A post-disaster rescue method, system and device using a group robot

Abstract

This invention discloses a method, system, and equipment for disaster relief using swarm robots. The method includes: simulating a disaster relief scenario; determining the hard rules that each robot in the swarm must follow based on its actions and states in the simulated scenario; constructing a functional relationship between the adoption rate of social rules and the performance indicators of the swarm robots; constructing a compromise decision model based on the performance indicators, the social rules, and the functional relationship; solving the compromise decision model using an adaptive linear programming algorithm and performing multi-objective visual trade-off analysis to obtain the range of social rule adoption rates that satisfies all the performance indicators; and directing the swarm robots to perform disaster relief based on the range of social rule adoption rates. This invention can minimize casualties and property damage in disaster relief.

Description

Technical Field

[0001] This invention relates to the field of disaster relief technology, and in particular to a disaster relief method, system and equipment using swarm robots. Background Technology

[0002] In recent years, the use of robots in disaster relief has attracted increasing attention from scholars. Disaster relief, as a complex systems engineering project, involves research in many fields such as dispatch and command, material distribution, and detection and sensing. Due to the unknown and dangerous nature of the post-disaster environment, people have increasingly turned their attention to rescue robots, hoping to use them to complete rescue work in dangerous and unknown environments. In actual disaster relief scenarios, the overall information is unknown, requiring rescue robots to explore the environment ahead, overcome the influence of rugged terrain, and avoid potential obstacles. In such situations, rescue robots need to locate trapped people and deliver relief supplies to designated areas in a timely manner. During this process, robots need to possess certain obstacle avoidance, off-road, and motion planning capabilities, and be able to adjust their movement status in a timely manner according to the dynamically changing environment. Furthermore, for some confined spaces where rescue personnel or large robots cannot enter for exploration, small swarm robots can better leverage their advantages. At the same time, to apply the designed small swarm robots to actual disaster relief problems, it is essential to simultaneously ensure multiple performance characteristics of the small swarm robots, such as reliability, efficiency, and low energy consumption, in order to minimize casualties and property damage. Summary of the Invention

[0003] Therefore, the purpose of this invention is to provide a disaster relief method, system, and equipment that utilizes swarm robots to minimize casualties and property losses during disaster relief.

[0004] To achieve the above objectives, the present invention provides the following solution:

[0005] A disaster relief method utilizing swarm robots includes:

[0006] Simulate post-disaster relief scenarios to obtain simulated scenarios;

[0007] The hard rules that each robot in the swarm of robots must follow are determined based on the actions and states of each robot in the simulation scenario.

[0008] Establish social rules; these social rules are used to avoid conflicts that may occur when the robot complies with the hard rules;

[0009] A functional relationship is established between the adoption rate of social rules and the performance indicators of the swarm robots; the adoption rate of social rules is the probability that the robots will adopt the corresponding social rules after a conflict occurs while adhering to the hard rules; the performance indicators include: the total time, total distance, and success rate of the swarm robots in disaster relief.

[0010] A compromise decision-making model is constructed based on the performance indicators, the social rules, and the functional relationships.

[0011] The compromise decision model is solved using an adaptive linear programming algorithm, and a multi-objective visual trade-off analysis is performed to obtain the range of social rule adoption rates that satisfy all the performance indicators.

[0012] The swarm of robots conducts disaster relief based on the adoption rate range of the social rules.

[0013] Optionally, the construction of social rules specifically includes:

[0014] Determine the types of conflicts between robots in the swarm;

[0015] The actions that the robot needs to perform to avoid a conflict are determined based on the type of conflict.

[0016] Social rules are constructed based on the actions that the robot needs to perform to avoid conflict.

[0017] Optionally, the expression for the Social Rule is as follows:

[0018] Social Rule =<C,DesA,RecA>

[0019] Where C represents the conflict that triggers the social rules, DesA represents the action the robot performs when the conflict occurs, and RecA represents the recommended action the robot should perform to avoid the conflict.

[0020] Optionally, a functional relationship is constructed between the adoption rate of social rules and the performance indicators of the swarm of robots, specifically including:

[0021] Collect sample data of multiple performance indicators under various combinations of social rule adoption rates;

[0022] Based on the sample data, a mapping relationship method based on response surface methodology is used to fit the functional relationship between the adoption rate of the social rules and the performance index.

