A method of deploying ground teams for space search and rescue based on grid technology

Abstract

The present invention discloses a method for deploying a space search and rescue ground team based on grid technology, the method comprising: determining a theoretical landing area with the theoretical landing point of the returner as the center, dividing the theoretical landing area into a number of grids and calculating the landing probability of each grid; taking the search and rescue area around the theoretical landing area as the deployment area of ​​the space search and rescue ground team, dividing the deployment area into grids and numbering them; taking the expected time for all space search and rescue ground teams to arrive at each spacecraft landing point grid from the deployment area as the optimization index, establishing the optimization problem of the deployment plan of the space search and rescue ground team; using the simulated annealing algorithm to solve the optimization problem, and outputting the standby grid number of each space search and rescue ground team. The method of the present invention can obtain the optimal deployment plan, so that the ground team can arrive at the returner landing site as soon as possible and carry out disposal work in time.

Description

Technical Field

[0001] The invention relates to the field of aerospace search and rescue, and in particular to a method for deploying aerospace search and rescue ground teams based on grid technology. Background Art

[0002] With the in-depth advancement of my country's manned space program and lunar exploration program, various space search and rescue missions are showing a trend of high frequency, normalization, and increasing emergency events, which requires the space search and rescue ground team to be able to quickly reach the returner landing point in a relatively short period of time. A scientific and reasonable ground team deployment plan is conducive to ensuring the safety of the team, improving the search and rescue efficiency of the team, and shortening the time required to complete the mission.

[0003] In the field of aerospace search and rescue deployment, there is currently a lack of corresponding mathematical models, which makes it impossible to quantify the effectiveness of deployment plans and scientifically solve deployment plans. A problem similar to the deployment problem is the force deployment problem, which requires the formulation of a force configuration plan based on the enemy situation, battlefield environment and mission objectives to maximize the satisfaction of mission requirements. At present, research on force deployment at home and abroad is basically concentrated in combat fields such as position attack and defense and air defense operations. Li Zhigang et al. established a deployment model based on two-layer integer programming for the deployment of position attack and defense forces. The first layer is the battalion-level unit decision-making layer, which aims to seize strongholds and eliminate enemy forces as soon as possible, and uses the fixed total number of each force as a constraint. The dynamic programming algorithm is used to determine the force allocation of the company-level unit. The second layer is the company-level unit decision-making layer. This layer uses the decision results of the first layer as a constraint condition, and the goal is to maximize the damage to the enemy per unit time. The dynamic programming algorithm is used to formulate the depth configuration plan of each force. In terms of air defense force deployment, Ren Jiye, Shen Maoxing and others regarded the outcome of the confrontation between the enemy and us as a probabilistic event, and used the linear programming method to solve the problem of air defense force deployment with the goal of maximizing the probability of defeating the enemy's attack. Kong Xiangyu introduced game theory into the problem of regional air defense force deployment. On the basis of measuring the air combat capability index of air defense forces, he considered the uncertain factors in the war, introduced random correction factors, and established a force deployment game model using the game matrix method. The particle swarm algorithm was used to solve the Nash equilibrium point of the model and obtain the mixed strategy of the combatants. Xiong Zhe and others considered the problem of air defense force deployment of mixed groups, comprehensively considered the importance of defending targets, the probability of successful defense, the composition of mixed forces and battlefield environmental factors, and used the weighted sum of the product of the combat effectiveness of the combat unit on the defended key points and the importance of the key points as the optimization objective function. With the number of troops as a constraint, an improved genetic algorithm was used to obtain the force deployment plan.

[0004] Compared with the traditional force deployment model, the deployment in the field of space search and rescue has the following two important differences:

[0005] First, the objects are different. The object of the space search and rescue mission is the returner, which is different from the traditional combat object in the field of force deployment. In the traditional force deployment model, there is a lack of ready-made returner models, and there is a lack of relevant research on how the returner status affects the force deployment.

[0006] Second, the purpose is different. The research on force deployment in the combat field is a decision-making process to strike the opponent as hard as possible and defend our goals in a hostile state of "life and death", while the purpose of space search and rescue deployment is to allow the ground team to reach the landing site of the returner as soon as possible and carry out disposal work in a timely manner.

