Urban emergency management rescue method assisted by low-altitude flight platform

Abstract

The invention discloses an urban emergency management rescue method assisted by a low-altitude flight platform, and particularly relates to the technical field of traffic control. Dynamically generating an optimal flight path and realizing real-time adjustment of the path by combining dynamic path planning and a reinforcement learning algorithm; task priority dynamic features and airspace risk features are analyzed, task allocation, priority ranking and resource dynamic adjustment are optimized through a task allocation algorithm, and multi-platform efficient cooperation is achieved; the flight platform executes rescue operation according to task allocation and feeds back task states and environment changes in real time; on the basis of advanced prediction of scheduling capability, a redundancy control strategy and an emergency response mechanism are combined to automatically adjust the operation state of the platform so as to avoid risks, so that the defects of real-time sensing, path planning, multi-platform cooperation and communication guarantee in a dynamic complex urban airspace in the prior art are effectively overcome; and the efficiency, the safety and the adaptive capacity of the low-altitude flight platform in emergency rescue are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic control, and particularly to an urban emergency management and rescue method assisted by a low-altitude flight platform. Background Art

[0002] Urban emergency management and rescue assisted by a low-altitude flight platform refers to the technology and methods of using low-altitude flight platforms (such as drones, helicopters, small aircraft, etc.) to provide auxiliary support in urban emergencies (such as natural disasters, fires, traffic accidents, medical emergencies, etc.); these platforms usually have the characteristics of rapid response, high mobility, and real-time information collection, and can provide efficient emergency services in cases where ground traffic is blocked or the situation is complex; for example, in a fire in a high-rise building in the city, the fire spreads quickly and is at a high position, and traditional fire-fighting equipment may be difficult to quickly control the fire; a low-altitude flight platform can carry a thermal imager to monitor the location and temperature distribution of the fire source, assisting firefighters in formulating a precise fire-fighting plan; some drones can also carry fire-fighting devices or water spraying systems and directly participate in the fire-fighting operation, especially suitable for areas that are difficult to access, such as the top of high-rise buildings or narrow alleys.

[0003] The existing technologies have the following deficiencies: In a complex urban environment, when existing low-altitude flight platforms are used for emergency rescue, there is a lack of a real-time perception, path planning, and obstacle avoidance coordination mechanism for dynamic high-density urban traffic scenarios. Especially in severe emergencies such as high-rise building fires or earthquakes that cause urban communication interruptions, existing platforms are difficult to achieve cooperation among multiple flight platforms in a dynamic and complex urban airspace and cannot complete rescue tasks efficiently and safely; this limitation leads to the low-altitude flight platform being prone to affecting the overall rescue efficiency due to flight conflicts, delays, or uneven resource allocation in an emergency, and may even cause secondary accidents. Summary of the Invention

[0004] The purpose of the present invention is to provide an urban emergency management and rescue method assisted by a low-altitude flight platform to solve the deficiencies in the background art.

[0005] To achieve the above purpose, the present invention provides the following technical solutions: An urban emergency management and rescue method assisted by a low-altitude flight platform, comprising the following steps: S1: Using multi-sensor fusion technology to perform real-time perception of a complex urban environment. The sensors include lidar, cameras, ultrasonic sensors, and inertial measurement units, and are used to collect three-dimensional environmental data, dynamic obstacle information, and flight platform operation status data of the target area; S2: Implementing real-time communication and data sharing among low-altitude flight platforms through distributed self-organizing network technology. In the case of partial or complete interruption of the urban communication network, a relay node mechanism is adopted to ensure the efficient transmission of information among platforms; S3: Based on the real-time perceived environmental data, use the dynamic path planning algorithm to generate the optimal flight path, and combine the reinforcement learning algorithm to dynamically adjust the path during flight to avoid obstacles; S4: Analyze the dynamic characteristics of the priority of emergency tasks and the airspace risk characteristics, evaluate the emergency task scheduling ability of the low-altitude flight platform, and according to the evaluation results, re-schedule multiple flight platforms through the task allocation algorithm. The re-scheduling includes task allocation, priority sorting, and dynamic adjustment of task resources; S5: The flight platform performs rescue operations according to the assigned tasks, including thermal imaging monitoring, material delivery, fire extinguishing operations, and positioning of trapped personnel, and feeds back the task execution status and environmental change information to the command center through the real-time communication network; S6: Predict in advance the abnormal situations of the emergency task scheduling ability of the low-altitude flight platform within the subsequent fixed time period, and before the occurrence of abnormal events, automatically adjust the platform operation status based on the preset redundant control strategy and emergency response mechanism to avoid risks.

[0006] Preferably, in S4, after analyzing the dynamic characteristics of the priority of emergency tasks, generate a task completion rate deviation index. The method for obtaining the task completion rate deviation index is as follows: Define as the actual task completion rate at time t, and define as the expected task completion rate at time t; Determine the time window size n, and assign weights , and normalize the sum of weights: ; Calculate the weighted moving average of the actual task completion rate , and the expression is: ; Use and the expected completion rate [[ID=I27]]to calculate the task completion rate deviation index, and the expression is: ; In the formula, FCHK is the task completion rate deviation index.

[0007] Preferably, in S4, after analyzing the airspace risk characteristics of emergency tasks, generate a flight conflict rate anomaly index. The method for obtaining the flight conflict rate anomaly index is as follows: Define the nodes of the graph structure. Each node represents a low-altitude flight platform. The node characteristics include flight status and airspace risk data. The edges represent the relationship between flight platforms, and the weights of the edges reflect the conflict risk. The initial characteristics of each node are vectors , and the characteristics of each edge are vectors . In each round of propagation, each node receives information from its neighbor nodes and updates its own characteristics: ; In the formula, is the information from node j to node i at the l-th layer, Generate a function for messages to summarize neighbor information: ; where is the sum of the information of all neighbor nodes received by node i, is the neighbor set of node i, and update the node features using the received messages: ; where is the feature update function; the node features of the last layer are used to predict the conflict probability of each node: ; is the conflict probability of node i, and σ is the activation function; calculate the expected conflict probability of each node based on historical data ; for each node, calculate the deviation between the actual and expected conflict probabilities , and the expression is: ; summarize the deviations of all nodes and calculate the Flight Conflict Rate Abnormality Index FCRAI, and the expression is: ; where N is the total number of nodes in the graph, is the Flight Conflict Rate Abnormality Index.

