Unmanned aerial vehicle emergency processing method and system

Abstract

The invention provides an unmanned aerial vehicle emergency processing method and system, and belongs to the technical field of unmanned aerial vehicle emergency management, and the method comprises the steps: detecting historical related parameters and historical environment parameters of an unmanned aerial vehicle, analyzing the historical related parameters and the historical environment parameters, and setting a triggering condition; judging whether a triggering condition is met or not according to the real-time data of the unmanned aerial vehicle, determining a corresponding emergency condition from a condition-condition table, switching to an emergency model according to the emergency condition, and preparing to execute return flight operation; acquiring current position information and environment data, planning a return path by using an obstacle avoidance algorithm, and planning a flight strategy by using an intelligent algorithm; according to real-time relevant parameters of the unmanned aerial vehicle, a landable area around a path is determined, the landable area is screened in real time, a final landing area is determined, and landing operation is executed, so that the autonomous coping ability and safety of the unmanned aerial vehicle in an emergency are improved, and it is ensured that the unmanned aerial vehicle effectively executes tasks in a complex environment.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle emergency management, and in particular to a method and system for handling unmanned aerial vehicle emergency situations. Background Art

[0002] With the rapid development of drone technology, its application in various fields is becoming more and more extensive, including logistics and transportation, agricultural monitoring, environmental monitoring, etc. However, drones may encounter various emergencies during flight, which may cause the drone to be unable to return or land safely, resulting in equipment damage or data loss. The current technology has a slow response speed in emergency situations, cannot be evaluated in real time, and cannot effectively deal with sudden obstacles.

[0003] Therefore, the present invention provides a drone emergency handling method and system. Summary of the invention

[0004] The present invention provides a method and system for handling emergency situations of unmanned aerial vehicles. By analyzing the historical and real-time data of unmanned aerial vehicles, trigger conditions are set and emergency situations are judged. The system quickly switches to the emergency model and uses obstacle avoidance algorithms and intelligent algorithms to plan the return path to ensure the safe return of the unmanned aerial vehicle. At the same time, the system screens the landing area in real time and finally determines the safe landing location, thereby improving the autonomous response capability and safety of unmanned aerial vehicles in emergency situations and ensuring that unmanned aerial vehicles can effectively perform tasks in complex environments.

[0005] The present invention provides a method for handling unmanned aerial vehicle emergencies, comprising: Step 1: Detect historical related parameters and historical environmental parameters of the drone, analyze the historical related parameters and historical environmental parameters, and set trigger conditions; Step 2: Determine whether the trigger condition is met based on the real-time data of the drone, and determine the corresponding emergency situation from the condition-situation table, switch to the emergency model based on the emergency situation, and prepare to perform the return operation; Step 3: Obtain current location information and environmental data, use obstacle avoidance algorithm to plan the return path, and use intelligent algorithm to plan the flight strategy; Step 4: Determine the landing area around the path according to the real-time relevant parameters of the UAV, screen the landing area in real time, determine the final landing area, and perform the landing operation.

[0006] The present invention provides a method for handling unmanned aerial vehicle emergency situations, which detects historical relevant parameters and historical environmental parameters of the unmanned aerial vehicle, analyzes the historical relevant parameters and historical environmental parameters, and sets trigger conditions, including: Detecting historical related parameters of the UAV, extracting parameter features of the historical related parameters, and determining parameter change trends based on the extracted features; Obtain the historical environmental parameters corresponding to each historical related parameter, and analyze the parameter relationship between the historical related parameters and the historical environmental parameters based on the change trend; Set trigger conditions based on parameter relationships and change trends.

[0007] The present invention provides a method for handling unmanned aerial vehicle emergency situations, which obtains historical environmental parameters corresponding to each historical relevant parameter, and analyzes the parameter relationship between the historical relevant parameter and the historical environmental parameter based on a change trend, including: Calculating a first correlation coefficient and a second correlation coefficient between the historical related parameter and the historical environmental parameter, and generating a coefficient heat map according to the first correlation coefficient and the second correlation coefficient combined with the change trend; Based on the coefficient heat map, the historical related parameters and the corresponding historical environmental parameters are grouped, and the correlation characteristics of each parameter grouping result are determined according to the coefficient heat map corresponding to the parameter grouping result, and the parameter grouping results are clustered to determine the central characteristics of different parameter grouping results, and the parameter grouping results are principal component analyzed to determine the variation characteristics of different parameter grouping results; The parameter relationship between the historical related parameters and the corresponding historical environment parameters is determined according to the correlation characteristics, the central characteristics and the variation characteristics.

[0008] The present invention provides a method for handling unmanned aerial vehicle emergency situations, which sets trigger conditions according to parameter relationships and change trends, and includes: Collect historical abnormal events and their corresponding consequences, analyze the conditions under which the historical abnormal events occurred, determine the normal range of each historical related parameter based on the conditions, and set a mild abnormality threshold, a moderate abnormality threshold, and a severe abnormality threshold based on the normal range; Setting primary conditions according to parameter relationships and change trends, combining the primary conditions according to a mild abnormality threshold, a moderate abnormality threshold, and a severe abnormality threshold to obtain a logical combination result, and constructing a logic tree based on the logical combination result; The logic tree is tested in a simulation environment, and the structure of the logic tree is adjusted according to the test results to obtain the trigger conditions.

