Unmanned platform state updating method for large-scale cluster simulation

Abstract

The invention relates to an unmanned platform state updating method for large-scale cluster simulation, belongs to the technical field of unmanned platform cluster simulation, and solves the problems of large calculation amount and low speed when the existing unmanned platform cluster simulation is used for updating the state of an unmanned platform. Comprising the following steps: obtaining the type of each unmanned platform in an unmanned platform cluster and a control instruction of each unmanned platform; dividing the unmanned platform cluster into a rotor wing cluster and a fixed wing cluster based on the type of each unmanned platform, and initializing the state of each cluster; the cluster state comprises a cluster linear state and a cluster attitude state; and based on the cluster type, each cluster state at the current simulation moment and the control instruction of each unmanned platform, performing state updating of each cluster, taking each updated cluster state as each cluster state at the next simulation moment, and repeatedly executing the step to perform state updating until the simulation is finished. And completing state updating of the unmanned platform state at each simulation moment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned platform cluster simulation, and particularly relates to a method for updating the state of an unmanned platform for large-scale cluster simulation. BACKGROUND

[0002] In recent years, there has been a trend of using unmanned aerial vehicle (UAV) clusters to handle complex tasks rather than a single complex UAV. A UAV cluster is a distributed system composed of a large number of UAVs. The cluster relies on the cooperative control function of the platforms to improve the overall task capability of the cluster. In the cluster, almost every UAV no longer has global information, but communicates with surrounding UAVs to achieve group behavior. This form of cluster has less demand for a center or master individual than a traditional cluster, so even if some UAVs are damaged, the cluster still has the ability to perform tasks.

[0003] Currently, in order to ensure that the UAV platform cluster can better perform tasks in the actual environment, the UAV platform cluster is usually simulated and tested. However, in the existing UAV platform cluster simulation system, the state of the UAV platform is updated one by one during the simulation process, and the dynamics model is mostly a nonlinear model, which has a large amount of calculation and slow speed, and cannot support large-scale precise dynamics equation calculation of more than 1000 nodes. Therefore, there is an urgent need for a method for updating the state of an unmanned platform for large-scale cluster simulation. SUMMARY

[0004] In view of the above analysis, the embodiments of the present application aim to provide a method for updating the state of an unmanned platform for large-scale cluster simulation, to solve the problems of large amount of calculation and slow speed in updating the state of the unmanned platform in the existing UAV platform cluster simulation.

[0005] The embodiments of the present application provide a method for updating the state of an unmanned platform for large-scale cluster simulation, comprising the following steps:

[0006] S1, obtaining the types of each unmanned platform in the unmanned platform cluster and the control instructions of each unmanned platform; wherein the control instructions include a desired position or a desired speed, and the types of the unmanned platforms include rotary-wing UAVs and fixed-wing UAVs;

[0007] S2, dividing the unmanned platform cluster into rotary-wing clusters and fixed-wing clusters based on the types of each unmanned platform, and initializing the states of each cluster; the states of the clusters include cluster linear states and cluster attitude states;

[0008] S3, based on the cluster type, the state of each cluster at the current simulation time and the control instruction of each unmanned platform, updating the state of each cluster, and repeating the step S3 to update the state of each cluster until the simulation ends, and completing the state update of the unmanned platform at each simulation time.

[0009] Compared with the prior art, the present application can achieve at least one of the following beneficial effects:

[0010] The unmanned platform state update method for large-scale cluster simulation provided by the present application can support the state calculation of a super-large unmanned system cluster with thousands of nodes, consume less hardware resources, and support the state calculation of a cluster with thousands of nodes by a consumer-grade image processor.

[0011] The above technical solutions can be combined with each other in the present application to achieve more preferred combination solutions. Other features and advantages of the present application will be described in the subsequent specification, and some advantages will become apparent from the specification, or will be understood by implementing the present application. The purposes and other advantages of the present application can be achieved and obtained from the contents specifically pointed out in the specification and the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0012] The accompanying drawings are included to provide a further understanding of the present application, and are incorporated herein and constitute a part of the detailed description. The drawings illustrate embodiments of the present application and, together with the description, serve to explain the principles of the present application. In the drawings:

[0013] Figure 1 The unmanned platform state update method for large-scale cluster simulation provided by the present application;

[0014] Figure 2 The distributed simulation middleware schematic diagram provided by the present application;

[0015] Figure 3 The parallel computing schematic diagram provided by the present application;

[0016] Figure 4 The rotor cluster state update schematic diagram provided by the present application;

[0017] Figure 5A fixed-wing cluster state updating schematic diagram provided for an embodiment of the present application is shown in the figure;

[0018] Figure 6 A rotor unmanned aerial vehicle physical model composition schematic diagram provided for an embodiment of the present application is shown in the figure;

[0019] Figure 7 A power unit model schematic diagram provided for an embodiment of the present application is shown in the figure;

[0020] Figure 8 A multi-rotor unmanned aerial vehicle force analysis schematic diagram provided for an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0021] The preferred embodiments of the present application will be described in detail below with reference to the drawings, wherein the drawings constitute a part of this application and are used to explain the principles of the embodiments of the present application, but are not used to limit the scope of the present application.

[0022] One specific embodiment of the present application discloses a state updating method for large-scale cluster simulation of unmanned platforms, as shown in the figure, comprising the following steps: Figure 1

[0023] S1, obtaining the types of each unmanned platform in the unmanned platform cluster and the control instructions of each unmanned platform; wherein the control instructions include expected positions or expected speeds, and the types of the unmanned platforms include rotor unmanned aerial vehicles and fixed-wing unmanned aerial vehicles;

[0024] S2, dividing the unmanned platform cluster into rotor clusters and fixed-wing clusters based on the types of each unmanned platform, and initializing the states of each cluster; the cluster states include cluster linear states and cluster attitude states;

[0025] S3, based on the cluster types, the states of each cluster at the current simulation time, and the control instructions of each unmanned platform, updating the states of each cluster, and repeating step S3 to update the states of each cluster as the states of each cluster at the next simulation time until the simulation is completed, thereby completing the state updating of the unmanned platform states at each simulation time.

