i is the length of segment i.

[0053] In a preferred embodiment, the radar threat probability

[0054]

[0055] R is the one-way propagation distance of electromagnetic waves; R h is the maximum unidirectional propagation distance of electromagnetic waves at height h, R 4 It is the fourth power of the one-way propagation distance of electromagnetic waves; It is the fourth power of the maximum one-way propagation distance of the electromagnetic wave at height h. Since the electromagnetic waves used by radar propagate in a straight line, and the energy loss of the emitted electromagnetic wave after being reflected by the target is equal to the fourth power of the one-way propagation distance of the electromagnetic wave R 4 It is inversely proportional to the probability of radar detecting the target and is related to the Poisson distribution, and then the radar threat probability is calculated through this formula.

[0056] The formula for calculating the radar threat cost is:

[0057]

[0058] m is the evenly divided flight segment, R ka For radar threats, L i is the flight segment, ka is the radar threat center, d ka is the distance between each segmentation point and ka, n R is the total number of radar threat points, and U is a different UAV. When a UAV moves along the flight path, it is necessary to consider whether the UAV is in the threat area, divide the flight segment into m parts, determine the impact of the threat center on the UAV, and thus determine the radar threat cost.

[0059] In a preferred embodiment, the probability of a pre-set striking object attacking a UAV is Poisson distributed with respect to the horizontal attack distance R, and the probability of a striking object attacking a UAV is eM -1 The striking object is launched outward with the striking object launching device as the center.

[0060] Threat probability of the striker

[0061]

[0062] RM h =RM max -kb|h-0.5H|, Kb=(RM max +RM min ) / 0.5H;RM max is the effective attack range of the object, RM min is the minimum attack radius, H is the attack height, RM h is the attack radius of the striking object at height h, RM is the horizontal attack distance, eM -1 is the attack probability of the strike object; in order to prevent the UAV from being hit by the strike object during movement, the attack range of the strike object at the height h is determined to determine the threat cost of the strike object.

[0063] The formula for calculating the threat cost of a strike object is:

[0064]

[0065] m is the evenly divided flight segment, M k To combat physical threats, L i is the flight segment, kb is the threat center of the object, d kb is the distance between each segmentation point and kb, n Rm is the total number of threat points of the strike object; U is different UAVs.

[0066] In a preferred embodiment, the terrain threat cost is confirmed by a terrain threat model, and the terrain threat model is established based on terrain elevation data and terrain relief data; the setting methods of the terrain threat cost and the terrain threat model are both existing technologies and will not be described here.

[0067] Reference Figure 3 As shown, the flight path generated in S1.3 is explained:

[0068] A1, A2, A3 and A4 are drones that perform different tasks respectively; the circular areas in the figure represent radar threats, strike threats or terrain threats; while reducing radar threats, strike threats and terrain threats, drones also add energy consumption costs to shorten the distance between the starting point Q and the target point.

[0069] Table 2 shows the flight path planning data of UAV A1, UAV A2, UAV A3 and UAV A4:

[0070]

[0071] Table 2

[0072] In the above S3, the drone includes an existing drone, and the specific structure of the drone is not described here; the drone is connected to the service layer by signals, and the drone receives instructions from the service layer to adjust the flight action. The drone is equipped with sensors, and the sensors are used to collect flight trajectory data, flight speed data, flight altitude data, flight attitude data, drone position data and environmental parameter data.

[0073] The virtual model includes more than one part, and the shape and appearance of each part are adapted to the shape and appearance of each part of the drone, so that the shape of the virtual model is similar to the shape of the drone entity.

[0074] In a preferred embodiment, the virtual model is developed through the Unity3D physics engine platform, and the simulation component is configured to simulate the functions of the physical drone. The virtual model developed through the Unity3D physics engine platform can simulate the lateral linear motion, longitudinal linear motion, vertical linear motion, roll rotational motion, pitch rotational motion and yaw rotational motion of the drone. At the same time, a collider component is provided to simulate the effect of the drone entity colliding with obstacles in the environment. The Unity3D physics engine is used as the expression form of the virtual layer to realize the interaction between the virtual layer and the physical layer. At the same time, based on the Euler equation, the motion characteristics in the three directions of roll, pitch and yaw are analyzed, and mathematical equations are used to describe the lateral motion, longitudinal motion and vertical motion. The confirmation method of the lateral motion, longitudinal motion and vertical motion is a prior art and will not be repeated here.

[0075] In S5, in a preferred embodiment, a virtual model and a drone communicate using MQTT (Message Queuing Telemetry Transport).

[0076] Table 1 is a comparison table of delay times for different communication network modes:

[0077] Communication network mode Number of drones Delay time (ms) Packet size (byte) WIFI-4G 30 65.7-78.8 124 WIFI-5G 30 38.2-45.3 124 Cloud Services 30 16.4-26.7 124

[0078] Table 1

[0079] As shown in Table 1, by using message queue telemetry transmission, the messages are serialized, the data packet size is reduced, and the communication delay is reduced.