[0008] Preferably, in S4, convert the task completion rate deviation index and the flight conflict rate abnormality index into a comprehensive feature vector, use the comprehensive feature vector as the input of the machine learning model, and use the machine learning model to predict the emergency task scheduling ability value label of the low-altitude flight platform with each group of comprehensive feature vectors as the prediction target, and use minimizing the sum of the prediction errors of the emergency task scheduling ability value labels of all low-altitude flight platforms as the training target to train the machine learning model until the sum of the prediction errors reaches convergence and then stop the model training, and determine the emergency task scheduling ability value of the low-altitude flight platform according to the model output result, where the machine learning model is a polynomial regression model.

[0009] Preferably, compare the obtained emergency task scheduling ability value of the low-altitude flight platform with the pre-set emergency task scheduling ability reference threshold according to historical data. If the emergency task scheduling ability value of the low-altitude flight platform is greater than or equal to the pre-set emergency task scheduling ability reference threshold, it indicates that the emergency task scheduling ability of the low-altitude flight platform is high, and at this time, generate a high task scheduling ability signal and no adjustment is required; if the emergency task scheduling ability value of the low-altitude flight platform is less than the pre-set emergency task scheduling ability reference threshold, it indicates that the emergency task scheduling ability of the low-altitude flight platform is low, and at this time, generate a low task scheduling ability signal and scheduling adjustment is required.

[0010] Preferably, in S6, the abnormal situation of the emergency mission scheduling ability of the low-altitude flight platform within a subsequent fixed time period is predicted in advance, specifically: collecting the emergency mission scheduling ability values of the low-altitude flight platform that are less than the pre-set reference threshold of the emergency mission scheduling ability generated within the subsequent fixed time period, and establishing a corresponding data set, calculating the mean and standard deviation of the data set, and after analyzing it, predicting in advance the abnormal situation of the emergency mission scheduling ability of the low-altitude flight platform within the subsequent fixed time period according to the analysis result.

[0011] Preferably, if the mean value of the abnormal coefficient in the data set is greater than or equal to the reference threshold of the mean value of the abnormal coefficient, and the standard deviation of the abnormal coefficient is less than the reference threshold of the standard deviation of the abnormal coefficient, the overall level of the abnormal scheduling ability value is low and the distribution is concentrated, indicating that there are obvious scheduling ability problems in multiple platforms. At this time, a first-level warning signal is generated, and measures need to be taken immediately to reallocate tasks and adjust priorities; If the mean value of the abnormal coefficient is greater than or equal to the reference threshold of the mean value of the abnormal coefficient, and the standard deviation of the abnormal coefficient is greater than or equal to the reference threshold of the standard deviation of the abnormal coefficient, the overall level of the abnormal scheduling ability value is low and the distribution fluctuates greatly, indicating that there are serious problems in some platforms. At this time, a second-level warning signal is generated, and the reasons for the abnormal platforms need to be analyzed one by one; If the mean value of the abnormal coefficient is less than the reference threshold of the mean value of the abnormal coefficient, and the standard deviation of the abnormal coefficient is greater than or equal to the reference threshold of the standard deviation of the abnormal coefficient, it indicates that the instability of the platform scheduling performance increases. At this time, a third-level warning signal is generated, and the platform status needs to be monitored and analyzed; If the mean value of the abnormal coefficient is less than the reference threshold of the mean value of the abnormal coefficient, and the standard deviation of the abnormal coefficient is less than the reference threshold of the standard deviation of the abnormal coefficient, the overall level and distribution of the abnormal scheduling ability value are within the normal range, and no adjustment measures need to be taken.

[0012] In the above technical solution, the technical effects and advantages provided by the present invention are as follows: 1. By combining multi-sensor fusion, distributed self-organizing network technology, dynamic path planning, reinforcement learning, and task allocation algorithms, the present invention realizes real-time perception, efficient communication, and collaborative scheduling of low-altitude flight platforms in complex urban environments; by generating the task completion rate deviation index and flight conflict rate anomaly index, using the polynomial regression model to accurately predict the emergency mission scheduling ability of the platform, and based on the redundant control and warning mechanism to avoid abnormal events in advance, significantly improving the task completion efficiency and operation safety of low-altitude flight platforms in emergencies; especially in extreme situations such as urban communication interruption or high-density airspace, the present invention can dynamically optimize resource allocation and path planning to ensure the smooth completion of emergency missions.

[0013] 2. Through the real-time analysis and hierarchical early warning (Level 1, Level 2, Level 3) of abnormal scheduling capabilities, the present invention can accurately identify potential problems with the platform's scheduling capabilities and formulate targeted optimization strategies according to the severity of the problems, such as task reallocation, priority adjustment, or real-time monitoring, etc. While ensuring the efficient execution of rescue tasks, the overall system minimizes the risks of flight conflicts and resource waste, solves the deficiencies of the prior art in dynamic collaboration and path optimization, and has significant safety, reliability, and application value, providing innovative technical support for urban emergency management. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0015] Figure 1 It is a flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of 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 of the present invention without creative efforts fall within the scope of protection of the present invention.