[0009] The present invention provides a method for handling emergency situations of unmanned aerial vehicles, which determines whether a trigger condition is met according to the real-time data of the unmanned aerial vehicle, determines the corresponding emergency situation from the condition-situation table, switches to an emergency model according to the emergency situation, and prepares to perform a return operation, including: Compare the real-time data with the trigger conditions to determine whether the trigger conditions are met, and record the trigger event and the corresponding trigger type; According to the trigger type and the trigger event, the corresponding emergency situation is searched from the predefined condition-situation table, and the identified emergency situation is classified, and the corresponding emergency model is matched from the classification-emergency model library according to the classification result; The execution conditions corresponding to the emergency model are set according to the emergency situation, and the drone prepares to perform the return operation based on the execution conditions.

[0010] The present invention provides a method for emergency handling of a drone, which obtains current position information and environmental data, uses an obstacle avoidance algorithm to plan a return path, and uses an intelligent algorithm to plan a flight strategy, including: Obtain the current location information and environmental data of the drone in an emergency, use the obstacle avoidance algorithm, and perform initial path planning from the current location information point to the control point; Obstacle points on the initial path are obtained based on the current position information and environmental data, and the obstacle points are used as repulsive force sources and the control points are used as attractive force sources. The next path point is calculated according to the repulsive force source and attractive force source, and the initial path is corrected in combination with the intelligent algorithm and the next path point to obtain the return path and determine the flight strategy; Perform a return-to-home operation according to the flight strategy.

[0011] The present invention provides a method for handling unmanned aerial vehicle emergency situations, which calculates the next path point according to the repulsive force source and the attractive force source, and comprises: ,in, represents the next path point position vector of the drone at time t+Δt; represents the current position vector of the drone at time t; m represents the mass of the drone; t represents time; The attraction gain coefficient representing the position vector G of the drone toward the attraction source; represents the time step, which means the time interval from the current time to the next time; G represents the position vector of the attraction source; represents the position vector of the UAV at time τ; The repulsive force gain coefficient representing the position vector of the drone facing the repulsive force source; represents the position vector of the i-th repulsive force source; i represents the index of the repulsive force source; The position vector representing the source of the repulsive force and the position vector of the drone at time τ The modulus value of The position vector representing the source of the repulsive force and the position vector of the drone at time τ The vector difference of .

[0012] The present invention provides a method for emergency handling of a drone, which determines a landing area around a return path according to real-time related parameters of the drone, screens the landing area in real time, determines a final landing area, and performs a landing operation, including: Determine the landing standard after each emergency situation occurs from the parameter-landing standard table, and determine the landing area within the return path and the preset range of the control point based on the landing standard; Based on the real-time relevant parameters of the drone, the landing area is screened in real time, and then the final landing area is determined and the landing operation is performed.

[0013] The present invention provides a drone emergency handling system, comprising: Trigger condition module: detects historical related parameters and historical environmental parameters of the drone, analyzes the historical related parameters and historical environmental parameters, and sets trigger conditions; Emergency situation module: judge whether the trigger condition is met according to the real-time data of the drone, determine the corresponding emergency situation from the condition-situation table, switch to the emergency model according to the emergency situation, and prepare to perform the return operation; Path module: obtains current location information and environmental data, uses obstacle avoidance algorithm to plan the return path, and uses intelligent algorithm to plan the flight strategy; Landing area module: Determine the landing area around the path according to the real-time relevant parameters of the UAV, screen the landing area in real time, determine the final landing area, and perform landing operations.

[0014] Compared with the existing technology, the beneficial effects of this application are as follows: by analyzing the historical and real-time data of the drone, setting the trigger conditions and judging the emergency situation, quickly switching to the emergency model, and using the obstacle avoidance algorithm and intelligent algorithm to plan the return path, the drone can return safely. At the same time, the landing area can be screened in real time, and the safe landing location can be finally determined, thereby improving the autonomous response capability and safety of the drone in emergency situations, and ensuring that the drone can effectively perform tasks in complex environments.

[0015] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0016] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 It is a flowchart of a method for handling unmanned aerial vehicle emergency situations provided by an embodiment of the present invention; Figure 2 It is a structural schematic diagram of a drone emergency handling system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0019] Embodiment 1: The embodiment of the present invention provides a method for handling unmanned aerial vehicle emergency situations. Figure 1 As shown, including: Step 1: Detect historical related parameters and historical environmental parameters of the drone, analyze the historical related parameters and historical environmental parameters, and set trigger conditions; Step 2: Determine whether the trigger condition is met based on the real-time data of the drone, and determine the corresponding emergency situation from the condition-situation table, switch to the emergency model based on the emergency situation, and prepare to perform the return operation; Step 3: Obtain current location information and environmental data, use obstacle avoidance algorithm to plan the return path, and use intelligent algorithm to plan the flight strategy; Step 4: Determine the landing area around the path according to the real-time relevant parameters of the UAV, screen the landing area in real time, determine the final landing area, and perform the landing operation.