[0080] Each drone has a unique ID, and the ID of the virtual model corresponds to the ID of the drone; this makes it easy to distinguish different drone entities, and through the association of IDs, the physical drone corresponding to the virtual entity can be accurately found. The unique device key is used for identity authentication and secure communication between a virtual model and a drone, ensuring a close connection between the physical layer and the virtual layer, so that the virtual layer can obtain and update the status information of the drone in real time. Using device keys for identity authentication and secure communication is a prior art and will not be repeated here.

[0081] In the service layer, the configuration data of each virtual drone is stored, including but not limited to flight parameters, sensor data and communication protocols. These configuration data can be adjusted according to actual needs, which facilitates subsequent upgrades and optimizations.

[0082] S5 also includes that in a preferred embodiment, the user layer uses a VR device to connect to the virtual layer signal, so that the operator of the user layer can directly interact with the virtual model. If the UAV flight needs to be actively controlled, the UAV flight is actively controlled by outputting a control signal through the user layer. The user layer and the virtual layer are connected by a VR device, and the user can directly interact with the virtual model, and control the UAV by manipulating the virtual model. When the real-time aircraft data of the physical layer is synchronized to the virtual layer, the user layer monitors the flight status of the UAV through the VR device, and can manipulate the virtual model through the VR device. Then the virtual model sends the control signal to the physical layer through the service layer, thereby realizing the user layer's active control of the UAV. The VR device is a prior art and will not be described here.

[0083] In this specification, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0084] In the description of this specification, the description with reference to the terms "preferred embodiment", "further embodiment", "other embodiments" or "specific example" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0085] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A multi-UAV control method based on digital twins, characterized by: The following steps are involved: S1. Preset the UAV path planning standard and generate more than one flight path according to different tasks to be performed; S2, comparing the utility scores of the flight paths corresponding to the same task to be executed, and setting the flight path with the highest utility score as the preliminary flight path of the current task to be executed; S3. Establish a digital twin platform for drones, construct twin data layer, virtual layer, user layer and service layer. The twin data layer collects historical flight data of drones; one virtual model in the virtual layer corresponds to one drone; S4, the virtual layer optimizes the flight path and transmits the optimized flight path to the service layer so that the UAV moves along the optimized flight path; S5. Synchronize the current flight data of the UAV in the digital twin platform and monitor the flight status of the UAV in real time through the virtual layer.

2. The multi-UAV control method based on digital twin according to claim 1 is characterized in that: S1 specifically includes the following models: S1.

1. Preset radar threat probability, strike threat probability and terrain threat model; S1.

2. Preset an energy cost function, and determine the radar threat cost, strike threat cost and terrain threat cost in the flight path through the energy cost function; S1.

3. Calculate the flight path of the drone through the objective function. The objective function is V(x) = ω R cost R +ω M cost M +ω E cost E +ω L cost L ; where ω R +ω M +ω E +ω L =1; cost R is the radar threat cost, cost M To combat the threat cost, cost E is the terrain threat cost, cost L is the energy consumption cost; R is the weight of radar threat cost, ω M is the weight of the threat cost of the strike object, ω E is the weight of terrain threat cost, ω L is the weight of energy consumption cost.

3. The multi-UAV control method based on digital twin according to claim 2 is characterized in that: The energy consumption cost of the preset UAV is positively correlated with the flight length; the energy consumption cost function is: Among them l i is the length of segment i.

4. The multi-UAV control method based on digital twin according to claim 3 is characterized in that: Radar threat probability R is the one-way propagation distance of electromagnetic waves; R h is the maximum unidirectional propagation distance of electromagnetic waves at height h, R 4 It is the fourth power of the one-way propagation distance of electromagnetic waves; is the fourth power of the maximum one-way propagation distance of electromagnetic waves at height h; The formula for calculating the radar threat cost is: m is the evenly divided flight segment, R ka For radar threats, L i is the flight segment, ka is the radar threat center, d ka is the distance between each segmentation point and ka, n R is the total number of radar threat points, and U is the number of different UAVs.

5. The multi-UAV control method based on digital twin according to claim 3 is characterized in that: Threat probability of the striker RM h =RM max -kb|h-0.5H|, Kb=(RM max +RM min ) / 0.5H;RM max is the effective attack range of the object, RM min is the minimum attack radius, H is the attack height, RM h is the attack radius of the striking object at height h, RM is the horizontal attack distance, e -1 is the attack probability; The formula for calculating the threat cost of an attack object is: m is the evenly divided flight segment, M kb To combat physical threats, L i is the flight segment, kb is the threat center of the object, d kb is the distance between each segmentation point and kb, n Rm is the total number of threat points of the strike object; U is different UAVs.

6. The multi-UAV control method based on digital twin according to claim 1 is characterized in that: A virtual model and a drone are authenticated and communicate securely via a unique device key.

7. The multi-UAV control method based on digital twin according to claim 3 is characterized in that: A virtual model and a drone communicate using MQTT.

8. The multi-UAV control method based on digital twin according to claim 1 is characterized in that: After S5, there is also S6; S6. If the physical environment changes, the flight path is regenerated, and then after obtaining the current flight data of the UAV on the current flight path, the current flight path is optimized through the virtual layer; the UAV moves along the optimized flight path.

Images