[0023] Optionally, an adaptive linear programming algorithm is used to solve the compromise decision model, and a multi-objective visual trade-off analysis is performed to obtain the range of social rule adoption rates that satisfy all the performance indicators, specifically including:

[0024] The compromise decision model is solved using an adaptive linear programming algorithm to obtain a single performance index that satisfies the minimum deviation under different weight combinations.

[0025] Draw a ternary plot of a single performance metric that satisfies the minimum deviation, and delineate the acceptable range;

[0026] The overlapping portions of the acceptable ranges in the superimposed ternary plots of all individual performance metrics that satisfy the minimum deviation represent the acceptable ranges of all performance metrics that satisfy the minimum deviation.

[0027] The range of social rule adoption rates that satisfy all performance metrics is determined based on the acceptable range of all performance metrics that satisfy the minimum deviation.

[0028] This invention also provides a disaster relief system using swarm robots, comprising:

[0029] The simulation module is used to simulate post-disaster relief scenarios and obtain the simulation scenario.

[0030] A hard rule construction module is used to determine the hard rules that each robot in the swarm of robots must follow based on the actions and states of each robot in the simulation scenario.

[0031] A social rule construction module is used to construct social rules; the social rules are used to avoid conflicts when the robot complies with the hard rules.

[0032] A function relationship construction module is used to construct a functional relationship between the social rule adoption rate and the performance indicators of the swarm of robots; the social rule adoption rate is the probability that the robot adopts the corresponding social rule after a conflict occurs while adhering to the hard rule; the performance indicators include: the total time, total distance, and success rate of the swarm of robots in disaster relief;

[0033] The compromise decision-making model construction module is used to construct a compromise decision-making model based on the performance indicators, the social rules, and the functional relationships.

[0034] The solution module is used to solve the compromise decision model using an adaptive linear programming algorithm and to perform multi-objective visual trade-off analysis to obtain the range of social rule adoption rates that satisfy all the performance indicators.

[0035] An execution module is used to enable the swarm of robots to conduct disaster relief based on the adoption rate range of the social rules.

[0036] The present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the disaster relief method of the above-mentioned application of swarm robots.

[0037] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described disaster relief method using swarm robots.

[0038] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0039] (1) This invention introduces social rules and social rule adoption rate to resolve conflicts among group robots, which not only characterizes the degree of autonomy of robots from the side, but also realizes the quantification of the degree of autonomy of robots.

[0040] (2) Based on hard rules, this invention initially designs a group robot that can perform tasks. Then, by analyzing conflicts, corresponding social rules are formulated, and an appropriate adoption rate is selected to gradually improve the group robot, so as to carry out disaster relief more quickly and effectively.

[0041] (3) This invention uses a compromise decision model to explore a satisfactory solution within a certain range, which not only solves the problem of coupling of multiple conflicting performance indicators, but also makes the swarm robot more adaptable and robust. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 A flowchart of a disaster relief method using swarm robots provided by the present invention;

[0044] Figure 2 This is a simulation image of a disaster relief scene;

[0045] Figure 3 A diagram showing the area division of the boxes;

[0046] Figure 4 This is a schematic diagram illustrating the conflicting forces acting on the robot as it rotates the box.

[0047] Figure 5 A diagram illustrating how robots can resolve steering conflicts by adhering to social rules governing steering.

[0048] Figure 6 A spectrum diagram illustrating the adoption rate of social rules;

[0049] Figure 7 A flowchart for solving a compromise decision model using an adaptive programming algorithm;

[0050] Figure 8 A ternary graph for objective 1;

[0051] Figure 9 This is a schematic diagram of the acceptable range in the ternary graph for objective 1;

[0052] Figure 10 This is a schematic diagram of the acceptable range in the ternary graph for objective 2;

[0053] Figure 11 This is a schematic diagram of the acceptable range in the ternary graph for objective 3;

[0054] Figure 12 A superimposed diagram of ternary graphs for three objectives. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] The purpose of this invention is to provide a disaster relief method, system, and equipment that utilizes swarm robots to minimize casualties and property damage during disaster relief efforts.

[0057] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0058] Example 1

[0059] like Figure 1 As shown, the disaster relief method using swarm robots provided by this invention includes the following steps:

[0060] Step 101: Simulate the post-disaster relief scenario to obtain the simulation scenario.

