Unmanned aerial vehicle deployment method in mountain dense forest emergency rescue scene and related equipment

Abstract

The invention provides an unmanned aerial vehicle deployment method, device, equipment and medium in a mountain dense forest emergency rescue scene, and the method comprises the steps: building a non-line-of-sight avoidance virtual force model, and solving the total resultant force of virtual forces borne by an unmanned aerial vehicle according to the non-line-of-sight avoidance virtual force model; constructing a particle swarm optimization model by taking each unmanned aerial vehicle deployment scheme as a particle, and optimizing the speed and position of the particle by combining the total resultant force of the virtual force of each unmanned aerial vehicle so as to dynamically adjust the deployment position of the unmanned aerial vehicle group; and iteratively solving the particle swarm optimization model until a preset positioning precision requirement is met or a maximum iteration number is reached, and outputting an optimal deployment position. According to the invention, the number of non-line-of-sight links can be effectively reduced by dynamically adjusting the deployment position of the unmanned aerial vehicle group; and a particle swarm optimization model is combined to solve the problems that an artificial potential energy field method has high dependence on the initial deployment position of the unmanned aerial vehicle and the initial position needs to be adjusted for multiple times, and the positioning precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of UAV communication, and in particular to a method, device, equipment and storage medium for deploying UAVs in an emergency rescue scenario in mountainous dense forests. Background Art

[0002] In emergency scenarios in complex terrains such as mountainous dense forests, the precise positioning of ground search and rescue personnel is a difficult problem. On the one hand, the success rate and accuracy of GPS positioning will be greatly reduced in occluded areas such as forest coverage, terrain undulations, and valley bottoms, and even positioning rejection may occur; on the other hand, after natural disasters such as geology and floods, ground communication base stations are damaged, and ground search and rescue personnel cannot obtain communication and positioning services, and position information cannot be transmitted to the ground search and rescue command center in real time, increasing the difficulty of emergency rescue command. UAVs, with the characteristics of high mobility, flexibility, and all-weather operation, can carry communication and positioning network equipment to provide positioning services for ground search and rescue personnel. In related technologies, the particle swarm optimization (PSO) algorithm is used to optimize the deployment of UAVs to provide positioning services for ground search and rescue personnel. However, due to the complex and diverse terrain of the disaster area, in the case of limited UAVs, it is easy to form non-line-of-sight links between UAVs and ground search and rescue personnel, and the traditional PSO algorithm does not reduce the number of non-line-of-sight links. Therefore, the positioning accuracy is poor. The artificial potential field (APF) method can quickly calculate according to the terrain environment, effectively avoid collisions, and guide UAVs to move towards the target area. However, in complex scenarios such as mountainous dense forests, the APF method has a high dependence on the initial position of the UAV and is prone to falling into local minima. Summary of the Invention

[0003] The present invention provides a method, device, equipment and storage medium for deploying UAVs in an emergency rescue scenario in mountainous dense forests, so as to solve the defects that the traditional UAV deployment method does not reduce the number of non-line-of-sight links in complex scenarios such as mountainous dense forests, resulting in low positioning accuracy, and has a high dependence on the initial position of the UAV and requires multiple adjustments of the initial position.

[0004] The present invention provides a method for deploying UAVs in an emergency rescue scenario in mountainous dense forests, including: Constructing a virtual force model for non-line-of-sight avoidance, and solving the total resultant force of the virtual forces received by the UAV according to the virtual force model for non-line-of-sight avoidance; Taking each UAV deployment plan as a particle to construct a particle swarm optimization model, and optimizing the speed and position of the particles by combining the total resultant force of the virtual forces of each UAV to dynamically adjust the deployment position of the UAV swarm; Iteratively solve the particle swarm optimization model until the preset positioning accuracy requirement is met or the maximum number of iterations is reached, and output the optimal deployment position.

[0005] According to the method for deploying unmanned aerial vehicles in the mountainous dense forest emergency rescue scenario provided by the present invention, the constructing a virtual force model for non-line-of-sight avoidance and solving the total resultant force of the virtual forces received by the unmanned aerial vehicles according to the virtual force model for non-line-of-sight avoidance includes: Obtain the virtual forces received by each unmanned aerial vehicle, where the virtual forces include anti-collision virtual repulsive forces, positioning virtual attractive forces, and non-line-of-sight avoidance virtual forces; Based on the anti-collision virtual repulsive force, the positioning virtual attractive force, and the non-line-of-sight avoidance virtual force, construct a virtual force model for non-line-of-sight avoidance, calculate the resultant force of the anti-collision virtual repulsive force, the positioning virtual attractive force, and the non-line-of-sight avoidance virtual force through the virtual force model for non-line-of-sight avoidance, and take the resultant force as the total resultant force of the virtual forces.

[0006] According to the method for deploying unmanned aerial vehicles in the mountainous dense forest emergency rescue scenario provided by the present invention, the obtaining the virtual forces received by each unmanned aerial vehicle includes:

[0007] The anti-collision virtual repulsive force between the i-th unmanned aerial vehicle and an obstacle or another unmanned aerial vehicle k is expressed as:

[0008]

[0009] where K rep represents the virtual repulsive force factor, represents the effective threshold of the virtual repulsive force, β is a constant representing the reduction factor, p i represents the coordinate of the unmanned aerial vehicle, and o k represents the coordinate of the center of the obstacle or the unmanned aerial vehicle;