[0020] In this embodiment, the history-related parameters are various data collected by the drone during the flight, such as flight altitude, recording the real-time altitude of the drone, speed, flight speed of the drone, battery power, current remaining battery power, signal strength, wireless signal strength, etc.

[0021] In this embodiment, the historical environmental parameters are data related to the flight environment of the drone, such as weather conditions, meteorological data such as wind speed, temperature, humidity, terrain information, terrain features of the flight area (such as mountains, buildings, etc.), historical locations of obstacles, and the locations and types of previously detected obstacles.

[0022] In this embodiment, by detecting and extracting the characteristics of the historical parameters of the drone, analyzing its changing trends, and obtaining the corresponding historical environmental parameters, trigger conditions are set based on the relationship between these parameters. The trigger conditions are the standards for initiating emergency handling procedures under specific circumstances. For example, the battery power is less than 20%: When the battery power is less than 20%, the return procedure is automatically triggered.

[0023] In this embodiment, the condition-situation table is a predefined database that lists different trigger conditions and their corresponding emergency situations, for example, trigger condition: battery power is less than 20%, emergency situation: the drone needs to return to avoid power outage.

[0024] In this embodiment, emergency situations refer to emergencies or abnormal conditions that may cause safety risks or mission failures during the operation of the drone. These situations may be caused by external environment, equipment failure or human factors, such as equipment failure, low battery of the drone, sensor failure, communication interruption, etc., environmental changes, sudden strong winds, heavy rain, lightning and other severe weather, obstacles, unexpected buildings, flocks of birds or other obstacles on the flight path, system anomalies, software crashes, data transmission errors, etc.

[0025] In this embodiment, the emergency model is a response framework or strategy developed for specific emergency situations, which is intended to guide the behavior and decision-making process of the drone when encountering these situations. The emergency model usually includes specific operating steps, decision criteria and response strategies. For example, the return model automatically plans the return path in the event of equipment failure or low battery. The obstacle avoidance model adjusts the flight path in real time to avoid collision when an obstacle is detected during flight. The landing model selects a suitable landing area and lands safely in the event of severe weather or other emergencies.

[0026] In this embodiment, the emergency model and the emergency situation, when the drone detects a specific emergency situation, it will identify the situation according to the condition-situation table and switch to the corresponding emergency model. The emergency model provides specific operational guidance for the drone in emergency situations to ensure that the drone can respond quickly and effectively to the emergency situations encountered. The execution results of the emergency model can be fed back to the system to help optimize future emergency handling processes and model designs, and improve the safety and reliability of the drone.

[0027] In this embodiment, the environmental data includes real-time information of the surrounding environment, such as weather conditions, wind speed, temperature, obstacle locations, etc. For example, the current wind speed is 10 m / s, and there are obstacles such as trees and buildings around.

[0028] In this embodiment, the obstacle avoidance algorithm is an algorithm for calculating a safe path, which can identify and avoid obstacles, such as the A* algorithm or the Dijkstra algorithm, which is used to find the best path in a complex environment.

[0029] In this embodiment, the landing area refers to an area within the return path and the control point range that meets the landing criteria, for example, an open grassland that meets the landing criteria can be used as the landing point of the drone.

[0030] In this embodiment, the real-time related parameters are key data that the UAV continuously monitors during flight, including, for example, the altitude, speed, location information, and obstacle information of the surrounding environment of the UAV.

[0031] In this embodiment, real-time screening refers to the process of dynamically evaluating and selecting landing areas based on real-time relevant parameters. For example, during flight, the drone analyzes the surrounding environment in real time and automatically eliminates areas that do not meet landing standards, such as obstacles or uneven ground.

[0032] In this embodiment, the final landing area is the most suitable location for the drone to land safely after real-time screening. For example, after real-time screening, a flat area with no obstructions is selected as the final landing point.

[0033] In this embodiment, the landing operation is a landing process performed by the drone according to the determined final landing area. For example, the drone starts a landing program, gradually lowers its altitude, and lands safely in the final landing area.

[0034] The working principle and beneficial effects of the above technical solution are: by analyzing the historical and real-time data of the drone, setting the trigger conditions and judging the emergency situation, quickly switching to the emergency model, and using the obstacle avoidance algorithm and intelligent algorithm to plan the return path to ensure the safe return of the drone. At the same time, the landing area is screened in real time, and the safe landing location is finally determined, thereby improving the autonomous response capability and safety of the drone in emergency situations, and ensuring that the drone can effectively perform tasks in complex environments.

[0035] Embodiment 2: The embodiment of the present invention provides a method for handling emergency situations of a drone, which detects historical related parameters and historical environmental parameters of the drone, analyzes the historical related parameters and historical environmental parameters, and sets trigger conditions, including: Detecting historical related parameters of the UAV, extracting parameter features of the historical related parameters, and determining parameter change trends based on the extracted features; Obtain the historical environmental parameters corresponding to each historical related parameter, and analyze the parameter relationship between the historical related parameters and the historical environmental parameters based on the change trend; Set trigger conditions based on parameter relationships and change trends.