[0061] The constructed simulation scenario is as follows Figure 2 As shown, Figure 2The simulation scenario shown is a disaster relief scenario built in Webots software. The robots need to rescue the trapped people. The success of the rescue mission depends on whether the swarm of robots can successfully transport boxes containing supplies to the trapped people. There will be obstacles during the transport, and the boxes need to be turned to pass through smoothly.

[0062] In addition, to simplify the task, the following assumptions are made regarding the post-disaster relief scenario:

[0063] (1) The scene is a bounded area.

[0064] (2) All the people were trapped in the same place.

[0065] (3) The rescue mission is considered complete when the robot transports supplies to the trapped people.

[0066] The first assumption aligns with reality. In actual disaster relief, a building on fire is typically a bounded area, as is a city affected by an earthquake. In this paper's task definition, the disaster relief area is a bounded rectangular region. While the second assumption doesn't perfectly match reality (in reality, trapped people may be scattered in multiple locations), it doesn't significantly impact the experimental results. By adding trapped people from different locations to the initial rescue of a group (one-dimensional), the scope can be easily expanded to rescue groups in multiple locations (multi-dimensional). The second assumption is added simply to simplify the simulation's time requirement. The third assumption also aligns with reality regarding the resources mentioned. For example, in fire rescue, firefighters need to transport fire-fighting robots to the trapped population to extinguish the fire and achieve the rescue mission.

[0067] Step 102: Determine the hard rules that each robot must follow based on its actions and states in the simulation scenario.

[0068] Based on a post-disaster relief simulation scenario, this paper analyzes the performance requirements for swarm robots and preliminarily designs the hard rules that each robot in the swarm must follow to achieve the task. The robots make decisions based on these hard rules, execute corresponding actions, and achieve the swarm robot task through mutual cooperation. These hard rules are the most fundamental rules that ensure the swarm robots can accomplish the task, and the robots must adhere to them.

[0069] Define the hard rules for swarm robots:

[0070] (1) Hard rules refer to the rules that robots must follow when making decisions and taking actions. Only by following all hard rules can a robot complete a task.

[0071] (2) When designing hard rules, analyze the actions that the robot needs to perform to achieve the task, and design hard rules based on actions and states.

[0072] (3) In order to ensure the adaptability and efficiency of the group of robots, the design of hard rules should start from the individual robot. The designed hard rules should ensure that the robot can choose the action that maximizes its own interests when performing the task.

[0073] Take, for example, multiple robots working together to push a box:

[0074] Setting: At least two robots must cooperate to move the box. During the task, the area is divided into 8 zones based on the box's position, including 4 turning zones, such as... Figure 3 As shown.

[0075] Design hard rules: The robot moves to the nearest turning area based on its own position before performing the box-pushing action. The robot makes decisions and acts according to the defined hard rules for turning.

[0076] The algorithm for the hard rules governing the steering of the boxes when the robots are collaboratively pushing them is as follows:

[0077]

[0078] In a swarm of robots, there is no subordinate relationship between the robots; each robot is completely autonomous and has full autonomy, and is not affected by the decisions of other robots.

[0079] In the actual execution of hard rules, due to the complex relationships between robots and the fact that each robot pursues its own interests to maximize its own, conflicts arise among the robots in a swarm. This results in poor performance indicators such as efficiency, success rate, adaptability, energy consumption, and robustness of the swarm robots, making it unable to meet task requirements.

[0080] For example, in the three robot collaborative box-pushing cases mentioned above, they all follow the same hard turning rules, which may result in the robots simultaneously rotating the boxes in two opposing areas. Figure 4 As shown, the robot's forces conflicted. Although the box was eventually able to be turned, the robot's efficiency in completing the task of turning the box was low.

[0081] Step 103: Construct social rules; social rules are used to avoid conflicts that occur when robots follow hard rules.

[0082] Conflicts can occur when robots execute actions according to hard rules, affecting the performance of swarm robots. To improve performance, corresponding social rules are formulated by analyzing the causes of these conflicts to avoid them. The introduction of social rules does not affect the final completion of the swarm robot task; it only affects performance aspects such as efficiency and reliability. Robots can autonomously decide whether to comply with these rules.