[0010] The user u j exerts a virtual attractive force on the unmanned aerial vehicle subgroup g j =[p i , p j , p k is expressed as: where K att represents the coefficient factor of the positioning virtual attractive force, and P is the constructed artificial potential field; Define to represent the unit normal vector of the straight line connecting the i-th unmanned aerial vehicle and the j-th ground search and rescue personnel, and the non-line-of-sight avoidance virtual force exerted by the ground user on the unmanned aerial vehicle is expressed as: where K eva represents the non-line-of-sight avoidance virtual force coefficient, p iRepresents the coordinate point of the i-th drone, u j Represents the coordinate point of the j-th ground search and rescue personnel. According to the drone deployment method in the mountainous dense forest emergency rescue scenario provided by the present invention, taking each drone deployment plan as a particle to construct a particle swarm optimization model, and combining the total resultant force of the virtual forces of each drone to optimize the velocity and position of each particle to dynamically adjust the deployment position of the drone swarm, including: Generate an initial particle swarm, set the initial position set of the drone swarm, the initial connection matrix between the drones and the ground search and rescue personnel, and the initial average positioning accuracy; For each particle, calculate the average positioning accuracy of the current particle; If the average positioning accuracy of the current particle is better than the historical optimal value, update the global optimal solution, and record the optimal accuracy and its corresponding drone position and connection matrix; Combine the inertia weight, learning factor, random number, and the total resultant force of the virtual forces received by the drones to jointly update the velocity of the particles; Adjust the position of the drones according to the updated velocity, ensuring that the maximum flight speed is not exceeded; After completing the round iteration and particle iteration, return the global optimal deployment position and its corresponding average positioning accuracy. According to the drone deployment method in the mountainous dense forest emergency rescue scenario provided by the present invention, the method of jointly updating the velocity of the particles by combining the inertia weight, learning factor, random number, and the total resultant force of the virtual forces received by the drones includes: Obtain the algorithm input parameters, and the algorithm input parameters include the number of drones N, the number of ground users M, the transmission power of the drone signal, the environmental noise power, the positioning accuracy threshold ∈, and the maximum moving speed v of the drones max ; Obtain the number of particles in the particle swarm according to the number of drones N and the number of ground search and rescue personnel M; Input the transmission power of the drone signal and the environmental noise power into the channel model, and obtain the total resultant force of the virtual forces on each particle according to the channel transmission loss; According to the number of particles, the positioning accuracy threshold ∈, the maximum moving speed v of the drones max And the total resultant force of the virtual forces on the particles to update the velocity of the particles, and the velocity update formula is p i = p i + V i where ω is the inertia weight, c1, c2 are the learning factors, r1, r2 are random numbers in [0, 1], V i = [vp,x , v p,y , v p,z is the velocity vector of the i-th unmanned aerial vehicle (UAV). represents the total resultant force of the virtual forces on the particle, g best is the historical best position, p i is the position of the particle. According to the UAV deployment method provided by the present invention in the mountainous and dense forest emergency rescue scenario, the positioning accuracy is calculated through the time difference of arrival model between the ground user and the UAV, and the positioning result is evaluated based on the geometric dilution of precision factor. According to the UAV deployment method provided by the present invention in the mountainous and dense forest emergency rescue scenario, the input parameters of the acquisition algorithm include: The input parameters of the acquisition algorithm are obtained through the channel model, and the channel model includes line-of-sight link loss and non-line-of-sight link loss: where f represents the carrier frequency, d0 represents the reference distance, c represents the speed of light; n LoS represents the path loss factor in the line-of-sight link, n NLoS represents the path loss factor in the non-line-of-sight link, L F (d0) represents the path loss of free space when the reference distance is d0 between the i-th UAV and the j-th ground user, L Slant represents the additional loss in the forest area where the connection line between the UAV and the ground user is inclined, X LoS represents the forest area shadow fading in the line-of-sight link, X NLoS represents the forest area shadow fading in the non-line-of-sight link, d i,j is the distance between the i-th UAV and the j-th ground search and rescue personnel.

[0011] The present invention also provides a UAV deployment device in the mountainous and dense forest emergency rescue scenario, including: The first construction module is used to construct a virtual force model for non-line-of-sight avoidance, and solve the total resultant force of the virtual forces received by the UAV according to the virtual force model for non-line-of-sight avoidance; The second construction module is used to construct a particle swarm optimization model with each UAV deployment plan as a particle, and optimize the speed and position of the particle by combining the total resultant force of the virtual forces of each UAV to dynamically adjust the deployment position of the UAV swarm; The output module is used to iteratively solve the particle swarm optimization model until the preset positioning accuracy requirement is met or the maximum number of iterations is reached, and output the optimal deployment position.

[0012] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the method for deploying drones in the mountainous dense forest emergency rescue scenario as described in any one of the above is implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for deploying drones in the mountainous dense forest emergency rescue scenario as described in any one of the above is implemented.

[0014] The method, device, equipment, and storage medium for deploying drones in the mountainous dense forest emergency rescue scenario provided by the present invention construct a virtual force model for non-line-of-sight avoidance, and solve the total resultant virtual force received by the drones according to the virtual force model for non-line-of-sight avoidance; use each drone deployment plan as a particle to construct a particle swarm optimization model, and optimize the speed and position of the particles in combination with the total resultant virtual force of each drone to dynamically adjust the deployment positions of the drone swarm; iteratively solve the particle swarm optimization model until the preset positioning accuracy requirement is met or the maximum number of iterations is reached, and output the optimal deployment position. By dynamically adjusting the deployment positions of the drone swarm, the present invention can effectively reduce the number of non-line-of-sight links, and solves the problem that the artificial potential field method has a high dependence on the initial deployment position of the drones and requires multiple adjustments of the initial position by combining the particle swarm optimization model, thereby improving the positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 is a schematic flow chart of the method for deploying drones in the mountainous dense forest emergency rescue scenario provided by the embodiment of the present invention; Figure 2 is a deployment scenario diagram of the drone positioning network in the complex terrain emergency scenario provided by the embodiment of the present invention; Figure 3 is a schematic diagram of the comparison result of the convergence curve and standard deviation of the algorithm positioning result provided by the embodiment of the present invention; Figure 4 is a comparison diagram of the positioning results of the number of drones provided by the embodiment of the present invention; Figure 5 is provided by the embodiment of the present invention Comparison diagram of positioning results and standard deviations under different values; Figure 6It is a schematic diagram of the functional structure of a drone deployment device in a mountainous dense forest emergency rescue scenario provided by an embodiment of the present invention; Figure 7 It is a schematic diagram of the functional structure of an electronic device provided by an embodiment of the present invention. Detailed implementation manners