[0026] Specifically, the position mode and the speed mode in the simulation system can be switched with each other; in the position mode, the control instruction is the expected position; in the speed mode, the control instruction is the expected speed.

[0027] In implementation, the updating of the states of each cluster based on the cluster types, the states of each cluster at the current simulation time, and the control instructions of each unmanned platform comprises:

[0028] ​If the cluster type is a rotor cluster, based on the rotor cluster state at the current simulation time and the control instructions of each unmanned platform in the cluster, the cluster linear state update and the cluster attitude state update of the cluster are performed;

[0029] If the cluster type is a fixed-wing cluster, based on the fixed-wing cluster state at the current simulation time and the control instructions of each unmanned platform in the cluster, the cluster linear state and the cluster attitude state update of the cluster are performed.

[0030] In a specific implementation, if the cluster type is a rotor cluster, the cluster linear state update is performed in the following manner:

[0031] Based on the rotor cluster state at the current simulation time and the control instructions of each unmanned platform in the cluster, the rotor cluster linear state change rate is obtained;

[0032] The rotor cluster linear state change rate is integrated to obtain the updated cluster linear state of the rotor cluster; wherein,

[0033] The elements in the cluster linear state of the rotor cluster are linear state vectors of each rotor unmanned aerial vehicle, and the linear state vector includes the position, velocity, and sum of the rotation speeds of the propellers of the rotor unmanned aerial vehicle.

[0034] Specifically, the rotor unmanned aerial vehicle includes four propellers; the rotor cluster linear state change rate is expressed as:

[0035]

[0036] wherein,

[0037]

[0038] In the formula, X represents the rotor cluster linear state at the current simulation time, X' represents the rotor cluster linear state increment input at the current simulation time, M i represents the dynamic coefficient vector of the i-th rotor unmanned aerial vehicle in the rotor cluster, n represents the total number of rotor unmanned aerial vehicles in the rotor cluster, represents the linear state vector change rate of the i-th rotor unmanned aerial vehicle in the rotor cluster, X i represents the linear state vector of the i-th rotor unmanned aerial vehicle in the rotor cluster at the current simulation time, X i ' represents the linear state vector increment input of the i-th rotor unmanned aerial vehicle in the rotor cluster at the current simulation time, P i , V i respectively represent the position and velocity of the i-th rotor unmanned aerial vehicle in the rotor cluster at the current simulation time, ω i,k represents the propeller rotation speed of the k-th propeller of the i-th rotor unmanned aerial vehicle in the rotor cluster at the current simulation time, respectively, T represents a transpose, wherein the propeller rotating speed of each propeller of the rotor unmanned aerial vehicle at the current simulation moment is obtained based on a control instruction.

[0039] It should be noted that the propeller rotating speed of each propeller of each rotor unmanned aerial vehicle at the current simulation moment is obtained based on a control instruction, and a control system in the simulation system calculates the relative speed of each propeller based on the control instruction of the rotor unmanned aerial vehicle through a dynamic model of the rotor unmanned aerial vehicle and a control algorithm, wherein the control algorithm includes PID, model predictive control, self-disturbance control, etc., and the rotating speed of the propeller is dynamically adjusted according to the difference between the current position or speed of the rotor unmanned aerial vehicle and the expected position or speed, and the process can be realized by the prior art, which will not be described here.

[0040] More specifically, the air resistance factor of the i-th rotor unmanned aerial vehicle in the rotor cluster and the propeller rotating speed factor are respectively represented as:

[0041]

[0042] In the formula, m i represents the mass of the i-th rotor unmanned aerial vehicle in the rotor cluster, p represents the air density of a flight environment, A i represents the cross-sectional area of the i-th rotor unmanned aerial vehicle in the rotor cluster, represents the air resistance coefficient, D i,p represents the propeller diameter of the i-th rotor unmanned aerial vehicle in the rotor cluster, C T represents the tension coefficient. In specific implementation, if the cluster type is a rotor cluster, the cluster attitude state is updated in the following manner:

[0043] Based on the rotor cluster state at the current simulation moment, a first change rate of the rotor cluster attitude state in the unmanned aerial vehicle coordinate system is obtained;

[0044] The first change rate of the rotor cluster attitude state in the unmanned aerial vehicle coordinate system is integrated to obtain a cluster attitude angular velocity, and then the cluster attitude angular velocity is coordinate-converted to obtain a second change rate of the rotor cluster attitude state in the geodetic coordinate system; wherein the elements in the cluster attitude angular velocity are the attitude angular velocities of each rotor unmanned aerial vehicle.

[0045] The second change rate of the rotor cluster attitude state in the geodetic coordinate system is integrated to obtain the updated cluster attitude state of the rotor cluster; wherein,

[0046] The elements in the cluster attitude state are the attitude angles of each rotor unmanned aerial vehicle.

[0047] Specifically, the attitude angles include a yaw angle, a pitch angle, and a roll angle.