[0036] In this embodiment, feature extraction is to extract meaningful indicators or patterns from historical related parameters, such as the average battery power, calculating the average battery power within a certain period of time, the fluctuation range of the flight speed, analyzing the maximum and minimum values ​​of the speed, the standard deviation of the signal strength, and evaluating the stability of the signal strength.

[0037] In this embodiment, the parameter change trend is the change pattern of the historical relevant parameters of the drone within a certain time range, such as the battery power decrease trend, the battery power gradually decreases during the flight, the signal strength decreases trend, as the flight distance increases, the signal strength gradually weakens, and the flight altitude changes, the flight altitude increases or decreases in a specific area.

[0038] In this embodiment, the parameter relationship is the mutual influence and dependence between historical related parameters and historical environmental parameters, such as battery power and temperature. In a high temperature environment, the battery power decreases faster. Flight altitude and terrain. In complex terrain, the change in flight altitude is related to the distribution of obstacles. Signal strength and distance. As the distance from the base station increases, the signal strength may decrease.

[0039] The working principle and beneficial effects of the above technical solution are: by detecting and extracting the characteristics of the historical parameters of the UAV, analyzing its changing trends, and obtaining the corresponding historical environmental parameters, trigger conditions are set based on the relationship between these parameters so as to respond quickly under specific circumstances. This can provide the UAV with real-time environmental adaptability during flight, improve the UAV's response speed to emergencies, and thus improve safety.

[0040] Embodiment 3: The embodiment of the present invention provides a method for handling emergency situations of a drone, which obtains historical environmental parameters corresponding to each historical relevant parameter, and analyzes the parameter relationship between the historical relevant parameter and the historical environmental parameter based on the change trend, including: Calculating a first correlation coefficient and a second correlation coefficient between the historical related parameter and the historical environmental parameter, and generating a coefficient heat map according to the first correlation coefficient and the second correlation coefficient combined with the change trend; Based on the coefficient heat map, the historical related parameters and the corresponding historical environmental parameters are grouped, and the correlation characteristics of each parameter grouping result are determined according to the coefficient heat map corresponding to the parameter grouping result, and the parameter grouping results are clustered to determine the central characteristics of different parameter grouping results, and the parameter grouping results are principal component analyzed to determine the variation characteristics of different parameter grouping results; The parameter relationship between the historical related parameters and the corresponding historical environment parameters is determined according to the correlation characteristics, the central characteristics and the variation characteristics.

[0041] In this embodiment, the first correlation coefficient is a statistic that measures the strength and direction of the linear relationship between two variables. The Pearson correlation coefficient is usually used. For example, the correlation coefficient between the battery power and the flight altitude is calculated to obtain the strength of the relationship between them.

[0042] In this embodiment, the second correlation coefficient may refer to the correlation of a nonlinear relationship, such as the Spearman rank correlation coefficient, which is used to evaluate the monotonic relationship between two variables, for example, calculating the Spearman correlation coefficient between signal intensity and flight speed to evaluate the nonlinear relationship between them.

[0043] In this embodiment, the coefficient heat map uses a heat map to visualize the correlation coefficient matrix, and the depth of color represents the strength of the correlation between variables. For example, in the heat map, the correlation between battery power and flight altitude is represented by a dark color, while the correlation between signal strength and flight speed is represented by a light color.

[0044] In this embodiment, the grouping process is to group parameters with strong correlation into one group according to the correlation coefficient heat map, for example, the battery power, flight altitude and speed are grouped together because they have a high correlation.

[0045] In this embodiment, the parameter grouping result is the result of dividing the parameters according to the correlation. For example, the grouping result may be: group 1, battery power, flight altitude, speed; group 2, signal strength, ambient temperature.

[0046] In this embodiment, the correlation feature is a feature of the relationship between each parameter group. For example, in group 1, the battery power is negatively correlated with the flight altitude, which means that when the battery power decreases, the flight altitude may also decrease.

[0047] In this embodiment, cluster analysis is to analyze the parameter grouping results using a clustering algorithm to identify similarities, for example, using a K-means clustering algorithm to group different parameters into several categories to find groups with similar characteristics.

[0048] In this embodiment, the central feature is a representative feature of each cluster center. For example, in the clustering result, the central feature of group 1 may be a higher battery level and a relatively lower flight altitude.

[0049] In this embodiment, principal component analysis is a dimensionality reduction technique that simplifies the data set by extracting the main components and retains most of the information. For example, principal component analysis is performed on the parameter grouping results to extract several key factors that mainly affect the performance of the drone, such as battery power and flight speed.

[0050] In this embodiment, the variation feature is a feature that describes the change of a parameter. For example, in principal component analysis, the variability of battery power is large, which may affect the stability of flight.