[0007] Therefore, the space search and rescue deployment problem is quite different from the traditional force deployment problem, and the existing force deployment model cannot be directly used.

[0008] At present, the research on force deployment in the field of space search and rescue is still in its infancy, and there are mainly two problems:

[0009] First, there is a lack of ready-made mathematical models for space search and rescue. At present, there is still a lack of unified modeling of returners, search and rescue areas, and search and rescue teams. This leads to a lack of research basis for the search and rescue deployment problem, which seriously restricts the solution of the search and rescue deployment plan.

[0010] Second, there is a lack of quantitative description and solution methods for the search and rescue deployment problem, a lack of systematic research on the optimization indicators and constraints of deployment tasks, no directly available deployment model, and no ready-made solution algorithm that can be used to obtain search and rescue deployment plans. Summary of the invention

[0011] The purpose of the present invention is to overcome the above-mentioned technical defects and proposes a method for deploying ground teams for space search and rescue based on grid technology, including: completing modeling of the space search and rescue area based on grid technology; establishing a mathematical model of the returner based on the landing dispersion probability; taking the shortest time for the team corresponding to the plan to arrive at the landing point as an indicator, comprehensively considering the team safety and regional accessibility constraints, and establishing a mathematical model of the ground team deployment plan; and using a simulated annealing algorithm to solve and obtain the deployment plan.

[0012] To achieve the above object, the present invention proposes a method for deploying a space search and rescue ground team based on grid technology, the method comprising:

[0013] Determine the theoretical landing area with the theoretical landing point of the returner as the center, divide the theoretical landing area into several grids and calculate the landing probability of each grid;

[0014] The search and rescue area around the theoretical landing area is used as the deployment area for the space search and rescue ground team, and the deployment area is divided into grids and numbered;

[0015] Taking the expected time of all space search and rescue ground teams arriving at each spacecraft landing point grid from the deployment area as the optimization index, the optimization problem of the deployment plan of space search and rescue ground teams is established;

[0016] The simulated annealing algorithm is used to solve the optimization problem and output the standby grid number of each aerospace search and rescue ground team.

[0017] As an improvement of the above method, the theoretical landing area is determined with the theoretical landing point of the returner as the center, the theoretical landing area is divided into a number of grids and the landing probability of each grid is calculated; specifically, it includes:

[0018] The theoretical landing area is determined with the theoretical landing point of the returner as the center, and the theoretical landing area is divided into several grids: U D ={d 1 ,…,d D}, where D represents the total number of theoretical landing area grids, d i Indicates the number of the i-th grid, i = 1, 2, ..., D;

[0019] Taking into account the influence of the earth's gravity, air resistance, and the earth's rotation, the returner landing point prediction dynamics model is established as follows:

[0020] Let r be the center of mass position of the returner, and are the first-order and second-order derivatives of the returner relative to the geocentric system, then the dynamic equation of the returner's center of mass motion is: Where m is the mass of the returner, g is the acceleration due to gravity, and F D For pneumatic power, is the traction force, F k =-mω×(ω×r) is the Coriolis force, ω is the angular velocity of the earth's rotation;

[0021] Generate a large number of returner initial positions and returner initial velocities with random deviations, use them as the initial conditions of the returner center-of-mass motion dynamics equation, input them into the returner landing point prediction dynamics model, use the Runge-Kutta method to numerically solve the center-of-mass dynamics equation, simulate the returner landing process, and then obtain a series of returner landing points;

[0022] By counting the number of simulated landings of the returner in each grid area, the landing probability of the returner in each grid can be obtained: i The landing probability in is p i .