[0048] Specifically, the first change rate of the rotor cluster attitude state is represented as:

[0049]

[0050] wherein,

[0051] in the formula, represents an attitude angle acceleration of the i-th rotor unmanned aerial vehicle in the rotor cluster at the current simulation moment in the unmanned aerial vehicle coordinate system, are respectively components of the x, y, and z axes in the body coordinate system; J i represents a moment of inertia of the i-th rotor unmanned aerial vehicle in the rotor cluster, represents an attitude angle velocity of the i-th rotor unmanned aerial vehicle in the rotor cluster at the current simulation moment in the body coordinate system, wherein, are respectively components of the lateral axis, the longitudinal axis, and the vertical axis in the body coordinate system; τ i represents a gyroscopic moment of the i-th rotor unmanned aerial vehicle in the rotor cluster, wherein, τ i is defined as τ i,φ , τ i,θ , and τ i,ψ are respectively gyroscopic moments of the i-th rotor unmanned aerial vehicle in the rotor cluster in the roll channel, the pitch channel, and the yaw channel; represents a propeller moment of the i-th rotor unmanned aerial vehicle in the rotor cluster, and is defined as are respectively components of the x, y, and z axes in the body coordinate system.

[0052] More specifically, the propeller moment of the i-th rotor unmanned aerial vehicle in the rotor cluster is represented as:

[0053]

[0054] in the formula, C τ represents a torque coefficient, and d represents a rotor spacing of the rotor unmanned aerial vehicle.

[0055] Specifically, the second change rate of the rotor cluster attitude state is represented as:

[0056]

[0057] wherein,

[0058] wherein, represents the attitude angular velocity of the i-th rotor unmanned aerial vehicle in the rotor cluster at the current simulation moment in the earth coordinate system, wherein, respectively represent the change rates of the roll angle the pitch angle and the yaw angle of the i-th rotor unmanned aerial vehicle in the rotor cluster at the current simulation moment; represents the attitude angular velocity of the i-th rotor unmanned aerial vehicle in the rotor cluster at the current simulation moment in the unmanned aerial vehicle coordinate system; W i represents the transformation matrix of the i-th rotor unmanned aerial vehicle in the rotor cluster.

[0059] In specific implementation, if the cluster type is a fixed-wing cluster, the cluster linear state is updated in the following manner:

[0060] based on the fixed-wing cluster state at the current simulation moment and the control instructions of each unmanned platform in the cluster, a fixed-wing cluster linear state change rate is obtained;

[0061] integrating the fixed-wing cluster linear state change rate, an updated fixed-wing cluster cluster linear state is obtained; wherein,

[0062] the elements in the cluster linear state are linear state vectors of each fixed-wing unmanned aerial vehicle, and the linear state vector includes a roll state quantity, a pitch state quantity, a yaw state quantity, a forward state quantity, an up-down state quantity and a left-right state quantity.

[0063] Specifically, the roll state quantity is the angle of rotation around the longitudinal axis of the unmanned aerial vehicle body from the left side to the right side, the pitch state quantity is the angle of rotation around the transverse axis of the unmanned aerial vehicle body from the front to the back, and the yaw state quantity is the angle of rotation around the vertical axis of the unmanned aerial vehicle body from the top to the bottom; the forward state quantity is the length of linear motion along the longitudinal axis of the unmanned aerial vehicle body, the up-down state quantity is the length of linear motion along the vertical axis of the unmanned aerial vehicle body, and the left-right state quantity is the length of linear motion along the transverse axis of the unmanned aerial vehicle body.

[0064] Specifically, the fixed-wing cluster linear state change rate is represented as:

[0065]

[0066] wherein,

[0067]

[0068] wherein X f represents the linear state of the fixed-wing swarm at the current simulation time, represents the rate of change of the linear state of the fixed-wing swarm at the current simulation time, represents the first dynamic coefficient vector of the jth fixed-wing UAV in the fixed-wing swarm at the current simulation time, A j represents the second dynamic coefficient vector of the jth fixed-wing UAV in the fixed-wing swarm at the current time, n' represents the total number of fixed-wing UAVs in the fixed-wing swarm, represents the rate of change of the linear state vector of the jth fixed-wing UAV in the fixed-wing swarm, represents the linear state vector of the jth fixed-wing UAV in the fixed-wing swarm at the current simulation time, respectively represent the roll state quantity, the pitch state quantity, the yaw state quantity, the forward state quantity, the up-down state quantity and the left-right state quantity of the jth fixed-wing UAV in the fixed-wing swarm at the current simulation time, respectively represent the rate of change of the roll state quantity, the pitch state quantity, the yaw state quantity, the forward state quantity, the up-down control quantity and the left-right state quantity of the jth fixed-wing UAV in the fixed-wing swarm, T represents transposition.

[0069] More specifically, the first dynamic coefficient vector of the jth fixed-wing UAV in the fixed-wing swarm at the current simulation time A is represented as:

[0070]

[0071] wherein respectively represent the roll angle, the pitch angle and the yaw angle of the jth fixed-wing UAV in the fixed-wing swarm at the current simulation time, p j , q j , r j respectively represent the components of the attitude angular velocity of the jth rotor UAV in the fixed-wing swarm at the current simulation time in the body coordinate system in the lateral axis, the longitudinal axis and the vertical axis.

[0072] More specifically, the second dynamic coefficient vector of the jth fixed-wing UAV in the fixed-wing swarm at the current time A j is represented as:

[0073]

[0074] wherein g represents the acceleration of gravity, F j,x , F j,y , F j,z respectively represent the components of the resultant force of the jth fixed-wing UAV in the fixed-wing swarm at the current time in the body coordinate system in the lateral axis, the longitudinal axis and the vertical axis, represents the weight of the jth fixed-wing UAV in the fixed-wing cluster; wherein the resultant force of each fixed-wing UAV in the fixed-wing cluster is obtained based on the control instruction.

[0075] It should be noted that the resultant force of each fixed-wing UAV in the fixed-wing cluster is obtained based on the control instruction, in the control system in the simulation system, based on the control instruction of the fixed-wing UAV, the relative speed of each propeller is calculated through the dynamics model and control algorithm of the fixed-wing UAV, wherein the control algorithm includes PID, model predictive control, active disturbance rejection control, etc., the speed of the propeller is dynamically adjusted according to the difference between the current position or speed of the fixed-wing UAV and the expected position or speed, and then the total air force, thrust and gravity of the fixed-wing UAV are obtained based on the speed of each propeller of the fixed-wing UAV through dynamics and aerodynamics, and the resultant force is obtained by adding up. The above process is realized by the prior art, and will not be described in detail in the embodiment.