[0017] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0018] Figure 1 It is a flowchart of a drone deployment method in a mountainous dense forest emergency rescue scenario provided by an embodiment of the present invention. As Figure 1 shown, the drone deployment method in a mountainous dense forest emergency rescue scenario provided by an embodiment of the present invention includes: Step 101: Construct a virtual force model for non-line-of-sight avoidance, and solve the total resultant force of the virtual forces received by the drone according to the virtual force model for non-line-of-sight avoidance; Step 102: Use each drone deployment plan as a particle to construct a particle swarm optimization model, and optimize the speed and position of the particle by combining the total resultant force of the virtual forces of each drone to dynamically adjust the deployment position of the drone swarm; Step 103: Iteratively solve the particle swarm optimization model until the preset positioning accuracy requirement is met or the maximum number of iterations is reached, and output the optimal deployment position.

[0019] Traditionally, the particle swarm algorithm is used to optimize the drone deployment to provide positioning services for ground search and rescue personnel. However, due to the complex and diverse terrain of the disaster area, in the case of a limited number of drones, it is easy to form a non-line-of-sight link between the drones and the ground search and rescue personnel, and the traditional PSO algorithm does not reduce the number of non-line-of-sight links. Therefore, the positioning accuracy is poor. The artificial potential field method can be quickly calculated according to the terrain environment, effectively avoid collisions, and guide the drone to move towards the target area. However, in complex scenarios such as mountainous dense forests, the APF method has a high dependence on the initial position of the drone and is prone to falling into the situation of local minimum.

[0020] The UAV deployment method in the mountain dense forest emergency rescue scenario provided by the embodiments of the present invention constructs a virtual force model for non-line-of-sight avoidance, and solves the total resultant force of the virtual forces received by the UAV according to the virtual force model for non-line-of-sight avoidance; uses each UAV deployment plan as a particle to construct a particle swarm optimization model, and combines the total resultant force of the virtual forces of each UAV to optimize the speed and position of the particles to dynamically adjust the deployment positions of the UAV swarm; iteratively solves the particle swarm optimization model until the preset positioning accuracy requirement is met or the maximum number of iterations is reached, and outputs the optimal deployment position. The present invention can effectively reduce the number of non-line-of-sight links by dynamically adjusting the deployment positions of the UAV swarm, and combines the particle swarm optimization model to solve the problem that the artificial potential field method is highly dependent on the initial deployment position of the UAV and requires multiple adjustments of the initial position, improving the positioning accuracy.

[0021] In the embodiments of the present invention, the positions of the UAVs carrying the positioning network devices are deployed to provide reliable positioning services for the ground search and rescue personnel, so that the positions of the ground search and rescue personnel can transmit the positioning information back to the rescue command center in time in the case of positioning denial in the disaster area. The system architecture of the embodiments of the present invention is as Figure 1 shown. Assuming to provide positioning services for the ground search and rescue personnel in the mountain forest emergency scenario, deploy UAVs to form an aerial positioning network for ground search and rescue personnel to provide positioning services, and the UAVs establish connections with users using the same frequency band. Define and to represent the sets of UAVs and ground search and rescue personnel respectively. The area where the UAVs can be deployed can be expressed as . The position of the UAV can be described by , where ; the position of the ground search and rescue personnel can be described by . Then, the distance between the th UAV and the th ground search and rescue personnel can be calculated.

[0022] In the disaster area with complex terrain and numerous obstacles, the ranging error caused by non-line-of-sight links is one of the important reasons for the difficulty in ensuring positioning accuracy. The conventional idea for solving non-line-of-sight links is to identify and fit the ranging error caused by non-line-of-sight links, and add the ranging error caused by non-line-of-sight links to the positioning calculation during positioning solution. This method is greatly affected by environmental factors and is limited in practical applications. In the UAV deployment position solving method, the APF algorithm can be quickly deployed to this mountain dense forest emergency rescue, but it is greatly affected by the initial deployment position and has the disadvantages of being easily trapped in local minima and difficult to converge; the PSO algorithm has the advantage of easy convergence, but it does not have the ability to explicitly avoid non-line-of-sight links.

[0023] Based on any of the above embodiments, for the scenario of emergency rescue in mountainous dense forests, the embodiments of the present invention design three artificial potential fields: an anti-collision repulsive potential field, a positioning gravitational potential field, and a non-line-of-sight link avoidance potential field. The three virtual forces generated guide the unmanned aerial vehicle (UAV) to move in a reasonable direction. In the embodiments of the present invention, constructing the virtual force model for non-line-of-sight avoidance, and solving the total resultant force of the virtual forces received by the UAV according to the virtual force model for non-line-of-sight avoidance includes: Step 201, obtain the virtual forces received by each UAV, where the virtual forces include an anti-collision virtual repulsive force, a positioning virtual gravitational force, and a non-line-of-sight avoidance virtual force; Step 202, construct a virtual force model for non-line-of-sight avoidance based on the anti-collision virtual repulsive force, the positioning virtual gravitational force, and the non-line-of-sight avoidance virtual force. Calculate the resultant force of the anti-collision virtual repulsive force, the positioning virtual gravitational force, and the non-line-of-sight avoidance virtual force through the virtual force model for non-line-of-sight avoidance, and use the resultant force as the total resultant force of the virtual forces.

[0024] In the embodiments of the present invention, the APF with non-line-of-sight avoidance ability guides the UAV to avoid collisions with obstacles and improve the positioning accuracy of ground search and rescue personnel. According to actual requirements, the resultant force of the virtual forces received by the UAV can be expressed as the vector sum of three virtual forces:

[0025]

[0026] where, represents the virtual repulsive force generated between the i-th UAV and the obstacle k. The virtual repulsive force is designed to prevent collisions with obstacles or other UAVs. The magnitude of this force is determined by the distance between the UAV and the obstacle or other UAVs. As the distance to the obstacle decreases, the anti-collision virtual repulsive force increases; when the distance exceeds the threshold , the anti-collision virtual repulsive force tends to zero.