[0017] Embodiment, please refer to Figure 1 As shown, the method for urban emergency management and rescue assisted by a low-altitude flight platform in this embodiment includes the following steps: S1: Use multi-sensor fusion technology to perform real-time perception of the complex urban environment. The sensors include lidar, cameras, ultrasonic sensors, and inertial measurement units, which are used to collect three-dimensional environmental data, dynamic obstacle information, and flight platform operation status data of the target area; S2: Implement real-time communication and data sharing between low-altitude flight platforms through distributed self-organizing network technology. In the case where part or all of the urban communication network is interrupted, the relay node mechanism is adopted to ensure the efficient transmission of information between platforms; S3: Based on the real-time perceived environmental data, use a dynamic path planning algorithm to generate the optimal flight path, and combine a reinforcement learning algorithm to dynamically adjust the path during flight to avoid obstacles; S4: Analyze the dynamic characteristics of the priority and airspace risk characteristics of emergency tasks, evaluate the emergency task scheduling ability of the low-altitude flight platform. According to the evaluation results, re-schedule multiple flight platforms through a task allocation algorithm. The re-scheduling includes task allocation, priority ranking, and dynamic adjustment of task resources; S5: The flight platform performs rescue operations according to the assigned tasks, including thermal imaging monitoring, material delivery, fire extinguishing operations, and positioning of trapped personnel, and feeds back the task execution status and environmental change information to the command center through a real-time communication network; S6: Predict in advance the abnormal situations of the emergency task scheduling ability of the low-altitude flight platform within subsequent fixed time periods, and before abnormal events occur, automatically adjust the platform operation status based on preset redundant control strategies and emergency response mechanisms to avoid risks.

[0018] In S1, the multi-sensor hardware configuration includes: Equip with a high-precision lidar for detecting three-dimensional space information in the environment and obtaining the distance and shape of surrounding obstacles; Configuration parameters: scanning frequency, detection range, resolution; Install a high-definition camera to obtain real-time visual data of the target area and identify dynamic targets (such as vehicles, crowds); Configuration parameters: frame rate, resolution, viewing angle range; Arrange multiple ultrasonic sensors around the flight platform to detect close-range obstacles and make up for the perception deficiency of lidar and camera in short-distance blind areas; Configuration parameters: detection angle, detection distance range; Integrate an IMU module to record the attitude, acceleration, and angular velocity information of the flight platform in real time to ensure that the perception system adapts to the dynamic flight state; Configuration parameters: sampling rate, acceleration range, angular velocity range.

[0019] The lidar scans the environment and generates three-dimensional point cloud data representing the spatial distribution of obstacles; The camera captures video or image frames and extracts visual feature information such as color, texture, and shape; The ultrasonic sensor continuously detects surrounding close-range obstacles and generates distance information; The IMU real-time collects the attitude and motion state of the flight platform and provides a position and direction reference.

[0020] Align the output data of the lidar, camera, ultrasonic sensor, and IMU to the same timestamp to ensure real-time performance and data consistency; Based on the installation positions and viewing angles of the sensors, determine the relative positions and direction relationships of the sensor data in a unified coordinate system through calibration.

[0021] In S2, each low-altitude flight platform serves as a network node, equipped with a self-organizing network communication module that supports wireless communication protocols (such as Wi-Fi, ZigBee, LoRa, or dedicated wireless links). The node has the ability to form a network autonomously without relying on fixed infrastructure (such as base stations or routers). The network topology (such as star, mesh, or hybrid topology) is adjusted in real time based on the node location, communication link status, and mission requirements. When nodes are added, removed, or moved, the self-organizing network dynamically updates the path to ensure network stability and efficiency.

[0022] Basic data includes the location information, speed, attitude and perceived environment data (such as obstacle location) of the flight platform shared by each node; advanced data includes mission status, emergency instructions, dynamic path planning results and mission collaboration information.

[0023] Assign priorities to different types of data: High priority: real-time location and obstacle avoidance information, used to ensure flight safety; Medium priority: mission-related information (such as target location and status), used for collaborative scheduling; Low priority: historical data or statistical information of environmental perception, used for optimization strategy.

[0024] Use distributed time synchronization algorithms (such as NTP or IEEE 1588 protocol) to ensure consistent data timestamps; use consistency protocols (such as RAFT or Paxos) to resolve data conflicts and ensure the consistency of shared data.

[0025] When the communication link is partially interrupted or the flight platforms are distributed over a large area, the flight platform with the best location and signal quality is automatically selected as the relay node; the relay node can be switched dynamically, and the selection criteria include: maximizing the signal coverage range; sufficient remaining energy; and strong hardware processing capabilities.

[0026] The functions of relay nodes include: receiving and forwarding communication data from other platforms to extend the communication range; aggregating distributed data and forwarding it to task nodes to reduce the redundancy of multi-path communication; and using relay nodes to maintain communication stability when the direct communication link fails.

[0027] Use multi-hop routing protocols (such as AODV or DSR) to establish optimal communication paths between nodes, reduce latency and transmission energy consumption; monitor link quality in real time and dynamically adjust routing strategies.

[0028] When a city's communication network is completely interrupted (such as due to an earthquake or infrastructure damage), the flight platforms rely on self-organizing networks to maintain internal communications; relay nodes build a virtual backbone network to connect the flight platforms in various dispersed areas into a whole.

[0029] In the case of partial communication interruption (such as poor signal coverage in a certain area), combined with external available networks (such as satellite communication or restored cellular networks), the relay node acts as a bridge to ensure global information sharing; dynamically allocate communication bandwidth, and preferentially ensure the transmission of high-priority data according to task requirements; optimize communication power based on the remaining battery power and distance of the node to extend the working time of the flying platform; adopt spectrum allocation and interference cancellation technologies (such as frequency hopping technology or beamforming) to improve communication stability.

[0030] For example, when the communication infrastructure is damaged, the flying platform shares information such as the disaster area map and the locations of trapped people through self-organizing networks and relay nodes to ensure the coordination of rescue operations; when the flying platform cannot communicate directly due to occlusion or long distance, the relay node bridges the signal and transmits the status information of the fire detection and extinguishing device; the flying platform shares dynamic traffic information in real time through self-organizing networks and collaborates to complete traffic guidance and rescue task planning.

[0031] S3: Based on the real-time sensed environmental data, use a dynamic path planning algorithm to generate the optimal flight path, and combine a reinforcement learning algorithm to dynamically adjust the path during flight to avoid obstacles.

[0032] Use multi-sensor fusion technology (from S1) to obtain real-time environmental perception data, including: three-dimensional environmental models (such as point cloud data); dynamic obstacle information (such as the movement trajectories of vehicles and crowds); the status information of the flying platform (such as position, speed, direction).