[0076] In specific implementation, the elements in the cluster attitude state of the fixed-wing cluster are the attitude angles of each fixed-wing UAV, wherein the attitude angles include the yaw angle, the pitch angle and the roll angle.

[0077] Specifically, if the cluster type is a fixed-wing cluster, the attitude of each fixed-wing UAV is updated based on the attitude of each fixed-wing UAV at the current simulation time of the fixed-wing cluster, and then the updated cluster attitude state is obtained; wherein the attitude of each fixed-wing UAV is updated by integrating the attitude change rate of each fixed-wing UAV, and the attitude change rate is represented as:

[0078]

[0079] In the formula, are the change rates of the roll angle, the pitch angle and the yaw angle of the jth fixed-wing UAV in the fixed-wing cluster at the current simulation time, respectively.

[0080] In specific implementation, the simulation platform system in the embodiment is a network communication service system constructed uniformly by an angle system, and the network communication subsystem is planned and constructed as the message bus of the entire simulation platform. By connecting each business application with the network communication subsystem to realize network communication between each other, the topology structure of the network communication network of the simulation platform is simplified, the network communication mechanism is unified, the management, maintenance and upgrade difficulty of the network communication link can be effectively reduced, and the scalability of the system is improved.

[0081] The solving module uses a data distribution service of publish / subscribe architecture, is data-centric, can guarantee real-time, efficient and flexible distribution of data, and can meet the computing requirements of distributed simulation systems. The middleware uses a centerless communication architecture and provides a data-centric publish / subscribe communication model. In the publish / subscribe mode, the simulation communication nodes only need to subscribe to the data they are interested in or publish the data they can provide. Logically, data is directly propagated between simulation communication nodes, so it is very simple to establish a communication connection: the publisher registers the data it can provide, and the subscriber subscribes to the data it is interested in; the publisher only cares about the specific data it wants to deliver when publishing data; the subscriber only cares about the specific data it wants to receive when receiving information. The communication entities in the publish / subscribe mechanism have an asynchronous and loosely coupled relationship. Since data flows directly from the publisher to the subscriber without the participation of an intermediate server, it has very high transmission efficiency. From the perspectives of development and maintenance and communication performance, the application of the publish / subscribe mechanism to information interaction can better meet the communication requirements of current large-scale distributed simulation systems. As can be seen, the publish / subscribe model uses an event-based architecture, which is more convenient for dynamic expansion.

[0082] The advantages of the publish / subscribe mode are:

[0083] 1) The publisher and the subscriber do not need to know the existence of each other;

[0084] 2) A subscriber can receive data from multiple different publishers, and a publisher can send data to multiple different subscribers, which can realize many-to-many communication;

[0085] 3) The publisher and the subscriber do not need to keep synchronization, i.e., they do not need to be active at the moment of exchanging data, because there is a data-centric interface that supports storing and forwarding data to directly receive and send data.

[0086] As shown in Figure 2 , the system framework of the distributed simulation middleware based on the data distribution service is given. The framework includes two parts: one part is the data transmission related module, which is composed of a data transmission management interface, a communication entity manager, a data model manager and related API interfaces, and realizes the data transmission function based on the network communication subsystem; the other part is the simulation solving service related module, which mainly realizes the related simulation solving function on the basis of the communication middleware.

[0087] Based on the simulation middleware and the dynamics model of the unmanned system, the state solving part uses GPU parallel computing, uses the Cuda parallel computing architecture of Nvida to realize large-scale state solving, and the flow chart is as shown in Figure 3 , including the following steps:

[0088] a1, initializing the simulation environment and starting the rendering node, wherein the node responsible for graphics rendering is started to visualize the behavior of the unmanned platform during the simulation;

[0089] a2, starting the communication node, wherein the node responsible for communication is started to ensure information transmission between unmanned platforms and between the platform and the control center;

[0090] a3, allocating computing nodes according to the type and number of unmanned platforms, wherein the corresponding computing resources are allocated according to the type (such as rotor unmanned aerial vehicle, fixed-wing unmanned aerial vehicle) and number of unmanned platforms in the simulation;

[0091] a4, starting the computing node, wherein the allocated computing node is started, which will be responsible for processing the computing task of the unmanned platform;

[0092] a5, determining whether the simulation has reached the predetermined end condition, if so, the simulation stops; otherwise, updating the simulation clock to simulate the passage of time;

[0093] a6, when the rotor cluster node receives the clock information, the rotor cluster node state is updated, that is, the state of the rotor unmanned aerial vehicle cluster is updated according to the received clock information and the current task state, and then the latest state of the rotor unmanned aerial vehicle cluster is published to other nodes or the control center;

[0094] a7, when the fixed-wing cluster node receives the clock information, the fixed-wing cluster node state is updated, that is, the state of the fixed-wing unmanned aerial vehicle cluster is updated according to the received clock information and the current task state, and then the latest state of the fixed-wing unmanned aerial vehicle cluster is published to other nodes or the control center.

[0095] a8, return to step a5 and continue to check whether the simulation is ended, repeat steps a5-a8 until the simulation is ended.

[0096] Wherein, the state updating process of the rotor and fixed-wing unmanned aerial vehicle cluster nodes is shown in Figure 4 and Figure 5 respectively, which is updated based on the updating rule of the embodiment in the simulation system.