[0051] In this embodiment, the parameter relationship is determined by analyzing correlation characteristics, cluster analysis and principal component analysis to determine the mutual influence between different parameters. For example, through cluster analysis, a strong correlation between battery power and flight altitude is found, thereby establishing a relationship between the two; the parameter relationship is the mutual dependence and influence between different parameters. For example, there is a negative correlation between battery power and flight altitude, indicating that when the battery power decreases, the flight altitude may also decrease.

[0052] The working principle and beneficial effects of the above technical solution are: by calculating the correlation coefficient between historical relevant parameters and historical environmental parameters, a coefficient heat map is made to visualize the relationship between them. Based on the heat map, the parameters are grouped and clustered to identify correlation characteristics, central characteristics and variation characteristics, thereby establishing the relationship between the parameters, which helps to optimize the emergency response strategy of the drone and improve its decision-making ability in emergency situations.

[0053] Embodiment 4: The embodiment of the present invention provides a method for handling emergency situations of unmanned aerial vehicles, which sets trigger conditions according to parameter relationships and change trends, including: Collect historical abnormal events and their corresponding consequences, analyze the conditions under which the historical abnormal events occurred, determine the normal range of each historical related parameter based on the conditions, and set a mild abnormality threshold, a moderate abnormality threshold, and a severe abnormality threshold based on the normal range; Setting primary conditions according to parameter relationships and change trends, combining the primary conditions according to a mild abnormality threshold, a moderate abnormality threshold, and a severe abnormality threshold to obtain a logical combination result, and constructing a logic tree based on the logical combination result; The logic tree is tested in a simulation environment, and the structure of the logic tree is adjusted according to the test results to obtain the trigger conditions.

[0054] In this embodiment, historical abnormal events include drone crash, battery exhaustion, obstacle encounter, navigation signal interference, sudden climate change, software failure, etc., and the consequences of the events include drone crash, data loss, mission failure (such as failure to complete the rescue mission), etc.

[0055] In this embodiment, the occurrence conditions include environmental factors, such as extreme weather, complex terrain, electromagnetic interference, etc., equipment factors, low battery, sensor failure, GPS signal loss, etc.; human factors, operational errors, system configuration errors, etc.

[0056] In this embodiment, the normal range is derived based on the drone's own performance and historical related parameters, such as battery power 30% - 100%, GPS signal strength > 3 satellites, wind speed < 15 m / s, and temperature -10°C to 50°C.

[0057] In this embodiment, the mild abnormality threshold indicates that the system is in a mild abnormal state under a certain parameter or situation, but has not yet caused a serious impact on the system as a whole; the moderate abnormality threshold indicates that the system has entered a moderate abnormal state and may have begun to have a certain degree of impact on the normal operation of the system; the severe abnormality threshold indicates that the system has reached a severe abnormal state, which may have a serious impact on the normal operation of the system, or even cause catastrophic consequences.

[0058] In this embodiment, the primary conditions refer to parameters or conditions that can be directly observed and measured in the system. They are usually related to the normal operating status of the system. The abnormalities that have occurred in the UAV are also within the setting range of the primary conditions. For example, the primary condition corresponding to a UAV that has never had any abnormalities is a flight speed of 5-25 m / s. Another UAV has caused loss of control or failed to respond to control commands in time when the flight speed exceeded 20 m / s, so its corresponding primary condition is a flight speed of 5-18 m / s.

[0059] In this embodiment, the logical combination result is to combine the mild, moderate and severe abnormality thresholds according to the relationship and change trend between different parameters to determine the overall state of the system. For example, mild abnormality: if the battery power is <50% and the GPS signal is <5 satellites, it is marked as mild abnormality; moderate abnormality: if the battery power is <30% or the GPS signal is <3 satellites, it is marked as moderate abnormality; severe abnormality: if the battery power is <15% or the GPS signal is lost, it is marked as severe abnormality.

[0060] In this embodiment, the logic tree is a tree structure constructed based on the results of logical combination, which is used to describe the abnormal states and consequences that may occur in the system under different conditions, as well as the triggering conditions; the logic tree can help system analysts or decision makers understand the relationship between various parameters in the system, predict possible abnormal events, and take corresponding measures to deal with these abnormal events. For example, root node: drone status, sub-node 1: battery power, mild abnormality: <50%, moderate abnormality: <30%, severe abnormality: <15%, sub-node 2: GPS signal, mild abnormality: <5 satellites, moderate abnormality: <3 satellites, severe abnormality: lost, sub-node 3: wind speed, mild abnormality: >10 m / s, moderate abnormality: >15 m / s, severe abnormality: >20 m / s.

[0061] In this embodiment, by testing the logic tree in a simulation environment and observing the changes in different parameters and the results of logical combinations, the triggering conditions for the system to enter an abnormal state can be determined. These conditions are that a specific parameter reaches a certain threshold, or multiple parameters reach a certain state at the same time.

[0062] The working principle and beneficial effects of the above technical solution are: by collecting historical abnormal events and their consequences, analyzing the conditions for the occurrence of events, determining the normal range of each parameter, and setting abnormal thresholds, primary conditions are formulated based on parameter relationships and change trends, and a logic tree is generated through logical combination. The logic tree is tested and adjusted in a simulation environment to determine effective triggering conditions, thereby improving the responsiveness of drones in emergency situations, optimizing emergency decision-making processes, and improving the operational safety of drones.