[0033] Convert the real-time sensed data into an available input format: Generate a grid map: Divide the environment into passable and impassable areas; Construct a topological map: Represent the possible path options of the flying platform with nodes and edges; Extract the motion characteristics of dynamic obstacles for predicting their future positions.

[0034] Ensure that the path of the flying platform from the starting point to the target point meets the following conditions: Safety: Avoid all static and dynamic obstacles; Efficiency: The path is the shortest or the flight time is the least; Adaptability: Adapt to real-time environmental changes.

[0035] Path planning algorithms include: The A* algorithm uses heuristic search to generate an initial optimal path from the starting point to the end point; The heuristic function is adjusted according to flight costs (such as distance, energy consumption) and obstacle avoidance requirements; In some complex environments, the Dijkstra algorithm is combined with the Dijkstra algorithm to handle path planning in high-density obstacle areas; The Rapidly-Exploring Random Tree (RRT) algorithm is used to find a feasible path in high-dynamic scenarios and avoid getting stuck in local optimal solutions.

[0036] When real-time environmental data changes (such as the addition of obstacles), trigger path replanning; Use incremental algorithms to only recalculate the affected local paths to improve computational efficiency.

[0037] The design of the reinforcement learning model includes: State: The position and speed of the current flight platform, as well as the surrounding environment data (such as the distance and direction of obstacles); Action: The next movement direction or path selection of the flight platform; Reward: Positive rewards: Selecting a safe and fast path; Successfully avoiding obstacles; Penalties: Long paths, approaching obstacles or collisions; Algorithms include: Deep Q-Network (DQN): Using a deep learning model to approximate the Q-value function to handle path optimization problems in a continuous state space; Policy Gradient algorithm: Directly optimizing the action selection strategy of the flight platform to make path adjustment smoother and more adaptable to dynamic environments.

[0038] Through multiple simulations and tests in actual flights, continuously optimize the reinforcement learning model to enable it to have fast decision-making capabilities in complex dynamic environments; Based on dynamic obstacle information (from S1), predict the future positions of obstacles to form a dynamic safety area; If the current path may conflict with obstacles: Short-term adjustment: Adjust the flight direction through the reinforcement learning model to bypass obstacles; Long-term optimization: Re-calculate the global optimal path in combination with the path planning algorithm.

[0039] Use the operating state data of the flight platform (such as the attitude information provided by the IMU) to monitor and feedback the execution of path adjustment in real time; When the path deviates or the execution is abnormal, trigger a correction mechanism to ensure flight safety.

[0040] S4: Analyze the dynamic characteristics of the priority and airspace risk characteristics of emergency tasks, evaluate the emergency task scheduling ability of low-altitude flight platforms, and according to the evaluation results, re-schedule multiple flight platforms through a task allocation algorithm. The re-scheduling includes task allocation, priority ranking, and dynamic adjustment of task resources.

[0041] The dynamic characteristics of priority reflect the changes in the importance and urgency of emergency tasks. The influencing factors include: Task urgency: The quantification of the time sensitivity of the task. For example, the time requirement for a fire extinguishing task is more urgent than that of traffic control; Task importance: The impact of the task on the overall rescue operation. For example, the importance of delivering rescue medical supplies may be higher than that of environmental survey; Resource allocation constraints: Constraints on task execution such as the number of platforms and energy consumption limits.

[0042] The emergency mission scheduling ability of a low-altitude flight platform is reflected in the ability to adjust task priorities in real time and allocate resources reasonably; Task sorting mechanism: Use a multi-attribute decision-making model (such as the weighted scoring method) to calculate task priorities; Dynamic adjustment mechanism: When a new high-priority task appears, the system re-sorts and adjusts the scheduling plan in real time; Task allocation strategy: The platform preferentially allocates high-priority tasks according to the remaining energy, current location, and task requirements; Adopt task allocation algorithms (such as the auction algorithm or the minimum cost allocation algorithm) to optimize resource utilization.

[0043] Scheduling efficiency metrics include: Task completion rate: The proportion of high-priority tasks completed within the specified time; Response time: The average time from when the platform receives a task to when it starts executing the task; Resource utilization rate: The effectiveness of resource (such as energy, communication bandwidth) utilization.

[0044] The airspace risk characteristics describe the potential risk factors faced by the flight platform in the urban airspace, including: Building density: High building density areas increase the complexity of path planning and enhance the collision risk; Obstacle dynamics: Moving obstacles (such as vehicles, crowds) increase the difficulty of obstacle avoidance; Communication interruption risk: The urban communication network is affected by disasters or signal blockages, resulting in limited information sharing among platforms; Wind field disturbance: The airflow changes between high-rise buildings may affect the stable flight of the platform.

[0045] When scheduling, the platform needs to optimize the task path and allocation in combination with the airspace risk characteristics: In high-risk areas, preferentially adopt avoidance strategies or multi-platform collaboration to complete tasks; In high-risk areas, use more conservative path planning algorithms (such as RRT*) to increase the safety margin; In low-risk areas, use more efficient algorithms (such as A*) to improve scheduling efficiency; In high-risk airspace, limit the number of simultaneously operating flight platforms to reduce the risk of airspace conflicts; In low-risk airspace, increase the number of platforms to accelerate task execution.

[0046] Scheduling safety metrics include: Task success rate: The completion situation of tasks in high-risk airspace; Flight conflict rate: The probability of conflicts between platforms or between platforms and obstacles; Task interruption rate: The number or proportion of task interruptions caused by risks.

[0047] After analyzing the dynamic characteristics of the priorities of emergency tasks, a task completion rate deviation index is generated. The method for obtaining the task completion rate deviation index is as follows: Define as the actual task completion rate at time t (percentage, ranging from 0 to 100); Define as the expected task completion rate at time t (percentage, predicted from the target task model or historical data); Determine the time window size n, for example, the past 5 time points; Allocate weights , and normalize the sum of weights: ; Calculate the weighted moving average of the actual task completion rate , the expression is: ; Use and the expected completion rate to calculate the task completion rate deviation index, the expression is: ; In the formula, FCHK is the task completion rate deviation index.