[0023] As an improvement of the above method, the optimization problem of the deployment plan of the space search and rescue ground team is established by taking the expected time for all the space search and rescue ground teams to arrive at each spacecraft landing point grid from the deployment area as the optimization index; specifically including:

[0024] The grid number set of the potential area is denoted as Ω m , the subscript m represents the total number of grids available for layout;

[0025] When the returner lands on the i The kth space rescue ground team starts from the initial position P k Start and calculate the shortest path to reach grid d i Time required ki , then at least one space search and rescue ground team arrives at number d i The shortest time of the grid is K is the total number of space search and rescue ground teams;

[0026] Traversing all possible landing situations of the returner, the expected time to complete the search mission is Therefore, solving the optimal layout problem can be converted into solving the initial position P of the k-th search and rescue team. k , requiring P k Located in the potential area Ω m And the expected time to complete the task is the shortest, then the optimization problem is:

[0027]

[0028] As an improvement of the above method, the method further includes: for the kth space search and rescue ground team, calculating the pass factor of its adjacent grids in the theoretical landing area; specifically including:

[0029] The maximum ideal speed that the kth space rescue ground team can achieve in the theoretical landing area is V max , then the maximum speed of the team from grid m to grid n is V mn , then the transit factor γ of the team from grid m to grid n mn Defined as The maximum travel speed between two grids is determined based on the terrain data within each grid and the travel conditions of the squad's convoy. Using the topological graph method, the center point of each grid area is abstracted into a vertex, and adjacent grids are connected by directed edges. Each directed edge is assigned a weight, which represents the travel factor of the squad from the starting point to the end point of the directed edge.

[0030] As an improvement of the above method, when the returner lands on the i The kth space search and rescue ground team starts from the initial position P k Start and calculate the optimal path to reach grid d i Minimum time required t ki , including:

[0031] When the returner lands on the i The kth space search and rescue ground team starts from the initial position P k Starting from, the possible travel factor from two adjacent grids m to grid n is γ mn , then the speed V of the team between these two grids mn =V max γ mn , maximum speed V max ; The travel time from grid m to grid n is Among them l mn is the distance between the center points of grid m and grid n;

[0032] After obtaining the travel time of the kth space search and rescue ground team between each grid, the Dijkstra algorithm in graph theory is used to calculate the optimal path between the team and the returner landing grid and the corresponding shortest time t ki .

[0033] As an improvement of the above method, the simulated annealing algorithm is used to solve the optimization problem and output the standby grid number of each aerospace search and rescue ground team; specifically, it includes:

[0034] Step S1) The optimization objective function of the search and rescue deployment problem is recorded as Where S represents the deployment plan, which gives the grid numbers of all space search and rescue ground teams at the initial moment;

[0035] Step S2) Setting the initial annealing temperature T and the stop temperature T f and the number of inner loops N, select one of the feasible layout schemes as the initial value S of the scheme;

[0036] Step S3) The number of iterations l is set to 0;

[0037] Step S4) randomly generates a new set of layout schemes S′;

[0038] Step S5) calculate the annealing temperature difference Δt=t(S')-t(S); determine whether Δt<0 holds, if so, accept S' as the current optimal solution, set S equal to S', and proceed to step S6); otherwise, accept S' as the current optimal solution with a probability of exp(-Δt / T), set S equal to S', and proceed to step S6);

[0039] Step S6) add 1 to the value of l to determine whether l is less than N. If yes, proceed to step S4); otherwise, proceed to step S7);

[0040] Step S7) If the current temperature T is less than the stop temperature T f , go to step S3), otherwise, go to step S8);

[0041] Step S8) Update T with a temperature strictly less than T, so that T tends to 0; when T is greater than the stop temperature T f When , go to step S3), otherwise, go to step S9);

[0042] Step S9) outputs S as the final layout plan.

[0043] The advantages of the present invention are:

[0044] 1. The present invention establishes a grid model of the search area; by drawing on the geographic space information grid technology, the ground area is divided into a series of grids using regular rectangular units, and after the grids are given a pass factor, a directed graph is used to construct the grid model of the search area;

[0045] 2. The method of the present invention adopts a landing probability model to replace the previous latitude and longitude landing point prediction model, uses the returner landing point prediction dynamics model to obtain the landing probability of the returner in each grid, and uses the landing probability as the only attribute of the returner to establish a returner model for space search and rescue missions;

[0046] 3. The present invention establishes a mathematical model for the search and rescue deployment problem, takes the weighted shortest time for the team to reach the landing position of the returner as the optimization goal, takes the team safety and the number of teams as constraints, and establishes a discrete programming model that can solve the deployment plan;