[0027] In the embodiments of the present invention, obtaining the virtual forces received by each UAV includes:

[0028] The anti-collision virtual repulsive force between the i-th UAV and the obstacle or other UAV k is expressed as:

[0029]

[0030] where, K rep represents the virtual repulsive force factor, represents the effective threshold of the virtual repulsive force, β is a constant representing the reduction factor, p i represents the UAV coordinate, and o k represents the coordinate of the center of the obstacle or UAV;

[0031] User u jFor the drone subgroup g j = [p i , p j , p k The virtual gravitational force exerted is expressed as:

[0032]

[0033] where K att represents the coefficient factor for positioning the virtual gravitational force, and P is the constructed artificial potential field;

[0034] Define to represent the unit normal vector of the straight line connecting the i-th drone and the j-th ground search and rescue personnel. The non-line-of-sight avoidance virtual force exerted on the drone by the ground user is expressed as:

[0035]

[0036] where, K eva represents the non-line-of-sight avoidance virtual force coefficient, p i represents the coordinate point of the i-th drone, and u j represents the coordinate point of the j-th ground search and rescue personnel.

[0037] In the embodiment of the present invention, the APF is combined with the PSO algorithm, and the resultant force of three virtual forces is used to replace the local minimum part, so that the drone is jointly guided by the current global optimal position and the resultant force of the virtual forces, improving the positioning accuracy of the ground search and rescue personnel. Specifically, the anti-collision virtual repulsive force generated by the anti-collision repulsive force potential field can prevent the drone from colliding with obstacles or other drones, effectively improving the safety of drone deployment. The positioning gravitational potential field generates a positioning virtual gravitational force to guide the drone to move towards a place with higher positioning accuracy, ensuring the system performance of the positioning network. The non-line-of-sight avoidance virtual force generated by the non-line-of-sight link avoidance potential field will reduce the number of non-line-of-sight links between the drone and the ground user, skillfully solving the problem of poor positioning accuracy of the ground search and rescue personnel affected by non-line-of-sight links, and effectively ensuring the system performance of the positioning network. The traditional PSO algorithm is prone to falling into the local minimum problem. A reasonably designed APF can guide the particles in the PSO algorithm to continuously iterate and optimize, effectively avoiding the local minimum problem, and also solving the problem of difficult convergence and easy falling into the local minimum of the APF method.

[0038] Based on any of the above embodiments, constructing a particle swarm optimization model with each drone deployment plan as a particle, and optimizing the velocity and position of each particle by combining the total resultant force of the virtual forces of each drone to dynamically adjust the deployment position of the drone swarm, including:

[0039] Step 301: Generate an initial particle swarm, set the initial position set of the UAV swarm, the initial connection matrix between the UAVs and the ground search and rescue personnel, and the initial average positioning accuracy.

[0040] Step 302: For each particle, calculate the average positioning accuracy of the current particle.

[0041] Step 303: If the average positioning accuracy of the current particle is better than the historical optimal value, update the global optimal solution, and record the optimal accuracy and its corresponding UAV positions and connection matrix.

[0042] Step 304: Combine the inertia weight, learning factor, random number, and the total resultant force of the virtual forces received by the UAVs to jointly update the velocity of the particles.

[0043] Step 305: Adjust the positions of the UAVs according to the updated velocity, ensuring that the maximum flight speed is not exceeded.

[0044] Step 306: After completing the round iteration and particle iteration, return the global optimal deployment position and its corresponding average positioning accuracy.

[0045] In the embodiment of the present invention, the combination of the inertia weight, learning factor, random number, and the total resultant force of the virtual forces received by the UAVs to jointly update the velocity of the particles includes:

[0046] Step 3041: Obtain the algorithm input parameters, where the algorithm input parameters include the number of UAVs N, the number of ground search and rescue personnel M, the transmission power of the UAV signal, the environmental noise power, the positioning accuracy threshold ∈, and the maximum moving speed v of the UAVs max ;

[0047] Step 3042: Obtain the number of particles in the particle swarm according to the number of UAVs N and the number of ground search and rescue personnel M.

[0048] Step 3043: Input the transmission power of the UAV signal and the environmental noise power into the channel model, and obtain the total resultant force of the virtual forces on each particle according to the channel transmission loss.

[0049] Step 3044: Update the velocity of the particles according to the number of particles, the positioning accuracy threshold ∈, the maximum moving speed v of the UAVs max and the total resultant force of the virtual forces on the particles. The velocity update formula is

[0050]

[0051] p i = p i + V i

[0052] where ω is the inertia weight, c1 and c2 are learning factors, and r1 and r2 are random numbers in the range of [0, 1].

[0053] V i = [v p,x , v p,y , v p,z is the velocity vector of the i-th UAV. represents the total resultant force of the virtual forces on the particle, and g best is the historical best position, and p i is the position of the particle.

[0054] In the embodiments of the present invention, in order to provide a positioning service with as high positioning accuracy as possible for ground search and rescue personnel, it is desirable that the communication link between the UAV and the ground search and rescue personnel is a line-of-sight link, and the ranging error measured at this time is relatively small. Assume that a 0-1 variable α i,j ∈ {0, 1}, i =

[0055] 1, 2,..., N, j = 1, 2,..., M represents the connection relationship between the i-th UAV and the j-th ground search and rescue personnel, where 0 represents non-line-of-sight connection or non-connection, and 1 represents line-of-sight connection. Therefore, for each ground search and rescue personnel, a column vector α j = [α 1,j , α 2,j ,..., α N,j can be used to describe the connection status between the UAV swarm and the ground user j. Furthermore, α = [α1,..., α T can be used to represent the connection status between the UAV swarm and the ground search and rescue personnel. In order to improve the positioning accuracy as much as possible, it is necessary to make the number of line-of-sight links as large as possible. When the number of line-of-sight links is greater than or equal to 3, the provided positioning service can be considered to be high enough. M