[0083] The steps for establishing social rules are as follows:

[0084] Analyze the types of conflicts between robots in a swarm of robots;

[0085] Design the actions that the robot needs to perform to avoid conflict based on the type of conflict;

[0086] Social rules are formulated based on the actions of robots.

[0087] Social rules can be defined as:

[0088] Social Rule =<C,DesA,RecA>

[0089] Where C is the conflict that triggers the social rules, DesA is the action the robot takes when the conflict occurs, and RecA is the action the robot should take to avoid the conflict.

[0090] The rational formulation of social rules can effectively reduce the occurrence of conflicts among swarm robots, thereby improving their performance. Therefore, formulating social rules requires a thorough analysis of the principles underlying swarm robot conflicts and the development of corresponding solutions. Social rules can be formulated based on the actions that robots need to perform to resolve conflicts.

[0091] To address the conflict arising from the robot's collaborative pushing of the box to execute the hard rule of turning in step 102, a social rule of turning is defined:

[0092] turing Social Rule=<C,DesA,RecA> =<Steering conflict; robots pushing boxes in opposing steering areas; fewer robots in the same steering area moving to the adjacent steering area>.

[0093] Decisions that comply with the rules of the transition to a more social society, such as Figure 5As shown, firstly, the robot makes a decision based on hard turning rules, selecting and moving to the nearest turning area; then, the system senses a conflict, triggering social turning rules; finally, the robot re-decides according to social turning rules, resolves the conflict, and successfully turns the box. Although the robot following social turning rules abandons its most self-interested decision, it improves overall efficiency. Therefore, introducing social rules essentially changes the robot's self-interest, correcting its decisions from a global perspective, thereby improving the performance of swarm robots.

[0094] Step 104: Construct a functional relationship between the social rule adoption rate and the performance indicators of the swarm robots; the social rule adoption rate is the probability that a robot will adopt the corresponding social rule after a conflict occurs while adhering to hard rules; the performance indicators include: the total time, total distance traveled, and success rate of the swarm robots in disaster relief. Efficiency, energy consumption, and reliability are represented by total time, total distance traveled, and success rate, respectively.

[0095] Each social rule adoption rate is introduced to characterize the probability that the robot will abide by the social rule when it encounters a conflict corresponding to that social rule, thus reflecting the degree of autonomy of the robot. The functional relationship between multiple social rule adoption rates and performance indicators is obtained through response surface methodology. The specific steps are as follows: 1) Collect sample data of multiple performance indicators under multiple combinations of different social rule adoption rates; 2) Fit the polynomial function relationship between social rule adoption rates and different performance indicators based on the obtained sample data using the response surface methodology.

[0096] The introduction of social rules can alter the performance of swarm robots, but their impact on performance metrics can vary. For example, when measuring swarm robot performance using success rate and efficiency, social rules can be categorized as follows:

[0097] 1) Social rules that simultaneously improve efficiency and success rate.

[0098] 2) Social rules that prioritize efficiency at the expense of success rates;

[0099] 3) Social rules that prioritize increasing success rates at the expense of efficiency;

[0100] When formulating social rules by analyzing conflicts among swarm robots, the ideal situation is to simultaneously improve the efficiency and success rate of the swarm robots. In this case, the robots should follow these social rules with 100% probability to maximize system performance. However, how the robots follow the latter two types of social rules needs to be explored.

[0101] In swarm robotics, autonomy is considered one of the most fundamental attributes that define a robot, representing the degree to which a robot is not interfered by other external factors during decision-making and action. Autonomy is closely related to whether a robot adheres to social rules. When the robots in swarm robotics only abide by legal rules and not any social rules, the robots are completely autonomous, and the autonomy value is 1. They are not affected by the decisions of any other robots and make decisions and act completely independently. At this time, the swarm robot has a distributed organizational structure. When the robots follow all social rules, they consider the impacts brought by the decisions of all other robots. At this time, the robots have no autonomy, and the autonomy value is 0. At this time, the swarm robot is equivalent to a centralized distribution structure. Related research has shown that during the process of task execution, an autonomous decision-making framework based on dynamic adjustment can perform better in task realization. Therefore, inspired by dynamic autonomy, this invention uses the social rule adoption rate to represent autonomy, enabling the robot to select the optimal social rule adoption rate for different social rules, thereby achieving the dynamic change of the robot's autonomy and optimizing the performance of the swarm robot.