[0063] Embodiment 5: The embodiment of the present invention provides a method for handling emergency situations of unmanned aerial vehicles, which determines whether a trigger condition is met according to the real-time data of the unmanned aerial vehicle, determines the corresponding emergency situation from the condition-situation table, switches to an emergency model according to the emergency situation, and prepares to perform a return operation, including: Compare the real-time data with the trigger conditions to determine whether the trigger conditions are met, and record the trigger event and the corresponding trigger type; According to the trigger type and the trigger event, the corresponding emergency situation is searched from the predefined condition-situation table, and the identified emergency situation is classified, and the corresponding emergency model is matched from the classification-emergency model library according to the classification result; The execution conditions corresponding to the emergency model are set according to the emergency situation, and the drone prepares to perform the return operation based on the execution conditions.

[0064] In this embodiment, the classification-emergency model library is a database that stores different emergency situations and their corresponding processing models, and is classified according to the identified emergency situations. For example, emergency situation: GPS signal loss, emergency model: start automatic return mode, and use the last known position for navigation.

[0065] In this embodiment, the execution condition refers to the conditions that the drone must meet to perform emergency operations under a specific emergency model, for example, the execution condition: the battery power must be higher than 10% to perform the return operation.

[0066] In this embodiment, the return operation refers to a flight mode in which the UAV automatically or manually starts returning to the starting position after identifying an emergency situation. For example, the return operation: after the UAV receives a low-battery alarm, it automatically starts the return program and returns to the starting point along the shortest path.

[0067] The working principle and beneficial effects of the above technical solution are: by comparing real-time data monitoring with trigger conditions, it is determined whether the trigger conditions are met and the trigger event is recorded. According to the trigger type, the emergency situation is found from the condition-situation table, classified and matched with the emergency model, and the execution conditions are set according to the emergency situation. The drone prepares for the return operation accordingly to achieve rapid response.

[0068] Embodiment 6: The embodiment of the present invention provides a method for emergency handling of a drone, which obtains current position information and environmental data, plans a return path using an obstacle avoidance algorithm, and plans a flight strategy using an intelligent algorithm, including: Obtain the current location information and environmental data of the drone in an emergency, use the obstacle avoidance algorithm, and perform initial path planning from the current location information point to the control point; Obstacle points on the initial path are obtained based on the current position information and environmental data, and the obstacle points are used as repulsive force sources and the control points are used as attractive force sources. The next path point is calculated according to the repulsive force source and attractive force source, and the initial path is corrected in combination with the intelligent algorithm and the next path point to obtain the return path and determine the flight strategy; Perform a return-to-home operation according to the flight strategy.

[0069] In this embodiment, the real-time geographic location of the drone includes latitude, longitude and altitude information. For example, the drone is currently located at longitude 120.1, latitude 30.2, and altitude 150 meters.

[0070] In this embodiment, the initial path is a preliminary planned path from the current position information to the target control point, for example, a straight line path from the current position to the starting point.

[0071] In this embodiment, obstacle points are the locations of objects identified on the initial path that may hinder the flight of the drone, and also include no-fly zones and, for example, the coordinates of buildings and trees found on the path; control points are the target locations where the drone needs to return or land, such as the starting launch point of the drone or a designated landing area.

[0072] In this embodiment, the repulsive force source is the identified obstacle point, and the force generated causes the drone to stay away from these areas. For example, when the distance to the obstacle is close, the repulsive force generated prompts the drone to change direction; the attractive force source is the control point position, and the force generated causes the drone to move toward this position. For example, when returning to the starting point, the control point exerts an attractive force on the drone.

[0073] In this embodiment, the next waypoint is the next target position calculated based on the current position information, obstacles and control points, for example, a safe position next to an obstacle as the next flight target of the drone.

[0074] In this embodiment, the correction process is to adjust the initial path in combination with an intelligent algorithm to avoid obstacles and optimize the flight route. For example, after an obstacle is identified, the path is recalculated to avoid the obstacle and update the flight plan.

[0075] In this embodiment, the return path is the final determined safe flight path of the drone from the current position information to the control point, such as a zigzag path bypassing buildings and trees to ensure a safe return; the flight strategy is a specific flight plan formulated based on the return path, including speed, altitude, and flight mode, such as maintaining an altitude of 150 meters and a speed of 20 m / s during the return process, and lowering the altitude when approaching obstacles.

[0076] The working principle and beneficial effects of the above technical solution are: by obtaining the current position information and environmental data of the drone, the obstacle avoidance algorithm is used for initial path planning. Obstacle points on the path are identified as the source of repulsive force, control points are identified as the source of attractive force, the next path point is calculated and the initial path is corrected, and finally a safe return path and flight strategy are obtained, thereby ensuring the safe return of the drone in an emergency, being able to respond to complex environments in real time, and enhancing the reliability of the drone in various application scenarios.