[0048] A relatively large task completion rate deviation index means that the deviation between the actual task completion rate and the expected completion rate is significant; this situation usually reflects insufficient emergency task scheduling capabilities of the low-altitude flight platform, and there may be the following problems: The dynamic adjustment of task priorities is not precise enough, resulting in resource allocation deviations; The platform fails to fully adapt to environmental changes (such as increased airspace risks or communication interruptions) during task execution; Or the scheduling system fails to respond to emergencies in real time and optimize task paths and strategies in a timely manner; Therefore, a large deviation index indicates that the scheduling efficiency and stability of the platform need to be improved, especially during the execution of high-priority tasks.

[0049] A relatively small task completion rate deviation index indicates that the gap between the actual task completion rate and the expected completion rate is small, indicating strong emergency task scheduling capabilities of the low-altitude flight platform; This is reflected in the scheduling system's ability to accurately identify task priorities and reasonably allocate resources, effectively adjust path planning and platform collaboration in a dynamic environment; At the same time, the platform has strong adaptability and robustness in the face of emergencies, ensuring the efficient completion of tasks; A low deviation index reflects that the efficiency and stability of the scheduling system have reached a relatively high level, which is conducive to meeting the high-efficiency response requirements of emergency rescue.

[0050] After analyzing the airspace risk characteristics of emergency tasks, generate an abnormal flight conflict rate index. The method for obtaining the abnormal flight conflict rate index is: Define the nodes (Nodes, V) of the graph structure: Each node represents a low-altitude flight platform, and the node characteristics include flight status and airspace risk data: Flight status: position (x, y, z), speed, direction, etc.; Airspace risk: dynamic obstacle density, wind field disturbance intensity, etc.; Edges (Edges, E): The edges represent the relationships between flight platforms (such as distance, speed difference), and the weights of the edges reflect conflict risks; Edge characteristics: distance between platforms, relative speed, communication link quality, etc.

[0051] The graph is defined as G=(V,E), and the initial feature of each node is a vector , and the feature of each edge is a vector , in each round of propagation, each node receives information from neighbor nodes and updates its own features: ; In the formula, is the information from node j to node i at the l-th layer, Generate a message function (such as a multi-layer perceptron, MLP); aggregate neighbor information: ; where is the sum of the information of all neighbor nodes received by node i, is the neighbor set of node i; update the node features using the received messages: ; where is a feature update function (such as GRU or MLP); the node features of the last layer are used to predict the conflict probability of each node (platform): ; is the conflict probability of node i, and σ is an activation function (such as Sigmoid); Calculate the expected conflict probability of each node based on historical data ; for each node, calculate the deviation between the actual and expected conflict probabilities , and the expression is: ; aggregate the deviations of all nodes and calculate the flight conflict rate anomaly index FCRAI, and the expression is: ; where N is the total number of nodes in the graph (the number of flight platforms), is the flight conflict rate anomaly index.

[0052] A larger flight conflict rate anomaly index indicates that the actual flight conflict rate is much higher than the expected level, reflecting significant problems in the emergency task scheduling ability of low-altitude flight platforms; this may be due to the task scheduling system failing to effectively avoid path crossings between platforms, or the airspace management strategy not being fully adapted to complex environments (such as high-density obstacles, dynamic risks, etc.); in addition, the scheduling system may have deficiencies in resource allocation, task priority adjustment, or real-time path optimization, resulting in frequent conflicts; a larger FCRAI is a warning signal, and the scheduling strategy needs to be optimized to reduce conflict risks and improve scheduling efficiency and task execution safety.

[0053] A smaller flight conflict rate anomaly index indicates that the actual flight conflict rate is close to or lower than the expected level, indicating strong emergency task scheduling ability of low-altitude flight platforms; this reflects that the scheduling system can accurately plan paths, coordinate the dynamic operation of multiple platforms, and effectively avoid airspace conflict risks; especially in high-density airspace or complex task scenarios, a lower FCRAI indicates excellent performance of the scheduling system in real-time adjustment and optimization, being able to meet the efficient and safe emergency task requirements and ensure the stable operation of platforms and the success rate of tasks.

[0054] Convert the task completion rate deviation index and the flight conflict rate anomaly index into a comprehensive feature vector, and use the comprehensive feature vector as the input of the machine learning model. The machine learning model takes predicting the emergency task scheduling ability value label of the low-altitude flight platform for each group of comprehensive feature vectors as the prediction target, and takes minimizing the sum of the prediction errors of the emergency task scheduling ability value labels of all low-altitude flight platforms as the training target, and trains the machine learning model until the sum of the prediction errors reaches convergence and then stops the model training. Determine the emergency task scheduling ability value of the low-altitude flight platform according to the model output result, where the machine learning model is a polynomial regression model.

[0055] The method for obtaining the emergency task scheduling ability value of the low-altitude flight platform is as follows: Obtain the corresponding function expression from the comprehensive feature vector training data of the trained machine learning model: ; In the formula, is the output function of the model, FCHK is the task completion rate deviation index, FCRAI is the flight conflict rate anomaly index, is the emergency task scheduling ability value of the low-altitude flight platform.

[0056] Compare the obtained emergency task scheduling ability value of the low-altitude flight platform with the pre-set emergency task scheduling ability reference threshold according to historical data. If the emergency task scheduling ability value of the low-altitude flight platform is greater than or equal to the pre-set emergency task scheduling ability reference threshold, it indicates that the emergency task scheduling ability of the low-altitude flight platform is high. At this time, generate a high task scheduling ability signal and no adjustment is required; if the emergency task scheduling ability value of the low-altitude flight platform is less than the pre-set emergency task scheduling ability reference threshold, it indicates that the emergency task scheduling ability of the low-altitude flight platform is low. At this time, generate a low task scheduling ability signal and scheduling adjustment is required.