[0102] Figure 6 Figure 4 shows the spectrum of the social rule adoption rate, where the social rule adoption rate of the robot increases from left to right, while the autonomy decreases as the rule adoption rate increases. Similarly, the social rule adoption rate (R) is represented by a percentage between 0% and 100% and is divided into the following three categories:

[0103] 1) R = 0%: Selfish, individualistic;

[0104] 2) 0% < R < 100%: Partially autonomous, partially compromising;

[0105] 3) R = 100%: Collectivistic.

[0106] During the entire process of task implementation, when the social rule adoption rates formulated for different conflicts are different, the autonomy of the robot will also change accordingly. Therefore, the adaptive social rule adoption rate and the dynamically adjustable autonomy have the same underlying meaning. However, current research on dynamically adjustable autonomy is limited to the switching of the autonomy value between 0 and 1, and there is still a lack of corresponding research on how to quantify the magnitude of autonomy and truly achieve the dynamic change of autonomy. This invention uses the social rule adoption rate to represent the autonomy of the robot from the side and truly realizes the quantification of autonomy.

[0107] To explore the relationship between the adoption rate of social rules and the performance indicators of swarm robots, and to optimize the performance of swarm robots, this invention employs a mapping method based on response surface methodology (RSM) to fit the mapping relationship between the performance indicators of swarm robots and the adoption rate of social rules. RSM, proposed by Box et al., is an experimental design method that integrates experimental design and mathematical modeling for optimization. It involves conducting experiments at representative local points, regressing and fitting the functional relationship between factors and outcomes over a global scope, and obtaining the optimal level values ​​for each factor. RSM offers advantages such as fewer experiments, shorter experimental cycles, high precision, high accuracy in obtaining regression equations, good predictive performance, and the ability to simultaneously study the interactions between several factors.

[0108] The system response y and the design variable (i.e., the adoption rate of social rules) x can be expressed as the sum of an approximate function and the total error, that is:

[0109]

[0110] In the formula: It is an approximation function of the unknown function, x[x1, x2, ..., x]. n ] represents n-dimensional independent design variables; δ represents the total error, which includes random error, modeling error, and systematic error.

[0111] The approximate function can be represented by a polynomial response surface, i.e.:

[0112]

[0113] In the formula: β is the unknown coefficient; k is the number of design variables; Indicates the predicted response values: β0, β i β ii These are the offset term, linear offset, and second-order offset coefficients, respectively; β ij It is the interaction coefficient.

[0114] By combining equation (1) with equation (2) and performing a series of transformations, the coefficient matrix of the polynomial fitting function that best approximates the actual function can be obtained.

[0115]

[0116] in:

[0117]

[0118] The fitted response surface expression is then:

[0119]

[0120] Response surface methodology (RSM) can effectively connect design objectives (i.e., performance indicators) and design parameters (i.e., social rule adoption rates), providing specific expressions. By substituting the adoption rates of each social rule as X and the three design objectives as responses y into the response surface model, the corresponding parameter mapping relationship fitted using RSM is obtained.

[0121] Step 105: Construct a compromise decision model based on performance indicators, social rules, and functional relationships.

[0122] The compromise decision model, proposed by Mistree et al., is a decision support problem model based on robust design. It is a mathematical model that formally expresses the multi-objective compromise decision problem in multidisciplinary product optimization design. Designers often encounter decision problems involving multiple conflicting objectives. When determining the values ​​of design parameters, compromises must be made between the design objectives of multiple disciplines to obtain parameter values ​​that satisfy these objectives as much as possible. Therefore, the compromise decision model is actually a special type of choice decision. As a mathematical model for balancing multiple conflicting objectives, it solves the problem of searching the parameter space for parameter combinations that best satisfy the objective weight preferences, given the weight preferences of each objective. The mathematical framework of the compromise decision problem model is shown below:

[0123]

[0124] Step 106: Solve the compromise decision model using an adaptive linear programming algorithm and perform multi-objective visual trade-off analysis to obtain the social rule adoption rate range that satisfies all performance indicators. Specifically, this includes: solving the compromise decision model using an adaptive linear programming algorithm to obtain the individual performance indicators that satisfy the minimum deviation under different weight combinations; drawing a ternary graph of the individual performance indicators that satisfy the minimum deviation and defining the acceptable range; overlaying the ternary graphs of all individual performance indicators that satisfy the minimum deviation, with the overlapping portion of the acceptable range in the overlaid ternary graphs representing the acceptable range of all performance indicators that satisfy the minimum deviation; and determining the social rule adoption rate range that satisfies all performance indicators based on the acceptable range of all performance indicators that satisfy the minimum deviation.