[0047] 4. The method of the present invention uses a simulated annealing algorithm to solve the search and rescue deployment model and obtains the optimal deployment plan; this plan enables the ground team to arrive at the returner landing site as soon as possible and carry out disposal work in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is the traffic factor topology diagram of the present invention;

[0049] Figure 2 It is a flow chart of the method for deploying a space search and rescue ground team based on grid technology of the present invention;

[0050] Figure 3 It is a flow chart of solving the optimal potential solution based on the simulated annealing algorithm of the present invention;

[0051] Figure 4(a) is a schematic diagram of the theoretical landing area and the potential area;

[0052] FIG4( b ) is a schematic diagram of a first layout scheme obtained using the method of the present invention;

[0053] FIG4( c ) is a schematic diagram of a second layout scheme obtained using the method of the present invention;

[0054] FIG4( d ) is a schematic diagram of a third layout scheme obtained using the method of the present invention. DETAILED DESCRIPTION

[0055] The technical solution of the present invention is described in detail below with reference to the accompanying drawings.

[0056] The deployment of each team before executing a space search and rescue mission is called the deployment of the search and rescue team. The existing ground team deployment plan is based on historical mission experience and is formulated according to factors such as the landing area of ​​the space returner and the performance of the search and rescue vehicle. Due to the lack of mathematical models related to the deployment of space search and rescue teams and deployment solution algorithms, the existing deployment model has the disadvantages of insufficient scientificity of the deployment plan and the inability to quantify the search and rescue efficiency of the plan.

[0057] Therefore, the present invention proposes a method for deploying a space search and rescue ground team based on grid technology, which is specifically as follows:

[0058] 1. Technical approach

[0059] The method of ground team deployment for space search and rescue first constructs models of various factors that affect the deployment plan, including the geographical environment model of the search and rescue area, the ground team model and the returner model, so as to establish a mathematical model of space search and rescue; then, the time for the team to arrive at the returner landing point is used as the optimization indicator, and the safety of the team is used as the constraint condition to establish a mathematical description of the search and rescue deployment problem; finally, the simulated annealing algorithm is used to solve the deployment problem and obtain the deployment plan.

[0060] 1.1 Establishing a mathematical model

[0061] 1.1.1 Modeling of geographical factors in search and rescue areas

[0062] Geographical factors such as lakes, rivers, cliffs, valleys, mountains, fences, roads, etc. in the search and rescue area have a great impact on the search and rescue activities, and the geographical environment in the search and rescue area needs to be modeled first. Gridding is the basis of geographic information systems and path planning methods, and is a widely used method that can effectively represent the geographical environment. This patent draws on the method of geographic spatial information grids and uses regular grid units to grid the ground area, that is, to divide the search and rescue area into sufficiently small and regular rectangular grids, and give the grids corresponding geographical attributes to approximate the real surface environment, thereby achieving modeling of complex geographical environments.

[0063] The size of the rectangular grid is an important parameter for grid division. If the selected grid side length is too large, on the one hand, it will not be enough to approximate the real geographical environment, resulting in distortion of environmental modeling. On the other hand, it will cause the search and rescue team's detection range to be unable to cover the grid where it is located, resulting in the situation where even if the team and the returner are in the same grid, they cannot detect the returner. On the contrary, if the grid side length is too small, it will greatly increase the amount of calculation of the ground team's deployment plan, and may even make the problem unsolvable. Therefore, it is necessary to select an appropriate grid size according to the task requirements. In this patent model, the principle of side length parameter selection is to round down the detection equipment's recognition distance of the search and rescue target.

[0064] The passability attribute of a grid is an indicator that comprehensively reflects the search and rescue team's ability to pass within the grid. It directly affects the team's forward route planning, and then affects the choice of deployment plan. The maximum travel speed that the same vehicle can reach is different in different terrains. In the same terrain, the travel speeds of different types of vehicles are also different. At the same time, the passability attribute is also related to the team's travel direction. If there is an east-west gully in the grid, the vehicle can pass through the grid when traveling in the east-west direction, but it will be blocked when passing in the north-south direction. The present invention uses a passability factor to describe the passability attribute. The passability factor is a real number between 0 and 1. The closer it is to 1, the higher the passability and the greater the forward speed that the team can achieve. The calculation method of the passability factor is as follows. Assume that the maximum ideal speed that a team can achieve in the search and rescue area is V max , the team numbered m The maximum speed of the grid to the grid numbered n is V mn , then the team from the grid m The access factor γ to grid n mn Defined as In order to intuitively represent the traffic factor of the squad, the center point of each grid area is abstracted into a vertex by adopting the topological graph method. Adjacent grids are connected by directed edges. Each directed edge is assigned a weight, which represents the traffic factor of the squad from the starting point to the end point of the directed edge. Figure 1 Schematic diagram of the traffic factor topology of 4 grids.