[0097] Specifically, the state updating process of the rotor unmanned aerial vehicle cluster node includes the following steps:

[0098] b1, starting the cluster simulation system, initializing the cluster parameters, that is, setting the initial parameters of the simulation, including the initial state and related configuration of the cluster.

[0099] b2, allocating Host memory and Device memory, that is, allocating the required memory space for Host (host, such as CPU) and Device (device, such as GPU).

[0100] b3, initialize Host data (initial state of the cluster), that is, initialize the initial state data of the cluster in the Host memory.

[0101] b4, determine whether the simulation is ended, if yes, the simulation is ended; otherwise, determine whether the clock is updated, if no, wait for the clock to be updated; if yes, then

[0102] copy the Host data to the Device, that is, copy the state data of the cluster at the current time from the Host memory to the Device memory, and then perform the linear state update of the cluster, that is, calculate the linear state of the cluster on the Device, and then perform the attitude state update of the cluster, that is, calculate the attitude state of the cluster on the Device;

[0103] copy the Device data to the Host, that is, copy the state data of the cluster at the next time from the Device memory back to the Host memory;

[0104] b5, return to step b4, continue to check whether the simulation is ended, repeat steps b4-b5 until the simulation is ended.

[0105] Specifically, the state update process of the fixed-wing unmanned aerial vehicle cluster node includes the following steps:

[0106] c1, start the cluster simulation system, and initialize the cluster parameters, that is, set the initial parameters of the simulation, including the initial state of the cluster and related configurations.

[0107] c2, allocate the Host memory and the Device memory, that is, allocate the required memory space for the Host (host, such as CPU) and the Device (device, such as GPU).

[0108] c3, initialize the Host data (initial state of the cluster), that is, initialize the initial state data of the cluster in the Host memory.

[0109] c4, determine whether the simulation is ended, if yes, the simulation is ended; otherwise, determine whether the clock is updated, if no, wait for the clock to be updated; if yes, then

[0110] copy the Host data to the Device, that is, copy the state data of the cluster at the current time from the Host memory to the Device memory, and then perform the linear state update of the cluster, that is, calculate the linear state of the cluster on the Device;

[0111] perform the attitude state update on each unmanned platform in the cluster to obtain the updated attitude state of the cluster;

[0112] Copy the Device data to the Host, i.e. copy the cluster state data of the next time instant from the Device memory back to the Host memory;

[0113] b5, return to step b4, continue to check whether the simulation is ended, repeat steps b4-b5 until the simulation is ended.

[0114] It should be noted that the method in the embodiment is derived by the following:

[0115] The physical modeling of the unmanned aerial vehicle is mainly aimed at the rotor, fixed wing and vertical take-off and landing unmanned aerial vehicle. The vertical take-off and landing unmanned aerial vehicle can be simplified as a rotor flight in the take-off and landing stage and as a fixed wing flight in the high-speed flight stage. Therefore, the physical modeling mainly includes two parts of the rotor and the fixed wing.

[0116] ①Physical model of the rotor unmanned aerial vehicle:

[0117] As shown in Figure 6 , it mainly includes four parts of a power unit model, a control efficiency model, a rigid body kinematics model and a rigid body dynamics model.

[0118] The power unit model: the power unit is a whole power mechanism with a brushless direct current motor, an electronic speed controller and a propeller as a group. The input is a throttle command between 0 and 1, and the output is the propeller speed. In practice, a model with the input being the throttle command and the output being the propeller tension can also be established.

[0119] The power unit model is a whole power mechanism with a brushless direct current motor, an electronic speed controller and a propeller as a group. As shown in Figure 7 , the throttle command σ is an input signal between 0 and 1, and the bottom battery voltage U b is not controlled.

[0120] The electronic speed controller receives the throttle command σ and the battery output voltage U b to generate an equivalent average voltage U m .

[0121] First, input a voltage signal, and the motor rotates to a steady state speed ω ss . This relationship is usually linear, that is,

[0122] ω ss =C b U m +ω b =C b U b σ+ω b =C R σ+ω b

[0123] wherein C R =C bU b , C b and ω b are constants.

[0124] Secondly, given a throttle command, the motor reaches a steady state speed ω ss in a certain time, which determines the dynamic response of the motor, denoted as T m . In general, the dynamic process of a brushless DC motor can be simplified as a first order low pass filter, whose transfer function is

[0125]

[0126] The complete power unit model is summarized as follows:

[0127]

[0128] Control efficiency model: the input is the propeller speed, and the output is the pull force and torque. When the propeller speed is known, the pull force and torque can be calculated through the control efficiency model. The inverse process of the control efficiency model is called the control allocation model, that is, when the desired pull force and torque are obtained through the controller design, the required propeller speed can be solved through the allocation model.

[0129] Multi-rotor usually uses constant pitch propellers, so the calculation formula of propeller pull force (unit: N) and torque (unit: N·m) is as follows:

[0130]

[0131] Where N (unit: RPM, revolutions per minute) is the propeller speed, D p (unit: m) is the propeller diameter, C T and C M are the dimensionless pull force coefficient and torque coefficient, respectively. ρ is the air density of the flight environment, which is a function of the flight altitude h (unit: m) and temperature T (unit: °C), and is expressed as follows:

[0132]

[0133] Where the standard atmospheric density ρ0=1.293 kg / m 3 (0℃, 273K), and the atmospheric pressure P a (unit: Pa) can be further expressed as

[0134]

[0135] The pull force of the propeller i of the rotor unmanned aerial vehicle is expressed as:

[0136]

[0137] where N is the propeller rotation speed i (unit: RPM, revolutions per minute) and ω i The conversion relationship between them is as follows:

[0138]

[0139] The counter-torque of the propeller i of the rotor unmanned aerial vehicle (static model) is:

[0140]

[0141] The multi-rotor unmanned aerial vehicle is driven by multiple propellers, and the rotation speed ω i (i = 1, 2,..., n) determines the total thrust f and torque τ of the multi-rotor.