[0125] In swarm robotics, multiple performance objectives need to be met simultaneously, and different tasks place different demands on the swarm robots' performance metrics. Therefore, when designing swarm robots, a trade-off between these performance objectives should be achieved as much as possible. Simultaneously, swarm robots need to possess adaptability and robustness to change. Therefore, this invention chooses to use a compromise decision model optimization method for design optimization. The process of optimizing multiple performance metrics using the compromise decision model method is summarized in the following six steps:

[0126] Step 1: Analyze the swarm of robots to determine parameters, boundary constraints, and design requirements;

[0127] Step 2: Determine the design variables (social rule adoption rate) and multiple system performance objectives (SOS performance indicators) that need to be met for the swarm robot and their deviation variables;

[0128] Step 3: Determine appropriate performance target values;

[0129] Step 4: Construct the objective function relation with deviation variables;

[0130] Step 5: Solve for the target value and design parameters that minimize deviation using an adaptive programming algorithm; the process of solving the compromise decision model using an adaptive programming algorithm is as follows: Figure 7 As shown;

[0131] Step 6: Visualize the trade-offs to obtain the range of design variables (social rule adoption rate) that simultaneously meet the design objectives (performance indicators).

[0132] Taking total time, total distance, and success rate as examples representing efficiency, energy consumption, and reliability respectively, a ternary graph is drawn for each objective based on the data obtained from the adaptive programming algorithm. The ternary graph for objective 1 (i.e., total time) is shown below. Figure 8 As shown. The acceptable range for the ternary graph of each target is divided, such as... Figures 9-11 As shown. By overlaying the ternary diagrams of the three objectives, the overlapping portion of the acceptable range is the design range that simultaneously satisfies all three objectives. The overlay diagram of the ternary diagrams of the three objectives is shown below. Figure 12 As shown, the overlapping gray areas represent regions that simultaneously satisfy all three objectives.

[0133] Step 107: Based on the adoption rate range of social rules, the swarm of robots conducts disaster relief.

[0134] Example 2

[0135] In order to execute the method corresponding to Embodiment 1 above and achieve the corresponding functions and technical effects, a disaster relief system using swarm robots is provided below, including:

[0136] The simulation module is used to simulate post-disaster relief scenarios and obtain the simulation scenario.

[0137] The hard rule construction module is used to determine the hard rules that each robot in the swarm of robots must follow based on the actions and states of each robot in the simulation scenario.

[0138] The social rules building module is used to construct social rules; social rules are used to avoid conflicts that occur when robots adhere to hard rules.

[0139] The function relationship construction module is used to construct the functional relationship between the social rule adoption rate and the performance indicators of the swarm robots; the social rule adoption rate is the probability that the robot will adopt the corresponding social rule after a conflict occurs when it is following the hard rules; the performance indicators include: the total time, total distance and success rate of the swarm robots in disaster relief;

[0140] The compromise decision-making model building module is used to construct compromise decision-making models based on performance indicators, social rules, and functional relationships.

[0141] The solution module is used to solve the compromise decision model using an adaptive linear programming algorithm to obtain the range of social rule adoption rates that satisfy all performance indicators;

[0142] The execution module is used to enable swarm robots to conduct disaster relief based on the adoption rate range of social rules.

[0143] Example 3

[0144] Embodiment 3 of the present invention provides an electronic device, including a memory and a processor. The memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the disaster relief method using swarm robots as described in Embodiment 1.

[0145] The aforementioned electronic device may be a server.

[0146] Example 4

[0147] Embodiment 4 of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the disaster relief method using swarm robots as described in Embodiment 1.