[0065] V max The maximum travel speed between two grids can be determined artificially by collecting topographic data in each grid through manual survey and combining it with the actual travel conditions of the fleet.

[0066] 1.1.2 Ground search and rescue team modeling

[0067] The ground search and rescue team is modeled as a point target with position parameters, maximum travel speed and detection range parameters. In the deployment problem, the team position and maximum speed mainly affect the planning of the search and rescue path and the calculation of the time to complete the search and rescue, and the detection range is mainly used to determine the grid size.

[0068] Only when the travel path of each team is determined can the shortest time required for each team to reach the returner landing point be determined. Therefore, the planning of the optimal path is the basis for solving the deployment problem. In order to obtain the shortest path between the search and rescue team and the returner landing point, it is necessary to calculate the time required for the team to travel between grid areas from the traffic factor topology map. To this end, firstly set the maximum travel speed V of the search and rescue team max Multiply it by the corresponding traffic factor to get the speed of the team between two grids. For example, to get the time t required for the search and rescue team to reach grid j from grid i ij , the maximum speed V max Multiply by the traffic factor γ between grid i and grid j ij Get the speed V of the team between these two grids ij =V max γ ij , the corresponding travel time can be further obtained as Among them l ij is the distance between the center points of grid i and grid j. After obtaining the travel time of the team between each grid, the Dijkstra algorithm in graph theory can be used to obtain the optimal path and the shortest time required between the team and the returner landing grid.

[0069] 1.1.3 Returner Modeling

[0070] The landing information of the returner determines the search and rescue recovery area and has a great impact on the deployment plan. The previous landing point forecast belongs to the point forecast mode, which predicts the longitude and latitude of the theoretical landing point of the returner. Due to the existence of forecast errors, there may be a certain deviation between the actual landing point of the returner and the theoretical landing point, and the point forecast mode cannot provide the landing range information of the returner. In order to comprehensively consider the impact of forecast errors, the landing probability forecast mode is used to replace the previous point forecast mode in this patent model. The landing probability forecast mode refers to predicting the landing probability of the returner in each grid. At this time, the predicted landing point will be replaced by the predicted landing area and landing probability. The predicted landing probability can quantitatively reflect the landing distribution range of the returner.

[0071] In order to calculate the landing probability of the returner in each grid, the factors affecting the returner's landing point are first taken as model input parameters, and the influence of the earth's gravity, air resistance, and the earth's rotation are comprehensively considered to establish the returner landing point prediction dynamic model as follows: r is the position of the returner's center of mass, and are the first-order and second-order derivatives of the returner relative to the geocentric system, then the dynamic equation of the returner's center of mass motion is: Where m is the mass of the returner, g is the acceleration due to gravity, and F D For pneumatic power, is the traction force, Fk =-mω×(ω×r) is the Coriolis force, ω is the angular velocity of the earth's rotation.

[0072] Then, a large number of returner initial positions and returner initial velocities with random deviations are generated by computer, and used as the initial conditions of the returner center of mass motion dynamics equation. They are substituted into the landing point prediction model, and the center of mass dynamics equation is numerically solved using the Runge-Kutta method to simulate the returner landing process, thereby obtaining a series of returner landing points. Finally, the number of simulated landings of the returner in each grid area is counted to obtain the landing probability of the returner in each grid. After obtaining the landing probability of the returner, the landing range of the returner is quantitatively described, which can provide search and rescue target data support for the subsequent solution of the deployment problem.