[0057] Re-schedule multiple flight platforms through a task allocation algorithm. The re-scheduling includes task allocation, priority sorting, and dynamic adjustment of task resources. Specifically: Define the allocation goals including: maximizing task completion efficiency: ensuring that tasks are completed according to priorities while minimizing the total completion time; balancing resource usage: avoiding overloading of individual platforms or resource waste; minimizing conflict risks: ensuring that the paths and resources do not interfere with each other during the execution of multi-platform tasks.

[0058] The allocation algorithm design includes: Hungarian algorithm: applied to the one-to-one matching problem between tasks and flying platforms, finding the optimal task allocation scheme based on the cost matrix; constructing the cost matrix C(i,j) to represent the cost of platform i executing task j (such as task distance, energy consumption); finding the allocation scheme through an optimization algorithm to minimize the total cost; Auction Algorithm: platforms "bid" for tasks, submitting bids according to their capabilities and statuses, and the task is allocated to the platform with the lowest bid; Greedy algorithm based on priority: allocate tasks from high to low priority, and preferentially select the most suitable platform for high-priority tasks; each task is allocated to the most suitable flying platform, considering the following factors: the distance between the current position of the platform and the task position; the remaining energy of the platform; the task execution ability of the platform (such as load capacity, equipment support).

[0059] The priority of each task is comprehensively determined by the following factors: Urgency: the time sensitivity of the task to be completed, for example, delivering relief supplies is more urgent than a survey task; Importance: the degree of impact of the task on the overall goal, for example, a rescue task takes precedence over environmental monitoring; Environmental Risk: the risk level of the task area, for example, the priority of a task in a high-wind speed area may be reduced; Use the weighted scoring method to calculate the task priority: dynamically adjust the priority according to the change of task status (such as completion progress, remaining time); in case of task conflicts (such as multiple platforms trying to execute the same task), give priority to completing high-priority tasks.

[0060] The dynamic adjustment of task resources includes: platform resources: the power, payload, communication ability, etc. of the platform; task resource requirements: the requirements of each task for platform resources, such as execution time, power consumption, communication data volume, etc.; dynamically allocate the platform task load to avoid resource overload or idleness; for example, high-priority tasks can be preferentially allocated to platforms with higher energy; when the platform resources are insufficient, dynamically switch tasks to other platforms with sufficient resources and update the scheduling plan at the same time; multiple platforms cooperate to complete high-resource-demand tasks (such as joint monitoring or fire-fighting tasks); including using heuristic algorithms: dynamically allocate resources through heuristic rules (such as giving priority to meeting the most critical tasks); dynamic programming: in the case of a given total amount of resources and task requirements, find the optimal resource allocation strategy to maximize the task completion rate; multi-objective optimization: simultaneously optimize the task completion rate and resource utilization rate, and use genetic algorithms or particle swarm algorithms to solve the problem of dynamic resource adjustment.

[0061] S5: The flying platform performs rescue operations according to the assigned tasks, including thermal imaging monitoring, material delivery, fire extinguishing operations, and positioning of trapped persons, and feeds back the task execution status and environmental change information to the command center through a real-time communication network.

[0062] Use thermal imaging equipment to monitor the environment in the disaster area, identify the location of the fire source, trapped persons, or other abnormal heat sources; the flying platform enters the predetermined area and starts scanning with the thermal imager; the thermal imager collects infrared data and generates a thermal distribution map; identify high-temperature areas (such as the fire source) or low-temperature contrast areas (such as trapped persons); mark key points and generate a report containing the temperature distribution and possible risk areas; transmit the thermal distribution map and marked data to the command center in real time for formulating the next action strategy.

[0063] The flying platform carries emergency rescue materials (such as food, medicine, communication equipment) and drops them to the target area; determine the weight, volume, and delivery accuracy requirements of the materials, and select a suitable flying platform; load the materials and plan the best delivery path; the flying platform arrives at the delivery area and releases the materials through the on-board control system; cooperate with the vision or positioning system to ensure the delivery accuracy; use a camera or sensor to confirm the successful delivery of the materials and feed back the data to the command center.

[0064] The flying platform is used to extinguish fires, especially in high-rise buildings or inaccessible areas; use a thermal imager and environmental sensors to locate the fire source and determine the scope of the fire; be equipped with fire extinguishing devices (such as fire extinguishing bombs, water spraying equipment); adjust the fire extinguishing angle and intensity according to the scope of the fire and the wind direction; monitor the fire extinguishing effect through thermal imaging and transmit the real-time data of the temperature change of the fire source back to the command center.

[0065] The flying platform accurately locates trapped persons through sensors and vision technology to support rescue operations; the flying platform enters the search mode and scans the area using cameras, thermal imagers, and sound sensors; use intelligent algorithms (such as object detection models) to identify human forms or vital signs (such as heat signals, calls for help); transmit the real-time position of the trapped persons and the surrounding environment information to the command center to provide accurate coordinates and path suggestions for the rescue team.

[0066] The flying platform transmits the task execution status and environmental change information back to the command center through the communication network: Task status: the current progress of the task, problems during execution (such as equipment failures); Environmental changes: real-time information such as the position of dynamic obstacles, the spread trend of the fire source; Platform status: remaining battery power, position coordinates, communication quality.

[0067] The flying platform realizes real-time communication among multiple nodes through distributed self-organizing network; uses optimized communication protocols (such as LoRa or 5G) to ensure low-latency and high-reliability transmission; uniformly transmits data to the command center, and provides it for decision-makers to analyze in real time through visualization tools (such as heat maps and path maps); the command center analyzes the task completion situation according to the real-time feedback data and optimizes the next action; dynamically adjusts the task allocation and path planning of the flying platform; saves the task execution data, combines it with historical data for intelligent analysis, and improves the scheduling efficiency and accuracy of future tasks.

[0068] S6: Predict in advance the abnormal situations of the emergency task scheduling ability of the low-altitude flying platform in the subsequent fixed time period, and before the occurrence of abnormal events, automatically adjust the platform operation state based on the preset redundant control strategy and emergency response mechanism to avoid risks.

[0069] Collect the emergency task scheduling ability values of the low-altitude flying platform generated in the subsequent fixed time period that are less than the preset reference threshold of the emergency task scheduling ability, establish the corresponding data set, calculate the mean and standard deviation of the data set, and after analyzing it, predict in advance the abnormal situations of the emergency task scheduling ability of the low-altitude flying platform in the subsequent fixed time period.