[0142] As shown in Figure 8 , the total thrust of the propeller acting on the quadrotor is

[0143]

[0144] For the cross-shaped quadrotor, the torque generated by the propeller is

[0145]

[0146] The matrix form of the force and torque generated by the propeller of the unmanned aerial vehicle is:

[0147]

[0148] For the X-shaped quadrotor, the torque generated by the propeller is

[0149]

[0150] The matrix form of the force and torque generated by the propeller of the unmanned aerial vehicle is:

[0151]

[0152] Rigid body dynamics model: Dynamics involves both motion and force, and is related to the mass and moment of inertia of the object. Newton's second law, kinetic energy theorem and moment of inertia are often used to study the interaction between objects. The input of the multi-rotor dynamics model is the thrust and torque (pitching torque, rolling torque and yawing torque), and the output is the velocity and angular velocity. The rigid body kinematics model and dynamics model together constitute the commonly used multi-rotor flight control rigid body model.

[0153] The force on the rotor unmanned aerial vehicle is the propeller thrust F p , gravity G and air resistance F d , the direction of gravity is always perpendicular to the ground downward, and the size is:

[0154] G = mg

[0155] The air resistance is expressed as:

[0156]

[0157] The position dynamics model is:

[0158]

[0159] The attitude dynamics model is:

[0160]

[0161] where J e R 3×3 is the moment of inertia, is the moment of force on the UAV body axis, where the moments of the roll and pitch channels are generated by the propeller pulling force, and the yaw channel is generated by the propeller torque, is the gyroscopic moment.

[0162] The rigid body kinematics model: kinematics is independent of mass and force, and only studies variables such as position, velocity, attitude, and angular velocity. The input of the multi-rotor kinematics model is velocity and angular velocity, and the output is position and attitude.

[0163] The position state update equation is shown in the following formula, where p e = [x, y, z] T is the coordinate position of the UAV in the earth coordinate system, and v e is the velocity of the UAV in the earth coordinate system.

[0164]

[0165] The attitude of the UAV is represented by Euler angles, and the rate of change of the attitude angle and the rotation angular velocity of the body are:

[0166]

[0167] where Θ is the three attitude angles (Euler angles) of the UAV, and:

[0168]

[0169] That is:

[0170]

[0171] Under the condition of small perturbation, that is, under the premise that the change of each angle is small, the rate of change of the attitude angle and the rotation angular velocity of the body are approximately equal, so:

[0172]

[0173] The UAV on-board sensors include accelerometer, gyroscope, and GPS, etc. All the sensor models are represented as abstract interfaces, which can be replaced or added with new sensors according to the requirements:

[0174] (1) Gyroscope and accelerometer: Gyroscope and accelerometer are the core devices of inertial measurement unit (IMU). This module models them by adding white noise and bias drift over time to the real data of the UAV. The accelerometer needs to remove gravity from the real linear acceleration in the geodetic coordinate system and convert the linear acceleration to the body coordinate system before adding the bias drift and noise.

[0175] (2) Magnetometer: The tilt-dipole model of the earth's magnetic field is used to calculate the components of the real magnetic field on the ground in the body coordinate system by given geographic coordinates and add white noise specified in the data table.

[0176] (3) Global Positioning System: The Global Positioning System model in the simulation software has a waiting time (usually 200 ms), an update rate (usually 50 Hz), and horizontal and vertical position error estimation decay rates to simulate the gain over time. The decay rates in the horizontal and vertical directions are modeled based on a first-order low-pass filter with independent parameter settings.

[0177] Based on the mathematical model of a single-rotor UAV, the state calculation of a swarm rotor UAV is written as a state equation.

[0178]

[0179] Let X i be the linear state vector of the i-th UAV, where P i is the position in the geodetic coordinate system, V i is the velocity in the geodetic coordinate system, and ω ik is the rotation speed of the k-th propeller of the i-th UAV. The linear state vector equation of the i-th UAV is written as follows:

[0180]

[0181] where is the air resistance factor, is the propeller rotation speed factor.

[0182]

[0183] Let X = [X1, X2,..., X n ] T be the linear state vector of the UAV swarm, X' = [X'1, X'2,..., X'N] n ]T For the UAV cluster linear state vector input, the cluster linear state vector update equation is as follows:

[0184]

[0185] The attitude angle Θ of the i-th UAV i and the attitude angular velocity The update equation is shown as follows.

[0186]

[0187] Thus, the cluster attitude state update process of the rotor cluster in the embodiment is obtained.

[0188] 2. Fixed-wing UAV physical model:

[0189] Under the condition of neglecting elastic vibration and deformation, the motion of the UAV can be regarded as rigid body motion containing six degrees of freedom, including three rotations (roll, pitch and yaw) around the axis and three linear motions (forward, up and down and left and right) along the axis. In view of the actual flight environment of the current UAV, the following assumptions are made:

[0190] Assume that the aircraft is a rigid body (i.e. neglect the influence of the elasticity of the aircraft), and the mass is constant;

[0191] Assume that the earth is an inertial reference frame (i.e. the ground-fixed coordinate system is regarded as an inertial coordinate, and the influence of the earth's rotation and revolution is neglected);

[0192] The ox b z b plane in the body coordinate system is a reference plane, so the inertia products I xy and I yz are zero;

[0193] Neglect the curvature of the earth and regard the earth as a plane;

[0194] Assume that the gravitational acceleration does not change with the flight height;

[0195] Thus, a system of twelve nonlinear first-order differential equations can be established.