[0056] In a terrain-complex scenario, the obstacles can be simply modeled in the form of cylinders, and Q = {q = [x q,k , y q,k , z q,k , i = 1, 2,..., K} can be used to describe the set of obstacle positions, and the column vector r = [r1, r2,..., r k is used to describe the radius of the obstacles. T

[0057]

[0058] The design optimization objective is to minimize GDOP by optimizing the deployment positions of the UAV swarm and the number of non-line-of-sight links avg , then this optimization problem can be expressed as:

[0059] ​​Among them, the optimized objective function is the minimum average value of GDOP. The constraint C1 indicates that the number of line-of-sight links of the ground user j is at least 3. The constraints C2 and C3 indicate that the deployment position of the UAV needs to consider the factor of preventing collisions, that is, a safe distance needs to be maintained between UAVs, and a safe distance needs to be ensured between UAVs and obstacles. Among them, and respectively represent the safety distance threshold between UAVs and the safety distance threshold between UAVs and obstacles. The constraint C4 represents the position boundary where the UAV can be deployed to prevent the UAV from being deployed outside the disaster area.

[0060] In the above optimization problem, since the optimized variable p, the deployment position of the UAV swarm, is a continuous variable and the number of non-line-of-sight links α j is a discrete variable, this optimization problem is a constrained mixed-integer non-linear optimization problem and is not suitable for explicit solution by taking derivatives. A penalty function is introduced to transform the above optimization problem into an unconstrained optimization problem, and then the virtual force is used to optimize the number of non-line-of-sight links. The transformation idea is to introduce a penalty function for each constraint. When the constraint is violated, the penalty function will increase the value of the objective function, thereby forcing the optimization algorithm to find a solution that satisfies the constraint. Then the above optimization problem can be transformed into an unconstrained problem with penalty terms:

[0061] Among them, each P t represents the penalty function corresponding to the constraint. For the constraint C1: α i,j is a discrete 0, 1 variable. In order to make the penalty function differentiable at , the square form can be used as the penalty function, and the corresponding penalty function can be expressed as: Among them, λ1 represents the penalty coefficient corresponding to the constraint C1, and the value range is λ1≥0. For the constraint C2: and the constraint C3: The corresponding penalty term can be expressed as: Among them, λ2 and λ3 respectively represent the corresponding penalty coefficients. For the constraint C4: 0≤x p,i ≤L x , 0≤y p,i ≤L y , h min ≤z p,i ≤h maxThe penalty term can be represented by a linear penalty function, i.e.: where λ4 represents the linear penalty function. The APF algorithm is highly dependent on the initial positions of UAVs in the UAV deployment scenario. If there are many obstacles at the initial positions, it is very easy to cause the balance of forces due to numerous non-line-of-sight avoidance virtual forces, making it impossible to obtain the optimal solution and falling into the local minimum trap. By setting numerous different initial deployment schemes and recording the global optimal solutions of each iteration, the PSO algorithm can effectively prevent the APF algorithm from falling into the local minimum trap and effectively improve the positioning accuracy of the UAV positioning network. Therefore, when the APF algorithm is combined with the PSO algorithm, the APF guides the UAV to move towards areas with high positioning accuracy. The current global optimal solution is recorded in each iteration, and this global optimal solution also guides the UAV to move in the direction of this optimal solution. When the positioning accuracy reaches a certain threshold or the preset number of rounds is reached, the algorithm iteration ends. The UAV deployment method provided in the embodiments of the present invention for the mountain dense forest emergency rescue scenario is shown in Table 1. Table 1 UAV deployment algorithm steps for the mountain dense forest emergency rescue scenario This method first generates an initial particle swarm and sets the initial position set of the UAV swarm The initial connection matrix α between the UAV and the ground search and rescue personnel (0) and the initial average positioning accuracy In the optimization of the algorithm iteration rounds, the following operations are performed on each particle (a total of R particles, and each particle represents a deployment scheme): conditional judgment and speed and position update operations. Among them, the conditional judgment operation is that if the average positioning accuracy of the current particle is better than the historical optimal value then update the global optimal solution and record the optimal accuracy and its corresponding UAV position and the connection matrix α * . The speed update operation is to jointly update the particle speed V by combining the inertia weight ω, learning factors c1, c2, random numbers r1, r2, and the resultant force matrix received by the UAV. The position update operation is to update the UAV position according to the speed V to ensure that it does not exceed the maximum flight speed v max . After completing all rounds of iteration and particle iteration, return the global optimal deployment position p * and the corresponding average positioning accuracy Based on any of the above embodiments, the positioning accuracy calculates the distance through the time difference of arrival model between the ground user and the UAV, and evaluates the positioning result based on the geometric dilution of precision factor. The embodiment of the present invention adopts a TDOA (Time Difference of Arrival) positioning and solving algorithm with the UAV as the base station, and selects GDOP (Geometric Dilution of Precision) as the metric index for evaluating the positioning result. Define the GDOP value of the user located at u j =[x u,j ,y u,j ,z u,j as GDOP j , and use the average GDOP value of all users to evaluate the overall performance of the UAV positioning deployment. Define to represent the trace of the matrix. Based on the above description, GDOP can be expressed as where H=(F T F) -1 F T , and P dz represents the ranging error covariance matrix: where η,σ i represent the correlation coefficient and standard deviation of the Gaussian white noise of the communication link respectively, i = 1, 2,..., N. The TDOA model between the ground search and rescue personnel and the UAV base station p k (k = 2,..., N) relative to the UAV main base station p1 can be expressed as where represents the distance between the UAV base station p k and the j-th ground search and rescue personnel, represents the ranging error caused by noise during the TDOA measurement process, and this error follows a Gaussian distribution with a mean of 0 and a standard deviation of σ i =σ. Based on any of the above embodiments, the input parameters of the acquisition algorithm include: Obtain the input parameters of the algorithm through the channel model, and the channel model includes line-of-sight link loss and non-line-of-sight link loss: where f represents the carrier frequency, d0 represents the reference distance, c represents the speed of light; n LoS represents the path loss factor under the line-of-sight link, nNLoS Denotes the path loss factor in the non-line-of-sight link, L F L(d0) represents the path loss in free space when the reference distance is d0 between the ith UAV and the jth ground user Slant Denotes the additional loss in the forest area where the connection line between the UAV and the ground user is inclined, X LoS Denotes the forest area shadow fading in the line-of-sight link, X NLoS Denotes the forest area shadow fading in the non-line-of-sight link, d i,j Is the distance between the ith UAV and the jth ground search and rescue personnel In the embodiment of the present invention, for mountain forest areas with complex terrain, during the deployment process of the UAV, it is very easy for the connection line between the UAV and the ground user to have a certain angle with the horizontal plane. Therefore, it can be processed according to the channel model of the forest area inclination. The total fading between the UAV and the ground user can be defined as the sum of the log-distance shadow fading and the excess loss in the forest area. And due to mountain occlusion, calculations need to be carried out from two perspectives of the line-of-sight link and the non-line-of-sight link Where L F L(d0) represents the path loss in free space when the reference distance is d0 between the ith UAV and the jth ground search and rescue personnel, and can be expressed as Where f represents the carrier frequency, d0 represents the reference distance, c represents the speed of light; n LoS Denotes the path loss factor in the line-of-sight link, n NLoS Denotes the path loss factor in the non-line-of-sight link, and both are constants, satisfying n LoS < n NLoS ; X LoS Denotes the forest area shadow fading in the line-of-sight link, which is a Gaussian random variable with a mean of 0 and a standard deviation of σ LoS , X NLoS Denotes the forest area shadow fading in the non-line-of-sight link, which is a Gaussian random variable with a mean of 0 and a standard deviation of σ NLoS . L Slant Denotes the additional loss in the forest area where the connection line between the UAV and the ground search and rescue personnel is inclined, and can be expressed as Where A, C, E, G, H are environmental parameters, A is a multiplier factor, C is an exponential factor of the carrier frequency, E is an exponential factor of the transmission distance, θ is the elevation angle between the UAV and the user, G is a correction factor of the elevation angle θ, and H is an exponential factor of the elevation angle θ. The above parameters are affected by factors such as the type and density of forest vegetation Therefore, the total channel fading between the UAV i and the ground search and rescue personnel j can be expressed as: Furthermore, the signal-to-noise ratio (SNR) between the UAV i and the ground search and rescue personnel j can be expressed as: where P j represents the power allocated to the ground user j, and the power allocation matrix for all users can be expressed as P = [P1, P2, …, P M . φ j represents the noise power of the ground search and rescue personnel j. Based on any of the above embodiments, the embodiments of the present invention prove the advantages of the proposed artificial potential field - particle swarm algorithm in the UAV positioning and deployment problem through experimental simulation results. Assume that the ground search and rescue personnel are distributed in an area of 2.5 km × 2.5 km, set the initial flight altitude of the UAV to h = 300 m, and each UAV has the same maximum transmission power, i.e., P1 = P2 = … = P N = 30 dBM, the minimum SNR threshold required for the ground search and rescue personnel to be served is The carrier power is f = 1.4 GHz, and the noise power is φ = -110 dBm. The empirical parameters of the forest area channel are A = 0.25, B = 0.39, C = 0.25, E = 0, G = 0.05. The path loss exponent is α = 3.5, and the reference distance is d0 = 1 m. The positioning virtual gravitational coefficient factor K att = 200, the anti-collision virtual repulsive coefficient factor K rep = 1000, the non-line-of-sight avoidance virtual force coefficient factor K evd = 1000, the virtual force coefficient K for maintaining the UAV subgroup configuration main = 200, the convergence threshold is ε = 5 dB, the anti-collision virtual repulsive distance threshold The virtual force conversion threshold for maintaining the UAV subgroup configuration As Figure 3 , Figure 4 shown, the comparison results of the method provided by the embodiments of the present invention with other benchmark algorithms are presented. The benchmark algorithms for comparison include the APF algorithm, the PSO algorithm, and the PSO - APF algorithm. Among them, the APF algorithm