[0070] If the mean of the abnormal coefficient in the data set is greater than or equal to the reference threshold of the mean of the abnormal coefficient, and the standard deviation of the abnormal coefficient is less than the reference threshold of the standard deviation of the abnormal coefficient, the overall level of the abnormal scheduling ability value is relatively low and the distribution is concentrated, indicating that there are obvious scheduling ability problems in multiple platforms and the problems are highly concentrated; at this time, generate a first-level warning signal and immediate measures need to be taken, such as reallocating tasks, adjusting priorities or increasing resource redundancy.

[0071] If the mean of the abnormal coefficient is greater than or equal to the reference threshold of the mean of the abnormal coefficient, and the standard deviation of the abnormal coefficient is greater than or equal to the reference threshold of the standard deviation of the abnormal coefficient, the overall level of the abnormal scheduling ability value is relatively low and the distribution fluctuates greatly, indicating that there are serious problems in some platforms but the distribution of the problems is relatively discrete; at this time, generate a second-level warning signal and the reasons for the abnormal platforms need to be analyzed one by one and targeted optimization measures need to be implemented.

[0072] If the mean of the abnormal coefficient is less than the reference threshold of the mean of the abnormal coefficient, and the standard deviation of the abnormal coefficient is greater than or equal to the reference threshold of the standard deviation of the abnormal coefficient, the overall level of the abnormal scheduling ability value has not reached a significantly low level but fluctuates greatly, indicating that the instability of the platform scheduling performance increases; at this time, generate a third-level warning signal and the platform status needs to be monitored and analyzed to prevent further deterioration.

[0073] If the mean value of the anomaly coefficient is less than the reference threshold of the mean value of the anomaly coefficient, and the standard deviation of the anomaly coefficient is less than the reference threshold of the standard deviation of the anomaly coefficient, the overall level and distribution of the abnormal scheduling ability value are within the normal range, and no adjustment measures need to be taken.

[0074] It should be noted here that the importance of the first-level warning signal is greater than that of the second-level warning signal, and the importance of the second-level warning signal is greater than that of the third-level warning signal. Relevant personnel can take corresponding handling measures according to different warning signal levels.

[0075] The response measures for the first-level warning signal include: optimizing resource allocation to give priority to high-priority tasks; reallocating overloaded tasks to other platforms with sufficient resources; enabling standby platforms to execute tasks to increase redundancy and improve system stability; dynamically adjusting path planning to reduce conflicts and delays.

[0076] The response measures for the second-level warning signal include: analyzing the reasons for the low scheduling ability of each abnormal platform one by one, such as communication interruption, insufficient energy or path planning failure; providing personalized optimization strategies; focusing on optimizing the scheduling of tasks related to abnormal platforms to avoid affecting the normal operation of other platforms.

[0077] The response measures for the third-level warning signal include: real-time monitoring of platforms with large fluctuations in the scheduling ability value to ensure that their scheduling ability does not further decline; adjusting task allocation and path planning in advance to reduce the risk of future decline in scheduling ability.

[0078] In this embodiment, through multi-sensor fusion technology, the low-altitude flight platform can real-time sense the complex urban environment, collect three-dimensional environment data, dynamic obstacle information and platform operation status; using distributed self-organizing network technology to achieve real-time communication and data sharing between platforms, even in the case of communication network interruption, the relay node can ensure efficient information transmission; combining real-time sensing data, using dynamic path planning algorithm to generate the optimal path, and dynamically adjusting the path through reinforcement learning algorithm to avoid obstacles; analyzing and evaluating the dynamic characteristics of task priority and airspace risk characteristics, and using task allocation algorithm to reschedule multiple platforms, including task allocation, priority sorting and dynamic adjustment of task resources; the platform performs rescue operations such as thermal imaging monitoring, material delivery, fire extinguishing operation and personnel positioning according to the assigned tasks, and real-time feedbacks the task execution status and environmental change information through the communication network; finally, predicting in advance the abnormal scheduling ability within the subsequent fixed time period, combining redundant control strategy and emergency response mechanism, automatically adjusting the platform operation status to avoid risks, so as to ensure the efficient and safe completion of urban emergency rescue tasks.

[0079] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0080] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any combination thereof; when implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product; the computer program product includes one or more computer instructions or computer programs; when the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part; the computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices; the computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner; the computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more collections of available media; the available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium; the semiconductor medium can be a solid-state drive.

[0081] It should be understood that the term "and / or" in this article is only a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural; in addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood by referring to the context; those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware or a combination of computer software and electronic hardware; whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution; those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0082] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application.

Claims

1. Urban emergency management and rescue method assisted by low-altitude flight platform, characterized in that: It includes the following steps: S1: Use multi-sensor fusion technology to perform real-time perception of the complex urban environment. The sensors include lidar, cameras, ultrasonic sensors, and inertial measurement units, which are used to collect three-dimensional environmental data, dynamic obstacle information, and flight platform operation status data of the target area; S2: Implement real-time communication and data sharing between low-altitude flight platforms through distributed ad-hoc network technology. In the case where the urban communication network is partially or completely interrupted, a relay node mechanism is adopted to ensure the efficient transmission of information between platforms; S3: Based on the real-time perceived environmental data, use a dynamic path planning algorithm to generate an optimal flight path, and combine a reinforcement learning algorithm to dynamically adjust the path during flight to avoid obstacles; S4: Analyze the dynamic characteristics of the priority of emergency tasks and the airspace risk characteristics, evaluate the emergency task scheduling ability of low-altitude flight platforms. According to the evaluation results, re-schedule multiple flight platforms through a task allocation algorithm. The re-scheduling includes task allocation, priority sorting, and dynamic adjustment of task resources; S5: The flight platform performs rescue operations according to the assigned tasks, including thermal imaging monitoring, material delivery, fire extinguishing operations, and positioning of trapped personnel, and feeds back the task execution status and environmental change information to the command center through a real-time communication network; S6: Predict in advance the abnormal situations of the emergency task scheduling ability of low-altitude flight platforms within subsequent fixed time periods. Before an abnormal event occurs, based on a preset redundant control strategy and emergency response mechanism, automatically adjust the platform operation status to avoid risks.