[0148] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0149] This article uses specific examples to illustrate the principles and implementation methods of the invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. The described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

Claims

1. A disaster relief method using swarm robots, characterized in that, include: Simulate post-disaster relief scenarios to obtain simulated scenarios; The hard rules that each robot in the swarm of robots must follow are determined based on the actions and states of each robot in the simulation scenario. Constructing social rules; specifically including: determining the types of conflicts between robots in the swarm; determining the actions that the robots need to perform to avoid conflicts based on the conflict types; constructing social rules based on the actions that the robots need to perform to avoid conflicts; the social rules are used to avoid conflicts that occur when the robots comply with the hard rules; A functional relationship is established between the adoption rate of social rules and the performance indicators of the swarm robots; the adoption rate of social rules is the probability that the robots will adopt the corresponding social rules after a conflict occurs while adhering to the hard rules; the performance indicators include: the total time, total distance, and success rate of the swarm robots in disaster relief. A compromise decision-making model is constructed based on the performance indicators, the social rules, and the functional relationships. The compromise decision model is solved using an adaptive linear programming algorithm, and a multi-objective visual trade-off analysis is performed to obtain the range of social rule adoption rates that satisfy all the performance indicators. Based on the adoption rate range of the aforementioned social rules, the swarm of robots is instructed to conduct disaster relief.

2. The disaster relief method using swarm robots according to claim 1, characterized in that, The social rules The expression is as follows: Where C represents the conflict that triggers the social rules, DesA represents the action the robot performs when the conflict occurs, and RecA represents the recommended action the robot should perform to avoid the conflict.

3. The disaster relief method using swarm robots according to claim 1, characterized in that, Constructing a functional relationship between the adoption rate of social rules and the performance indicators of the swarm of robots, specifically including: Collect sample data of multiple performance indicators under various combinations of social rule adoption rates; Based on the sample data, a mapping relationship method based on response surface methodology is used to fit the functional relationship between the adoption rate of the social rules and the performance index.

4. The disaster relief method using swarm robots according to claim 1, characterized in that, An adaptive linear programming algorithm is used to solve the compromise decision model, and a multi-objective visual trade-off analysis is performed to obtain the range of social rule adoption rates that satisfy all the performance indicators, specifically including: The compromise decision model is solved using an adaptive linear programming algorithm to obtain a single performance index that satisfies the minimum deviation under different weight combinations. Draw a ternary plot of a single performance metric that satisfies the minimum deviation, and delineate the acceptable range; The overlapping portions of the acceptable ranges in the superimposed ternary plots of all individual performance metrics that satisfy the minimum deviation represent the acceptable ranges of all performance metrics that satisfy the minimum deviation. The range of social rule adoption rates that satisfy all performance metrics is determined based on the acceptable range of all performance metrics that satisfy the minimum deviation.

5. A disaster relief system utilizing swarm robots, characterized in that, include: The simulation module is used to simulate post-disaster relief scenarios and obtain the simulation scenario. A hard rule construction module is used to determine the hard rules that each robot in the swarm of robots must follow based on the actions and states of each robot in the simulation scenario. A social rule construction module is used to construct social rules; specifically, it includes: determining the conflict type among robots in the swarm; determining the actions that the robots need to perform to avoid conflict based on the conflict type; constructing social rules based on the actions that the robots need to perform to avoid conflict; the social rules are used to avoid conflicts that occur when the robots comply with the hard rules; A function relationship construction module is used to construct a functional relationship between the social rule adoption rate and the performance indicators of the swarm of robots; the social rule adoption rate is the probability that the robot adopts the corresponding social rule after a conflict occurs while adhering to the hard rule; the performance indicators include: the total time, total distance, and success rate of the swarm of robots in disaster relief; The compromise decision-making model construction module is used to construct a compromise decision-making model based on the performance indicators, the social rules, and the functional relationships. The solution module is used to solve the compromise decision model using an adaptive linear programming algorithm and to perform multi-objective visual trade-off analysis to obtain the range of social rule adoption rates that satisfy all the performance indicators. An execution module is used to enable the swarm of robots to conduct disaster relief based on the adoption rate range of the social rules.

6. An electronic device, characterized in that, It includes a memory and a processor, the memory being used to store a computer program, and the processor running the computer program to cause the electronic device to perform the disaster relief method using swarm robots as described in any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the disaster relief method using swarm robots as described in any one of claims 1-4.

Images