randomly initializes a group of UAVs and then guides the UAV group to complete the deployment through the constructed virtual potential field; the PSO algorithm is the standard PSO algorithm. In the embodiments of the present invention, the PSO algorithm is first used to obtain a better deployment position of the UAV swarm, and then the APF algorithm is used for optimized deployment. Figure 3Shows the comparison of the convergence curves of the positioning results of various algorithms and the comparison of the standard deviation of the convergence results when the number of drones is 7. Among them, Figure 3 (a) is a comparison chart of the convergence curves of the positioning results, Figure 3 (b) is a comparison chart of the standard deviation of the positioning results. The positioning result of the embodiment of the present invention is better than other algorithms. And it converges faster than the VF-PSO algorithm. As Figure 3 (b) shows, compared with other algorithms, the standard deviation of the method provided by the embodiment of the present invention is the smallest, indicating that it is more stable and reliable than other algorithms. Figure 4 Shows the comparison of the positioning results under different numbers of drones. The simulation results show that the positioning accuracy of the APF method and the PSO-APF method has no obvious relationship with the increase in the number of drones. However, the average positioning accuracy of the NAAPF-PSO algorithm provided by the embodiment of the present invention increases with the increase in the number of drones, and the gap between the average positioning accuracy of the algorithm and the PSO algorithm is the largest when the number of drones is 6, indicating that the NAAPF-PSO algorithm provided by the embodiment of the present invention has greater advantages when the number of drones is small, that is, the embodiment of the present invention is not easily trapped in a local optimal solution when the number of drones is small, but other algorithms are easily trapped in a local optimal solution. Figure 5 Analyzes the influence of the penalty term coefficient on the algorithm performance within different numerical ranges. Figure 5 (a) shows the iterative results of the average positioning accuracy under different λ values, that is, GDOP avg . The results show that the λ value has a minimal impact on the positioning accuracy of the deployment plan because all values reach similar accuracies after 200 iterations, meeting the required standards. Figure 5 (b) emphasizes the stability of the algorithm under different λ values. When λ = 100, the average standard deviation of the algorithm is the smallest, while when λ = 0.1, the average standard deviation of the algorithm is the largest. Therefore, the penalty term coefficient mainly affects the stability of the algorithm rather than the positioning accuracy. The method for deploying drones in the mountainous and dense forest emergency rescue scenario provided by the embodiment of the present invention uses drones equipped with positioning network devices to provide positioning services for ground search and rescue personnel in view of the complex geographical environment, numerous obstacles, and damaged ground communication environment in the mountainous and dense forest emergency rescue scenario. The APF provided by the embodiment of the present invention has the effect of avoiding non-line-of-sight links, and the PSO algorithm considers how to combine with APF, and the two complement each other, thereby improving the positioning accuracy of ground search and rescue personnel. The following describes the device for deploying drones in the mountainous and dense forest emergency rescue scenario provided by the present invention. The device for deploying drones in the mountainous and dense forest emergency rescue scenario described below can be mutually referred to with the method for deploying drones in the mountainous and dense forest emergency rescue scenario described above. Figure 6This is a schematic structural diagram of a drone deployment device in a mountainous dense forest emergency rescue scenario provided by an embodiment of the present invention. As Figure 6 shown, the drone deployment device in a mountainous dense forest emergency rescue scenario provided by an embodiment of the present invention includes: A first construction module 601, configured to construct a virtual force model for non-line-of-sight avoidance, and solve the total resultant force of the virtual forces received by the drone according to the virtual force model for non-line-of-sight avoidance; A second construction module 602, configured to construct a particle swarm optimization model with each drone deployment plan as a particle, and optimize the speed and position of the particle by combining the total resultant force of the virtual forces of each drone to dynamically adjust the deployment position of the drone swarm; An output module 603, configured to iteratively solve the particle swarm optimization model until a preset positioning accuracy requirement is met or the maximum number of iterations is reached, and output the optimal deployment position. For the drone deployment device in a mountainous dense forest emergency rescue scenario provided by an embodiment of the present invention, by constructing a virtual force model for non-line-of-sight avoidance, the total resultant force of the virtual forces received by the drone is solved according to the virtual force model for non-line-of-sight avoidance; a particle swarm optimization model is constructed with each drone deployment plan as a particle, and the speed and position of the particle are optimized by combining the total resultant force of the virtual forces of each drone to dynamically adjust the deployment position of the drone swarm; the particle swarm optimization model is iteratively solved until a preset positioning accuracy requirement is met or the maximum number of iterations is reached, and the optimal deployment position is output. The present invention can effectively reduce the number of non-line-of-sight links by dynamically adjusting the deployment position of the drone swarm, and combines the particle swarm optimization model to solve the problem that the artificial potential field method has a high dependence on the initial deployment position of the drone and requires multiple adjustments of the initial position, thereby improving the positioning accuracy. Figure 7 An example of a schematic physical structure diagram of an electronic device is shown as Figure 7As shown, the electronic device may include: a processor 710, a communications interface 720, a memory 730, and a communication bus 740. Among them, the processor 710, the communications interface 720, and the memory 730 complete communication with each other through the communication bus 740. The memory 730 includes computer programs, an operating system, and acquired data. The processor 710 can call the logical instructions in the memory 730 to execute the method for deploying unmanned aerial vehicles (UAVs) in the mountain dense forest emergency rescue scenario. The method includes: constructing a virtual force model for non-line-of-sight avoidance, and solving the total resultant force of the virtual forces received by the UAV according to the virtual force model for non-line-of-sight avoidance; constructing a particle swarm optimization model with each UAV deployment plan as a particle, and optimizing the speed and position of the particles in combination with the total resultant force of the virtual forces of each UAV to dynamically adjust the deployment position of the UAV swarm; iteratively solving the particle swarm optimization model until the preset positioning accuracy requirement is met or the maximum number of iterations is reached, and outputting the optimal deployment position. In addition, when the logical instructions in the above-mentioned memory 730 can be implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the related technology, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes. On the other hand, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is used to execute the method for deploying UAVs in the mountain dense forest emergency rescue scenario provided by the above-mentioned various methods. The method includes: constructing a virtual force model for non-line-of-sight avoidance, and solving the total resultant force of the virtual forces received by the UAV according to the virtual force model for non-line-of-sight avoidance; constructing a particle swarm optimization model with each UAV deployment plan as a particle, and optimizing the speed and position of the particles in combination with the total resultant force of the virtual forces of each UAV to dynamically adjust the deployment position of the UAV swarm; iteratively solving the particle swarm optimization model until the preset positioning accuracy requirement is met or the maximum number of iterations is reached, and outputting the optimal deployment position. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative work. Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the related technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments. Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for deploying unmanned aerial vehicles in an emergency rescue scenario in mountainous dense forests, characterized in that, Including: Construct a virtual force model for non-line-of-sight avoidance, and solve the total resultant virtual force on the UAV according to the virtual force model for non-line-of-sight avoidance; Use each UAV deployment plan as a particle to construct a particle swarm optimization model, and optimize the speed and position of the particle by combining the total resultant virtual force of each UAV to dynamically adjust the deployment position of the UAV swarm; Iteratively solve the particle swarm optimization model until the preset positioning accuracy requirement is met or the maximum number of iterations is reached, and output the optimal deployment position.