2. The method for urban emergency management and rescue assisted by a low-altitude flight platform according to claim 1, characterized in that: In S4, after analyzing the dynamic characteristics of the priority of emergency tasks, a task completion rate deviation index is generated. The method for obtaining the task completion rate deviation index is: Definition is the actual task completion rate at time t, and is defined as is the expected task completion rate at time t; determine the time window size n and assign weights , and normalize the sum of weights: ; Calculate the weighted moving average of the actual task completion rate , the expression is: ; Use and the expected completion rate to calculate the task completion rate deviation index, and the expression is: ; In the formula, FCHK is the task completion rate deviation index.

3. The method for urban emergency management and rescue assisted by a low-altitude flight platform according to claim 2, wherein: In S4, after analyzing the airspace risk characteristics of emergency tasks, a flight conflict rate anomaly index is generated. The method for obtaining the flight conflict rate anomaly index is: Nodes that define the structure of the graph, each node represents a low-altitude flight platform, the node features include flight status and airspace risk data, the edges represent the relationships between flight platforms, and the weights of the edges reflect the conflict risks. The initial feature of each node is a vector , and the feature of each edge is a vector . In each round of propagation, each node receives information from its neighbor nodes and updates its own features: ; where is the information from node j to node i at the l-th layer, is the message generation function, which aggregates neighbor information: ; where is the sum of all neighbor node information received by node i, is the set of neighbors of node i. Update the node features using the received messages: ; where is the feature update function; the node features of the last layer are used to predict the conflict probability of each node: ; is the conflict probability of node i, and σ is the activation function; according to historical data, calculate the expected conflict probability of each node ; for each node, calculate the deviation between the actual and expected conflict probabilities , and the expression is: ; aggregate the deviations of all nodes and calculate the flight conflict rate anomaly index FCRAI, and the expression is: ; where N is the total number of nodes in the graph, is the flight conflict rate anomaly index.

4. The method for urban emergency management and rescue assisted by a low-altitude flight platform according to claim 3, characterized in that: In S4, convert the task completion rate deviation index and the flight conflict rate anomaly index into a comprehensive feature vector. Use the comprehensive feature vector as the input of a machine learning model. The machine learning model takes predicting the emergency task scheduling ability value label of low-altitude flight platforms for each group of comprehensive feature vectors as the prediction target, and minimizing the sum of prediction errors for the emergency task scheduling ability value labels of all low-altitude flight platforms as the training target. Train the machine learning model until the sum of prediction errors reaches convergence and then stop the model training. Determine the emergency task scheduling ability value of low-altitude flight platforms according to the model output results. Among them, the machine learning model is a polynomial regression model.

5. The method for urban emergency management and rescue assisted by a low-altitude flight platform according to claim 4, characterized in that: Compare the obtained emergency task scheduling ability value of the low-altitude flight platform with the pre-set reference threshold of the emergency task scheduling ability based on historical data. If the emergency task scheduling ability value of the low-altitude flight platform is greater than or equal to the pre-set reference threshold of the emergency task scheduling ability, it indicates that the emergency task scheduling ability of the low-altitude flight platform is high. At this time, generate a high task scheduling ability signal and no adjustment is required. If the emergency task scheduling ability value of the low-altitude flight platform is less than the pre-set reference threshold of the emergency task scheduling ability, it indicates that the emergency task scheduling ability of the low-altitude flight platform is low. At this time, generate a low task scheduling ability signal and scheduling adjustment is required.

6. The method for urban emergency management and rescue assisted by a low-altitude flight platform according to claim 1, wherein: In S6, predict in advance the abnormal situation of the emergency task scheduling ability of the low-altitude flight platform within the subsequent fixed time period. Specifically: collect the emergency task scheduling ability values of the low-altitude flight platforms generated within the subsequent fixed time period that are less than the pre-set reference threshold of the emergency task scheduling ability, and establish a corresponding data set. Calculate the mean and standard deviation of the data set, and after analyzing it, predict in advance the abnormal situation of the emergency task scheduling ability of the low-altitude flight platform within the subsequent fixed time period according to the analysis results.

7. The method for urban emergency management and rescue assisted by a low-altitude flight platform according to claim 6, wherein: If the mean value of the abnormal coefficient in the data set is greater than or equal to the reference threshold of the mean value of the abnormal coefficient, and the standard deviation of the abnormal coefficient is less than the reference threshold of the standard deviation of the abnormal coefficient, the overall level of the abnormal scheduling ability value is low and the distribution is concentrated, indicating that there are obvious scheduling ability problems in multiple platforms. At this time, generate a first-level warning signal and immediate measures need to be taken to reallocate tasks and adjust priorities. If the mean value of the abnormal coefficient is greater than or equal to the reference threshold of the mean value of the abnormal coefficient, and the standard deviation of the abnormal coefficient is greater than or equal to the reference threshold of the standard deviation of the abnormal coefficient, the overall level of the abnormal scheduling ability value is low and the distribution fluctuates greatly, indicating that there are serious problems in some platforms. At this time, generate a second-level warning signal and the reasons for the abnormal platforms need to be analyzed one by one. If the mean value of the abnormal coefficient is less than the reference threshold of the mean value of the abnormal coefficient, and the standard deviation of the abnormal coefficient is greater than or equal to the reference threshold of the standard deviation of the abnormal coefficient, it indicates that the instability of the platform scheduling performance increases. At this time, generate a third-level warning signal and the platform status needs to be monitored and analyzed. If the mean value of the abnormal coefficient is less than the reference threshold of the mean value of the abnormal coefficient, and the standard deviation of the abnormal coefficient is less than the reference threshold of the standard deviation of the abnormal coefficient, the overall level and distribution of the abnormal scheduling ability value are within the normal range and no adjustment measures need to be taken.

Images