[0073] 1.2 Search and rescue deployment

[0074] The present invention describes the deployment problem of the ground team as follows:

[0075] The optimal deployment problem based on the theoretical landing probability is to determine the standby position of each team at the initial moment of the mission, given the landing probability of the returner in each grid in the theoretical landing area, so that each team is on standby outside the theoretical landing area and the expected time to complete the search and rescue mission is the shortest. The shortest search and rescue time refers to the minimum time for at least one ground team to reach the landing point of the returner from the initial moment.

[0076] The flow chart of the ground search and rescue team deployment algorithm is as follows: Figure 2 As shown in the figure, the algorithm needs to use the grid access factor, the landing probability of the returner in each grid, the maximum travel speed of each team and the deployment area as input parameters. The deployment area refers to the area where the search and rescue force vehicles can be deployed at the initial moment, which is defined as the search and rescue area excluding the theoretical landing area. The deployment solution algorithm outputs the standby grid number of each search and rescue team.

[0077] Assume that the grid number corresponding to the theoretical landing area is U D ={d 1 ,…,d D}, where D represents the total number of theoretical landing area grids, d i (i=1,2,…,D) represents the number of the i-th grid. The grid number set of the potential area is denoted by Ω m , the subscript m indicates that the total number of grids available for deployment is m. j The landing probability in is p j (j=1,2,…,D). Assume that the returner lands on the j The i-th search and rescue team starts from the initial position P i Start and arrive at grid d along the shortest path jThe time required is t ij , then the shortest time for at least one search and rescue team to arrive at the returner landing grid is Traversing all possible landing situations, we can see that the expected time to complete the search mission is Therefore, solving the optimal layout problem can be converted into solving the initial position P of the i-th search and rescue team. i , requiring P i Located in the potential area Ω m And the expected time to complete the task is the shortest, that is, to solve the optimization problem:

[0078]

[0079] In order to overcome the difficulty of completely solving the huge amount of calculation, this patent uses a simulation method to solve the ground team deployment problem and obtain a "near-optimal" reasonable deployment plan within the allowed time. In order to use the simulated annealing algorithm, the optimization objective function of the search and rescue deployment problem is recorded as Where S represents the deployment plan, which gives the grid number where the i-th search and rescue team is located at the initial moment. The algorithm flow chart of using simulated annealing algorithm to solve the ground team deployment problem is as follows Figure 3 shown.

[0080] The specific steps of the optimal layout solution algorithm are as follows:

[0081] (1) Set the initial annealing temperature T and the stop annealing temperature T f , the number of inner loops N, select one of the feasible layout schemes as the initial value S of the scheme;

[0082] (2) Iterate steps (3) to (5) N times;

[0083] (3) Randomly generate a new set of layout schemes S′;

[0084] (4) Calculate the annealing temperature difference Δt=t(S')-t(S);

[0085] (5) If Δt < 0, then accept S′ as the current optimal solution, that is, let S equal S′; otherwise, accept S′ as the current optimal solution with a probability of exp(-Δt / T);

[0086] (6) If the current temperature T is less than the stop temperature T f , then go to step (8), otherwise go to step (7);

[0087] (7) Use a temperature strictly less than T to update T. At this time, it is required that T tends to 0 as the number of cycle steps increases. When T is greater than the stop temperature T f When , go to step (2); otherwise, go to step (8);

[0088] (8) Output S as the final layout solution obtained by the algorithm.

[0089] 2. Technical Effect

[0090] The technical effect of the present invention is to provide a deployment plan for the search and rescue team. Two sets of simulation experiments are conducted to verify the rationality and demonstration of the ground team deployment algorithm of the present invention.

[0091] The first calculation example is shown in Figure 4(a). Assume that the area to be deployed is the grid corresponding to the edge of the search and rescue area. The search and rescue area is divided into 9×7 square grids, each grid is 6km long, and the theoretical landing area of ​​the returner is divided into 7×5 grids. Figure 4(b) is an abstract schematic diagram of the search and rescue area. The gray area in the figure represents the deployment area, the white grid represents the passable grid, and the black grid represents the inaccessible grid, which corresponds to the lake in the figure. The numbers in the white area in the figure simulate the landing probability of the returner in each grid with the center point of the area as the aiming point. It can be seen from the figure that the landing probability of the grid at the center point of the area is the largest, and the landing probability corresponding to the grid farther from the center point is smaller.