[0077] Embodiment 7: An embodiment of the present invention provides a method for handling an emergency of a drone, which calculates a next path point according to the repulsive force source and the attractive force source, and includes: ,in, represents the next path point position vector of the drone at time t+Δt; represents the current position vector of the drone at time t; m represents the mass of the drone; t represents time; The attraction gain coefficient representing the position vector G of the drone toward the attraction source; represents the time step, which means the time interval from the current time to the next time; G represents the position vector of the attraction source; represents the position vector of the UAV at time τ; The repulsive force gain coefficient representing the position vector of the drone facing the repulsive force source; represents the position vector of the i-th repulsive force source; i represents the index of the repulsive force source; The position vector representing the source of the repulsive force and the position vector of the drone at time τ The modulus value of The position vector representing the source of the repulsive force and the position vector of the drone at time τ The vector difference of .

[0078] The working principle and beneficial effects of the above technical solution are: calculating the next path point of the UAV at time t+Δt, combining the current position information and the gain coefficient of the attraction source G, dynamically adjusting the flight path, and the UAV avoiding obstacles in real time according to the distance and direction from the repulsion source by using the repulsion gain coefficient to ensure safe return, improve the UAV's response speed to emergencies, and be able to quickly adjust the flight path.

[0079] Embodiment 8: The embodiment of the present invention provides a method for emergency handling of a drone, which determines a landing area around a return path according to real-time related parameters of the drone, screens the landing area in real time, determines a final landing area, and performs a landing operation, including: Determine the landing standard after each emergency situation occurs from the parameter-landing standard table, and determine the landing area within the return path and the preset range of the control point based on the landing standard; Based on the real-time relevant parameters of the drone, the landing area is screened in real time, and then the final landing area is determined and the landing operation is performed.

[0080] In this embodiment, the parameter-landing standard table is a database containing various emergency situations and their corresponding landing standards, for example, including standards such as minimum landing area and maximum slope under different weather conditions (such as strong wind, rain and snow).

[0081] In this embodiment, the landing standard is a series of conditions that the drone needs to meet to land safely under specific circumstances. For example, in an urban environment, the landing area is required to be at least 10 square meters and there must be no obstacles higher than 1 meter.

[0082] The working principle and beneficial effects of the above technical solution are: the landing standards in emergency situations are formulated through the parameter-landing standard table, the return path and the landing area within the control point range are determined, the UAV monitors the relevant parameters in real time, screens out the areas that meet the landing standards, and finally determines a safe landing site and executes the landing operation.

[0083] Embodiment 9: The embodiment of the present invention provides a drone emergency handling system, such as Figure 2 As shown, including: Trigger condition module: detects historical related parameters and historical environmental parameters of the drone, analyzes the historical related parameters and historical environmental parameters, and sets trigger conditions; Emergency situation module: judge whether the trigger condition is met according to the real-time data of the drone, determine the corresponding emergency situation from the condition-situation table, switch to the emergency model according to the emergency situation, and prepare to perform the return operation; Path module: obtains current location information and environmental data, uses obstacle avoidance algorithm to plan the return path, and uses intelligent algorithm to plan the flight strategy; Landing area module: Determine the landing area around the path according to the real-time relevant parameters of the UAV, screen the landing area in real time, determine the final landing area, and perform landing operations.

[0084] The working principle and beneficial effects of the above technical solution are: by analyzing the historical and real-time data of the drone, setting the trigger conditions and judging the emergency situation, quickly switching to the emergency model, and using the obstacle avoidance algorithm and intelligent algorithm to plan the return path to ensure the safe return of the drone. At the same time, the landing area is screened in real time, and the safe landing location is finally determined, thereby improving the autonomous response capability and safety of the drone in emergency situations, and ensuring that the drone can effectively perform tasks in complex environments.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for handling unmanned aerial vehicle emergency situations, characterized in that: include: Step 1: Detect historical related parameters and historical environmental parameters of the drone, analyze the historical related parameters and historical environmental parameters, and set trigger conditions; Step 2: Determine whether the trigger condition is met based on the real-time data of the drone, and determine the corresponding emergency situation from the condition-situation table, switch to the emergency model based on the emergency situation, and prepare to perform the return operation; Step 3: Obtain the current location information and environmental data of the drone, use the obstacle avoidance algorithm to plan the return path, and use the intelligent algorithm to plan the flight strategy; Step 4: Determine the landing area around the path according to the real-time relevant parameters of the UAV, screen the landing area in real time, determine the final landing area, and perform the landing operation.

2. The method for emergency handling of a drone according to claim 1, characterized in that: Detecting the historical relevant parameters and historical environmental parameters of the drone, analyzing the historical relevant parameters and historical environmental parameters, and setting trigger conditions, including: Detecting historical related parameters of the UAV, extracting parameter features of the historical related parameters, and determining parameter change trends based on the extracted features; Obtain the historical environmental parameters corresponding to each historical related parameter, and analyze the parameter relationship between the historical related parameters and the historical environmental parameters based on the change trend; Set trigger conditions based on parameter relationships and change trends.