[0196] The application of Newton's second law in the inertial reference frame can establish the linear motion equation of the UAV under the action of external force and the angular motion equation under the action of external moment. Since it has been assumed that the mass of the UAV is constant and the ground-fixed coordinate system is an inertial system, the dynamic equation can be expressed as follows:

[0197] Linear motion equation of the UAV under the action of external force:

[0198]

[0199] Linear motion equation of UAV under the action of external resultant force:

[0200]

[0201] Where, external resultant force Total air force Engine thrust And gravity Total air force And engine thrust The resultant force is decomposed into (F x ,F y ,F z ) in the body coordinate system; The external resultant moment includes total air moment And engine thrust moment (M T ,N T ,L T ).

[0202] The equations of motion of the center of mass and the motion around the center of mass are projected on the body coordinate system, which is called "body-body" system motion equation.

[0203] State vector X T =[u,v,w,φ,θ,ψ,p,q,r,x g ,y g ,h].

[0204] Control input vector U T =[δ T ,δ e ,δ a ,δ r ], Where: δ T The throttle lever, for UAV, is the control quantity of the size of the propeller engine throttle; δ e , δ a , δ r The elevator deflection angle, aileron deflection angle and rudder deflection angle respectively.

[0205]

[0206] Motion equation set:

[0207]

[0208] Moment equation set:

[0209]

[0210] Where,

[0211]

[0212] Navigation equations:

[0213]

[0214] Or

[0215]

[0216] Solve a, β, V by the following formula:

[0217]

[0218] From the transfer relationship of the coordinate conversion matrix, γ, χ, μ can be obtained by the following equation:

[0219]

[0220] Similar to the rotor unmanned aerial vehicle cluster state solving equation, the linear state solving of the fixed-wing unmanned aerial vehicle is rewritten as shown below, wherein the linear state vector

[0221]

[0222] The matrix form of the linear state equation of the fixed-wing unmanned aerial vehicle cluster is:

[0223]

[0224] Due to the coupling of each channel in the attitude of the fixed-wing unmanned aerial vehicle, it cannot be expanded into a matrix form like the linear state vector, so the attitude change of different fixed-wing unmanned aerial vehicles in the cluster needs to be calculated separately.

[0225] Compared with the prior art, the unmanned platform state updating method for large-scale cluster simulation provided by the embodiment, by acquiring the types of each unmanned platform in the unmanned platform cluster and the control instructions of each unmanned platform, dividing the unmanned platform cluster into a rotor cluster and a fixed-wing cluster based on the types of each unmanned platform, and initializing the states of each cluster, updating the states of each cluster based on the cluster type, the states of each cluster at the current simulation time, and the control instructions of each unmanned platform, and repeating the step of taking the updated states of each cluster as the states of each cluster at the next simulation time until the simulation ends, the state of the unmanned platform at each simulation time is updated, which can support the state calculation of a super large-scale unmanned system cluster with more than one thousand nodes, consume less hardware resources, and a consumer-grade image processor can support the state calculation of a cluster with one thousand nodes.

[0226] Those skilled in the art can understand that all or part of the processes of the above-mentioned embodiment methods can be completed by instructing the relevant hardware by a computer program, and the program can be stored in a computer readable storage medium. The computer readable storage medium is a disk, an optical disk, a read-only memory, a random access memory, etc.

[0227] The above description is merely preferred specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A method for updating the state of an unmanned platform for large-scale cluster simulation, characterized in that: The following steps are involved: S1. Obtaining the type of each unmanned platform in the unmanned platform cluster and the control instructions of each unmanned platform; wherein the control instructions include a desired position or a desired speed, and the type of the unmanned platform includes a rotary-wing UAV and a fixed-wing UAV; S2. Dividing the unmanned platform cluster into a rotary-wing cluster and a fixed-wing cluster based on the type of each unmanned platform, and initializing the state of each cluster; the cluster state includes a cluster linear state and a cluster attitude state; S3. Based on the cluster type, the status of each cluster at the current simulation moment, and the control instructions of each unmanned platform, the status of each cluster is updated, and the updated cluster status is used as the cluster status of each cluster at the next simulation moment. Repeat step S3 to update the status until the simulation ends, completing the status update of the unmanned platform at each simulation moment.

2. The unmanned platform status update method for large-scale cluster simulation according to claim 1, characterized in that: The state of each cluster is updated based on the cluster type, the state of each cluster at the current simulation time, and the control instructions of each unmanned platform, including: If the cluster type is a rotor cluster, the cluster linear state and cluster attitude state are updated based on the rotor cluster state at the current simulation moment and the control instructions of each unmanned platform in the cluster; If the cluster type is a fixed-wing cluster, the cluster linear state and cluster attitude state of the cluster are updated based on the fixed-wing cluster state at the current simulation moment and the control instructions of each unmanned platform in the cluster.

3. The unmanned platform status update method for large-scale cluster simulation according to claim 2, characterized in that: If the cluster type is a rotor cluster, the cluster linear state is updated in the following way: Based on the rotor cluster state at the current simulation moment and the control instructions of each unmanned platform in the cluster, the linear state change rate of the rotor cluster is obtained; Integrate the rate of change of the rotor cluster linear state to obtain the updated cluster linear state of the rotor cluster; wherein, The elements in the cluster linear state of the rotor cluster are the linear state vectors of each rotor UAV, and the linear state vectors include the position, speed and the sum of the rotation speeds of each propeller of the rotor UAV.

4. The unmanned platform status update method for large-scale cluster simulation according to claim 3, characterized in that: If the cluster type is a rotor cluster, the cluster attitude status is updated in the following way: Based on the rotor cluster state at the current simulation moment, the first rate of change of the rotor cluster attitude state in the UAV coordinate system is obtained; Integrating the first rate of change of the rotor cluster attitude state in the UAV coordinate system to obtain a cluster attitude angular velocity, and then performing coordinate transformation on the cluster attitude angular velocity to obtain a second rate of change of the rotor cluster attitude state in the earth coordinate system; wherein the elements of the cluster attitude angular velocity are the attitude angular velocity of each rotor UAV; The second change rate of the rotor cluster attitude state in the geodetic coordinate system is integrated to obtain an updated cluster attitude state of the rotor cluster; wherein the elements in the cluster attitude state are the attitude angles of each rotor UAV.