[0092] The number of search and rescue teams is set to 8. To simplify the verification parameters, it is assumed that the travel speed of each team in the search and rescue area is the same, which is 60 km / h. At this time, the travel factor of each team between any two adjacent passable grids is 1, and the travel time between two grids can be obtained by dividing the straight-line distance between the midpoints of the grids by the vehicle speed.

[0093] The deployment scheme obtained by using the deployment algorithm of the present invention is shown in the dots in Figure 4 (b). The numbers below the dots represent the team numbers. The average search and rescue time corresponding to the deployment scheme is 15.53 minutes, the fastest task completion time is 12.00 minutes, and the slowest search and rescue can be completed within 18.00 minutes. It can be seen from the figure that the search and rescue teams in the optimal deployment scheme are relatively scattered in the region. This is because if the teams are concentrated in the area near the theoretical landing point at the initial moment, then when the returner has a small probability and a large landing deviation, the search and rescue team will not be able to reach the returner landing point in time, resulting in a long time to complete the search and rescue. The optimal deployment algorithm proposed in this article can automatically avoid the search and rescue team from "clustering" deployment, which is in line with previous deployment experience. At the same time, in this deployment scheme, the deployment of the team avoids several grids in the lower right corner. This is because if the team is arranged in the lower right corner, it will be blocked by the inaccessible area, and the team needs to detour to go to the returner landing area, which increases the time to complete the search and rescue and reduces the search and rescue efficiency.

[0094] When the parameters of the search and rescue area remain unchanged, the vehicle speed parameters are changed to study the optimal deployment plan when the team speeds are different. For this purpose, assuming that the speeds of search and rescue teams 1 and 2 are 80 km / h, and the speeds of the other search and rescue teams are 60 km / h, and other parameters remain unchanged, two optimal deployment plans can be obtained as shown in Figure 4 (c) and Figure 4 (d). The average time required to complete the task for both plans is 12.39 minutes. Therefore, for the optimal deployment problem, the solution is not unique, and there may be several different optimal plans.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention is described in detail with reference to the embodiments, it should be understood by those skilled in the art that any modification or equivalent replacement of the technical solutions of the present invention does not depart from the spirit and scope of the technical solutions of the present invention and should be included in the scope of the claims of the present invention.

Claims

1. A method for deploying a space search and rescue ground team based on grid technology, the method comprising: Determine the theoretical landing area with the theoretical landing point of the returner as the center, divide the theoretical landing area into several grids and calculate the landing probability of each grid; The search and rescue area around the theoretical landing area is used as the deployment area for the space search and rescue ground team, and the deployment area is divided into grids and numbered; Taking the expected time of all space search and rescue ground teams arriving at each spacecraft landing point grid from the deployment area as the optimization index, the optimization problem of the deployment plan of space search and rescue ground teams is established; The simulated annealing algorithm is used to solve the optimization problem and output the standby grid number of each aerospace search and rescue ground team; The theoretical landing area is determined with the theoretical landing point of the returner as the center, the theoretical landing area is divided into a number of grids and the landing probability of each grid is calculated; Specifically include: The theoretical landing area is determined with the theoretical landing point of the returner as the center, and the theoretical landing area is divided into several grids: U D ={d1,…,d D }, where D represents the total number of theoretical landing area grids, d i Indicates the number of the i-th grid, i = 1, 2, ..., D; Taking into account the influence of the earth's gravity, air resistance, and the earth's rotation, the returner landing point prediction dynamics model is established as follows: Let r be the center of mass position of the returner, and are the first-order and second-order derivatives of the returner relative to the geocentric system, then the dynamic equation of the returner's center of mass motion is: Where m is the mass of the returner, g is the acceleration due to gravity, and F D For pneumatic power, is the traction force, F k =-mω×(ω×r) is the Coriolis force, ω is the angular velocity of the earth's rotation; Generate a large number of returner initial positions and returner initial velocities with random deviations, use them as the initial conditions of the returner center-of-mass motion dynamics equation, input them into the returner landing point prediction dynamics model, use the Runge-Kutta method to numerically solve the center-of-mass dynamics equation, simulate the returner landing process, and then obtain a series of returner landing points; By counting the number of simulated landings of the returner in each grid area, the landing probability of the returner in each grid can be obtained: i The landing probability in is p i .