2. The method for deploying an unmanned aerial vehicle in an emergency rescue scenario in mountainous dense forests according to claim 1, wherein The constructing a virtual force model for non-line-of-sight avoidance and solving the total resultant virtual force on the UAV according to the virtual force model for non-line-of-sight avoidance includes: Obtain the virtual force on each UAV, where the virtual force includes an anti-collision virtual repulsive force, a positioning virtual attractive force, and a non-line-of-sight avoidance virtual force; Construct a virtual force model for non-line-of-sight avoidance based on the anti-collision virtual repulsive force, the positioning virtual attractive force, and the non-line-of-sight avoidance virtual force, calculate the resultant force of the anti-collision virtual repulsive force, the positioning virtual attractive force, and the non-line-of-sight avoidance virtual force through the virtual force model for non-line-of-sight avoidance, and use the resultant force as the total resultant virtual force.

3. The method for deploying an unmanned aerial vehicle in an emergency rescue scenario in mountainous dense forests according to claim 2, wherein, The obtaining the virtual force on each UAV includes: The anti-collision virtual repulsive force between the i-th UAV and an obstacle or another UAV k is expressed as: Among them, K rep represents the virtual repulsive force factor, represents the virtual repulsive force effective threshold, β is a constant representing the reduction factor, p i represents the UAV coordinate, o k represents the obstacle or the UAV center coordinate; User u j applies a virtual gravitational force to the drone subgroup g j = [p i , p j , p k and is expressed as: Among which K att represents the coefficient factor for positioning the virtual gravity, and P is the constructed artificial potential field; Definition Denote the unit normal vector of the straight line connecting the i-th drone and the j-th ground search and rescue personnel. The non-line-of-sight avoidance virtual force exerted by the ground search and rescue personnel on the drone is expressed as: Among them, K eva represents the NLOS avoidance virtual force coefficient, p i represents the coordinate point of the i-th UAV, u j represents the coordinate point of the j-th ground search and rescue personnel.

4. The method for deploying an unmanned aerial vehicle in an emergency rescue scenario in mountainous dense forests according to claim 1, wherein, The using each UAV deployment plan as a particle to construct a particle swarm optimization model and optimizing the speed and position of each particle by combining the total resultant virtual force of each UAV to dynamically adjust the deployment position of the UAV swarm includes: Generate an initial particle swarm, and set the initial position set of the UAV swarm, the initial connection matrix between the UAV and the ground search and rescue personnel, and the initial average positioning accuracy; For each particle, calculate the average positioning accuracy of the current particle; If the average positioning accuracy of the current particle is better than the historical optimal value, update the global optimal solution, and record the optimal accuracy and its corresponding UAV position and connection matrix; Combine the inertia weight, learning factor, random number, and the total resultant virtual force on the UAV to jointly update the speed of the particle; Adjust the position of the UAV according to the updated speed to ensure that the maximum flight speed is not exceeded; After completing the round iteration and particle iteration, return the global optimal deployment position and its corresponding average positioning accuracy.

5. The method for deploying an unmanned aerial vehicle in an emergency rescue scenario in mountainous dense forests according to claim 4, wherein, The combining the inertia weight, learning factor, random number, and the total resultant virtual force on the UAV to jointly update the speed of the particle includes: Obtain the algorithm input parameters, where the algorithm input parameters include the number N of unmanned aerial vehicles, the number M of ground search and rescue personnel, the transmission power of the unmanned aerial vehicle signal, the environmental noise power, the positioning accuracy threshold ∈, and the maximum moving speed v of the unmanned aerial vehicle max ; Obtain the number of particles in the particle swarm according to the number of UAVs N and the number of ground search and rescue personnel M; Input the transmission power of the UAV signal and the environmental noise power into the channel model, and obtain the total resultant virtual force on each particle according to the channel transmission loss; According to the number of the particles, the positioning accuracy threshold ∈, and the maximum moving speed v of the drone max and update the velocity of the particle according to the total resultant force of the virtual forces on the particles. The velocity update formula is p i = p i + V i ; where ω is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers in [0, 1], and V i = [v p,x , v p,y , v p,z is the velocity vector of the i-th UAV, represents the total resultant force of the virtual forces on the particle, and g best is the historical best position, and p i is the position of the particle.

6. The method for deploying an unmanned aerial vehicle in an emergency rescue scenario in mountainous dense forests according to claim 4, wherein The positioning accuracy is calculated through the time difference of arrival model between the ground user and the UAV, and the positioning result is evaluated based on the geometric dilution of precision factor.

7. The method for deploying an unmanned aerial vehicle in an emergency rescue scenario in mountainous dense forests according to claim 5, wherein The obtaining the algorithm input parameters includes: Obtain the algorithm input parameters through the channel model, where the channel model includes line-of-sight link loss and non-line-of-sight link loss: Line-of-sight link loss Non-line-of-sight link loss Among them, f represents the carrier frequency, d0 represents the reference distance, c represents the speed of light; n LoS represents the path loss factor under the line-of-sight link, n NLoS represents the path loss factor under the non-line-of-sight link, L F (d0) represents the path loss of free space at the reference distance d0 between the i-th drone and the j-th ground search and rescue personnel, L Slant represents the additional loss in the forest area due to the inclination of the connection line between the drone and the ground user, X LoS represents the forest area shadow fading under the line-of-sight link, X NLoS represents the forest area shadow fading under the non-line-of-sight link, d i,j is the distance between the i-th drone and the j-th ground search and rescue personnel.

8. An unmanned aerial vehicle deployment device in an emergency rescue scenario in mountainous dense forests, characterized in that, Including: A first construction module for constructing a virtual force model for non-line-of-sight avoidance and solving the total resultant virtual force on the UAV according to the virtual force model for non-line-of-sight avoidance; A second construction module, configured to use each UAV deployment scheme as a particle to construct a particle swarm optimization model, and optimize the velocity and position of the particle by combining the total resultant force of the virtual forces of each UAV to dynamically adjust the deployment positions of the UAV swarm; An output module, configured to iteratively solve the particle swarm optimization model until a preset positioning accuracy requirement is met or the maximum number of iterations is reached, and output the optimal deployment position.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, When the processor executes the program, it implements the UAV deployment method in the mountainous and forested emergency rescue scenario according to any one of claims 1 to 7.

10. A non-transitory readable storage medium, on which a computer program is stored, characterized in that, When the computer program is executed by a processor, it implements the UAV deployment method in the mountainous and forested emergency rescue scenario according to any one of claims 1 to 7.

Images