5. The unmanned platform status update method for large-scale cluster simulation according to claim 2, characterized in that: If the cluster type is a fixed-wing cluster, the cluster linear state is updated in the following way: Based on the fixed-wing cluster state at the current simulation moment and the control instructions of each unmanned platform in the cluster, the fixed-wing cluster linear state change rate is obtained; The fixed-wing cluster state change rate is integrated to obtain the updated cluster linear state of the fixed-wing cluster; wherein, the elements in the cluster linear state are the linear state vectors of each fixed-wing UAV, and the linear state vectors include rolling state quantity, pitch state quantity, yaw state quantity, forward state quantity, up and down state quantity and left and right state quantity.

6. The unmanned platform status update method for large-scale cluster simulation according to claim 5, characterized in that: The elements in the cluster attitude state of the fixed-wing cluster are the attitude angles of each fixed-wing UAV, wherein the attitude angles include yaw angle, pitch angle and roll angle; If the cluster type is a fixed-wing cluster, the attitude of each fixed-wing UAV in the fixed-wing cluster is updated separately based on the attitude of each fixed-wing UAV at the current simulation moment, and then the updated cluster attitude state is obtained; among them, the attitude of each fixed-wing UAV is updated by integrating the attitude change rate.

7. The unmanned platform status update method for large-scale cluster simulation according to claim 3, characterized in that: The rotor UAV includes 4 propellers; the linear state change rate of the rotor cluster Expressed as: in, X=[X1,X2,...,X n ] T X′=[X1′,X2′,...,X n ′] T Where X represents the linear state of the rotor cluster at the current simulation moment, X′ represents the incremental input of the linear state of the rotor cluster at the current simulation moment, and M i represents the dynamic coefficient vector of the i-th rotor UAV in the rotor cluster, n represents the total number of rotor UAVs in the rotor cluster, represents the linear state change rate of the i-th rotor UAV in the rotor cluster, X i represents the linear state vector of the i-th rotor UAV in the rotor cluster at the current simulation moment, X i ′ represents the linear state vector increment input of the i-th rotor UAV in the rotor cluster at the current simulation moment, P i 、V i Respectively represent the position and speed of the i-th rotor UAV in the rotor cluster at the current simulation time, ω k,i represents the propeller speed of the kth propeller of the i-th rotor UAV in the rotor cluster at the current simulation moment, They represent the air resistance factor and propeller speed factor of the i-th rotor UAV in the rotor cluster, respectively, and T represents the transpose. The propeller speed of each propeller of the rotor UAV at the current simulation moment is obtained based on the control instruction.

8. The unmanned platform status update method for large-scale cluster simulation according to claim 7, characterized in that: The first rate of change of the rotor cluster attitude state Expressed as: in, Where, It represents the attitude angular acceleration of the i-th rotor UAV in the rotor cluster in the UAV coordinate system at the current simulation moment, They are The components of the x, y, and z axes in the body coordinate system; J i represents the moment of inertia of the i-th rotor UAV in the rotor cluster, represents the attitude angular velocity of the i-th rotor UAV in the rotor cluster in the body coordinate system at the current simulation moment, where They are The components of the horizontal, longitudinal and vertical axes in the body coordinate system; τ i represents the gyroscopic torque of the i-th rotor UAV in the rotor cluster, where τ i Defined as τ i,φ , τ i,θ , τ i,ψ are the gyroscopic moments of the i-th rotor UAV in the rotor cluster in the roll channel, pitch channel, and yaw channel respectively; It represents the propeller torque of the i-th rotor UAV in the rotor cluster and is defined as They are The components of the x, y, and z axes in the body coordinate system.

9. The unmanned platform status update method for large-scale cluster simulation according to claim 8, characterized in that: The second rate of change of the rotor cluster attitude state Expressed as: in, Where, represents the attitude angular velocity of the i-th rotor UAV in the rotor cluster in the geodetic coordinate system at the current simulation moment, where are the roll angles of the i-th rotor UAV in the rotor cluster at the current simulation moment. Pitch angle and yaw angle rate of change; W represents the attitude angular velocity of the i-th rotor UAV in the rotor cluster in the UAV coordinate system at the current simulation moment; i represents the transformation matrix of the i-th rotorcraft in the rotorcraft cluster.

10. The unmanned platform status update method for large-scale cluster simulation according to claim 5, characterized in that: Fixed-wing cluster linear state change rate Expressed as: in, Where, X f represents the linear state vector of the fixed-wing cluster at the current simulation moment, represents the rate of change of the linear state vector of the fixed-wing cluster at the current simulation moment, A represents the first dynamic coefficient vector of the jth fixed-wing UAV in the fixed-wing cluster at the current simulation moment, j represents the second dynamic coefficient vector of the jth fixed-wing UAV in the fixed-wing cluster at the current moment, n′ represents the total number of fixed-wing UAVs in the fixed-wing cluster, represents the linear state change rate of the jth fixed-wing UAV in the fixed-wing cluster, represents the linear state vector of the jth fixed-wing UAV in the fixed-wing cluster at the current simulation moment, They represent the rolling state, pitch state, yaw state, forward state, up-down state and left-right state of the j-th fixed-wing UAV in the fixed-wing cluster at the current simulation moment, They respectively represent the rates of change of the roll state, pitch state, yaw state, forward state, up and down control value, and left and right state value of the j-th fixed-wing UAV in the fixed-wing cluster, and T represents transpose.

Images