2. The method for deploying a space search and rescue ground team based on grid technology according to claim 1, characterized in that: The optimization problem of the deployment plan of the space search and rescue ground teams is established by taking the expected time for all the space search and rescue ground teams to arrive at each spacecraft landing point grid from the deployment area as the optimization index; specifically including: The grid number set of the potential area is denoted as Ω m , subscript m Indicates the total number of grids available for layout; When the returner lands on the i The kth space search and rescue ground team starts from the initial position P k Start and calculate the shortest path to reach grid d i Time required ki , then at least one space search and rescue ground team arrives at number d i The shortest time of the grid is K is the total number of space search and rescue ground teams; Traversing all possible landing situations of the returner, the expected time to complete the search mission is Therefore, solving the optimal layout problem can be converted into solving the initial position P of the k-th search and rescue team. k , requiring P k Located in the potential area Ω m And the expected time to complete the task is the shortest, then the optimization problem is:

3. The method for deploying a space search and rescue ground team based on grid technology according to claim 2, characterized in that: The method further includes: for the kth space search and rescue ground team, calculating the pass factor of the adjacent grids in the theoretical landing area; specifically including: The maximum ideal speed that the kth space rescue ground team can achieve in the theoretical landing area is V max , then the maximum speed of the team from grid m to grid n is V mn , then the transit factor γ of the team from grid m to grid n mn Defined as The maximum travel speed between two grids is determined based on the terrain data within each grid and the travel conditions of the squad's convoy. Using the topological graph method, the center point of each grid area is abstracted into a vertex, and adjacent grids are connected by directed edges. Each directed edge is assigned a weight, which represents the travel factor of the squad from the starting point to the end point of the directed edge.

4. The method for deploying a space search and rescue ground team based on grid technology according to claim 3 is characterized in that: When the returner lands on the i The kth space search and rescue ground team starts from the initial position P k Start, calculate the time-consuming optimal path to reach grid d i Minimum time required t ki , specifically including: When the returner lands on the i The kth space search and rescue ground team starts from the initial position P k Starting from, the possible travel factor from two adjacent grids m to grid n is γ mn , then the speed V of the team between these two grids mn =V max γ mn , maximum speed V max ; The travel time from grid m to grid n is Among them l mn is the distance between the center points of grid m and grid n; After obtaining the travel time of the kth space search and rescue ground team between each grid, the Dijkstra algorithm in graph theory is used to calculate the optimal path between the team and the returner landing grid and the corresponding shortest time t ki .

5. The method for deploying a space search and rescue ground team according to claim 4, characterized in that: The simulated annealing algorithm is used to solve the optimization problem and output the standby grid number of each aerospace search and rescue ground team; specifically includes: Step S1) The optimization objective function of the search and rescue deployment problem is recorded as Where S represents the deployment plan, which gives the grid numbers of all space search and rescue ground teams at the initial moment; Step S2) Setting the initial annealing temperature T and the stop temperature T f and the number of inner loops N, select one of the feasible layout schemes as the initial value S of the scheme; Step S3) The number of iterations l is set to 0; Step S4) randomly generates a new set of layout schemes S′; Step S5) calculate the annealing temperature difference Δt=t(S')-t(S); determine whether Δt<0 holds, if so, accept S' as the current optimal solution, set S equal to S', and proceed to step S6); otherwise, accept S' as the current optimal solution with a probability of exp(-Δt / T), set S equal to S', and proceed to step S6); Step S6) add 1 to the value of l to determine whether l is less than N. If yes, proceed to step S4); otherwise, proceed to step S7); Step S7) If the current temperature T is less than the stop temperature T f , go to step S3), otherwise, go to step S8); Step S8) Update T with a temperature strictly less than T, so that T tends to 0; when T is greater than the stop temperature T f When , go to step S3), otherwise, go to step S9); Step S9) outputs S as the final layout plan.

Images