3. The method for emergency handling of a drone according to claim 2, characterized in that: Obtain the historical environment parameters corresponding to each historical related parameter, and analyze the parameter relationship between the historical related parameters and the historical environment parameters based on the change trend, including: Calculating a first correlation coefficient and a second correlation coefficient between the historical related parameter and the historical environmental parameter, and generating a coefficient heat map according to the first correlation coefficient and the second correlation coefficient combined with the change trend; Based on the coefficient heat map, the historical related parameters and the corresponding historical environmental parameters are grouped, and the correlation characteristics of each parameter grouping result are determined according to the coefficient heat map corresponding to the parameter grouping result, and the parameter grouping results are clustered to determine the central characteristics of different parameter grouping results, and the parameter grouping results are principal component analyzed to determine the variation characteristics of different parameter grouping results; The parameter relationship between the historical related parameters and the corresponding historical environment parameters is determined according to the correlation characteristics, the central characteristics and the variation characteristics.

4. The method for emergency handling of a drone according to claim 3, characterized in that: Set trigger conditions based on parameter relationships and change trends, including: Collect historical abnormal events and their corresponding consequences, analyze the conditions under which the historical abnormal events occurred, determine the normal range of each historical related parameter based on the conditions, and set a mild abnormality threshold, a moderate abnormality threshold, and a severe abnormality threshold based on the normal range; Setting primary conditions according to parameter relationships and change trends, combining the primary conditions according to a mild abnormality threshold, a moderate abnormality threshold, and a severe abnormality threshold to obtain a logical combination result, and constructing a logic tree based on the logical combination result; The logic tree is tested in a simulation environment, and the structure of the logic tree is adjusted according to the test results to obtain the trigger conditions.

5. The method for emergency handling of a drone according to claim 1, characterized in that: Determine whether the trigger condition is met based on the real-time data of the drone, and determine the corresponding emergency situation from the condition-situation table. Switch to the emergency model based on the emergency situation and prepare to perform the return operation, including: Compare the real-time data with the trigger conditions to determine whether the trigger conditions are met, and record the trigger event and the corresponding trigger type; According to the trigger type and the trigger event, the corresponding emergency situation is searched from the predefined condition-situation table, and the identified emergency situation is classified, and the corresponding emergency model is matched from the classification-emergency model library according to the classification result; The execution conditions corresponding to the emergency model are set according to the emergency situation, and the drone prepares to perform the return operation based on the execution conditions.

6. The method for emergency handling of a drone according to claim 1, characterized in that: Obtain current location information and environmental data, use obstacle avoidance algorithms to plan the return path, and use intelligent algorithms to plan flight strategies, including: Obtain the current location information and environmental data of the drone in an emergency, use the obstacle avoidance algorithm, and perform initial path planning from the current location information point to the control point; Obstacle points on the initial path are obtained based on the current position information and environmental data, and the obstacle points are used as repulsive force sources and the control points are used as attractive force sources. The next path point is calculated according to the repulsive force source and attractive force source, and the initial path is corrected in combination with the intelligent algorithm and the next path point to obtain the return path and determine the flight strategy; Perform a return-to-home operation according to the flight strategy.

7. A method for handling unmanned aerial vehicle emergencies according to claim 6, characterized in that: Calculating the next path point according to the repulsive force source and the attractive force source includes: ,in, represents the next path point position vector of the drone at time t+Δt; represents the current position vector of the drone at time t; m represents the mass of the drone; t represents time; The attraction gain coefficient representing the position vector G of the drone toward the attraction source; represents the time step, which means the time interval from the current time to the next time; G represents the position vector of the attraction source; represents the position vector of the UAV at time τ; The repulsive force gain coefficient representing the position vector of the drone facing the repulsive force source; represents the position vector of the i-th repulsive force source; i represents the index of the repulsive force source; The position vector representing the source of the repulsive force and the position vector of the drone at time τ The modulus of The position vector representing the source of the repulsive force and the position vector of the drone at time τ The vector difference of .

8. The method for emergency handling of a drone according to claim 1, characterized in that: Determine the landing area around the return path according to the real-time relevant parameters of the UAV, screen the landing area in real time, determine the final landing area, and perform landing operations, including: Determine the landing standard after each emergency situation occurs from the parameter-landing standard table, and determine the landing area within the return path and the preset range of the control point based on the landing standard; Based on the real-time relevant parameters of the drone, the landing area is screened in real time, and then the final landing area is determined and the landing operation is performed.

9. A drone emergency handling system, characterized in that: include: Trigger condition module: detects historical related parameters and historical environmental parameters of the drone, analyzes the historical related parameters and historical environmental parameters, and sets trigger conditions; Emergency situation module: Determine whether the trigger condition is met based on the real-time data of the drone, determine the corresponding emergency situation from the condition-situation table, switch to the emergency model based on the emergency situation, and prepare to perform the return operation; Path module: obtains current location information and environmental data, uses obstacle avoidance algorithm to plan the return path, and uses intelligent algorithm to plan the flight strategy; Landing area module: Determine the landing area around the path according to the real-time relevant parameters of the UAV, screen the landing area in real time, determine the final landing area, and perform landing operations.

Images