Real-time monitoring system of intelligent unmanned aerial vehicle

Abstract

The invention discloses a real-time monitoring system for an intelligent unmanned aerial vehicle, and relates to the technical field of unmanned aerial vehicle monitoring, and the system comprises an environment sensing module, an intelligent positioning control module, a linkage decision module and a data feedback module: the environment sensing module obtains real-time environment data in an operation scene, and formulates an operation flight path of the unmanned aerial vehicle; the intelligent positioning control module is used for acquiring corresponding real-time space image data on an operation flight path of the unmanned aerial vehicle, identifying a retrieval target and performing positioning analysis; the linkage decision module records a target motion track in real time and predicts the target motion track, performs abnormal probability distribution analysis, and determines an unmanned aerial vehicle linkage model; planning an unmanned aerial vehicle decision according to the unmanned aerial vehicle linkage model; according to the invention, the bird repelling efficiency of the operation scene and the unmanned aerial vehicle linkage are improved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle (UAV) monitoring, and in particular to a real-time monitoring system for intelligent UAVs. Background Art

[0002] With the rapid development of drone technology, its application in agriculture, airports, power inspection and other fields is becoming more and more extensive. However, in the above scenarios, there is a threat to the safe operation of scene operations caused by bird activities, and traditional bird-repelling methods have problems such as low efficiency and poor accuracy. Using bird-repelling drones for scene operations to repel birds has become a new trend. However, current drone bird-repelling still has defects. Most of them are single scene inspections, and they track targets and expel them. The expulsion route of bird targets is unpredictable, and there is a lack of linkage and coordination with drone deployment, resulting in poor bird-repelling effect and low accuracy of current bird-repelling drones. Summary of the Invention

[0003] The purpose of the present invention is to provide a real-time monitoring system for an intelligent unmanned aerial vehicle to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solutions: A real-time monitoring system for intelligent drones, comprising an environmental perception module, an intelligent positioning control module, a linkage decision module, and a data feedback module: The environmental perception module acquires real-time environmental data in the operation scene through the environmental acquisition drone, and formulates the drone operation flight trajectory route based on the operation scene environmental data; the intelligent positioning control module collects the corresponding real-time spatial image data on the drone operation flight trajectory according to the drone operation flight, identifies and retrieves the target from the spatial image data by constructing a visual recognition model, and performs real-time target positioning analysis based on the recognition result; the linkage decision module identifies the target positioning data according to the visual recognition model, records the real-time target motion trajectory and performs real-time prediction and simulation of the target motion trajectory; performs abnormal probability distribution analysis based on the real-time prediction and simulation of the target motion trajectory, and determines the drone linkage model based on the analysis result; and plans drone decisions based on the drone linkage model.

[0005] Furthermore, the environment perception module includes an operation scene environment data acquisition unit and a flight trajectory formulation unit; The operation scene environment data acquisition unit obtains the safe operation range in the operation scene and collects real-time environment data based on the safe operation range; it collects environment data by combining an environment acquisition drone with a sensor device; the real-time environment data includes temperature data, wind speed data, humidity data, pressure data, etc. The flight trajectory formulation unit determines the operating parameter settings of the drone based on the real-time operating scene environment data, and formulates the operating inspection route of the drone; the operating parameter settings of the drone refer to determining the flight parameters of the drone based on the real-time environment, and its parameters include speed parameters, power parameters, steering angle, etc.

[0006] Furthermore, the safe operating range is the safe hemispherical space when equipment or personnel are operating in the operating scene; the operating inspection route refers to the inspection flight route of the drone within the safe space in the operating scene; and interference detection is performed on the inspection flight route by the drone.

[0007] Furthermore, the intelligent positioning control module includes a visual retrieval unit and a positioning analysis unit; The visual retrieval unit collects real-time spatial image data along the inspection route of the UAV and transmits it remotely to the monitoring terminal; the monitoring terminal performs continuous time target recognition on the spatial image data by building a visual recognition model; The positioning analysis unit performs positioning analysis on the target recognition results of the real-time spatial image data on the UAV operation inspection route according to the visual recognition model; if the recognition result is that there is no target, the flight positioning data of the UAV is continuously recorded and transmitted to the monitoring end; if the recognition result is that there is a target, the positioning of the UAV and the target are recorded at the same time, and the spatial relative position between the UAV and the target is determined; wherein, the spatial relative position is determined by spatially dotting the UAV and the target respectively, taking the spatial circumscribed spheres of the UAV and the target respectively, taking the centers of the spheres as the spatial positioning nodes of the UAV and the target respectively, and connecting the two nodes to obtain the spatial relative position data between the UAV and the target; the positioning record of the UAV and the target and the spatial relative position data are transmitted to the monitoring end.

[0008] Furthermore, the specific steps of performing continuous time target recognition on spatial image data by constructing a visual recognition model are as follows: By filtering the scene background pixels of the real-time spatial image data collected by the UAV, dynamic flying targets in the spatial image data are extracted; By performing pixel decomposition on the dynamic flying target image data, the contour pixel points of the dynamic flying target image are obtained, and by performing fusion processing on the contour pixel points, adjacent contour pixel points are connected respectively to determine the slope value of each pixel point connection segment; by performing difference calculation on the slopes between adjacent pixel point connection segments, by setting a slope judgment threshold, if the absolute value of the slope difference between adjacent pixel point connection segments is less than or equal to the slope judgment threshold, the pixel points on the adjacent pixel point connection segment are screened out; otherwise, the pixel points are retained; thereby obtaining the contour posture image data of the dynamic flying target; According to the fusion processing results of the contour pixels of the dynamic flight target image, the remaining contour pixels of the dynamic flight target image data are labeled; by mapping the posture image data of the dynamic flight target in the continuous time space image data to the same coordinate system, the spatial position change data of the contour pixels with the same label in the continuous time space image data are recorded, and the motion trajectory of the pixels with the same label in the continuous time is obtained by line processing; by comparing the curvature values ​​of the continuous time motion trajectory of adjacent labeled contour pixels, combined with the contour posture image data of the dynamic flight target in the continuous time, the database is matched and searched to determine the target; when performing matching retrieval, the initial retrieval is performed through the contour posture image data of the dynamic flight target in the continuous time; secondly, the re-retrieval is performed based on the curvature values ​​of the continuous time motion trajectory of the adjacent labeled contour pixels, and the curvature threshold is set. The database retrieval targets with a value less than the curvature threshold are screened out, and the type distribution of the retained database retrieval targets after screening is calculated. By determining the target's proportion weight, if the occupation weight is higher than the set threshold, it is judged to be a confirmed target; otherwise, it is fed back to the monitoring end for manual judgment by the staff; wherein, the spatial position change data of the same-label contour pixel points in the continuous time-space image data record at least one complete trajectory movement, which is expressed as the spatial position change data of the same-label contour pixel points beginning to have a trajectory repetition, then the recording is stopped; otherwise, if the spatial position change data of the same-label contour pixel points are all at one point in the continuous time, the response time is set, and the recording is continued within the response time until a complete trajectory movement is taken; if the correct data is still not recorded within the response time, the recording is stopped, and the feedback is fed back to the monitoring end for manual judgment.

[0009] Furthermore, the linkage decision module includes a target trajectory tracking and analysis unit and a linkage decision unit; The target trajectory tracking and analysis unit determines the target's real-time motion trajectory based on the target's real-time positioning data and the recorded target positioning data; and predicts and simulates the target's motion trajectory based on the target's real-time motion trajectory; The linkage decision-making unit performs an abnormal analysis of the operation scene based on the real-time motion trajectory prediction simulation, divides the safe operation range of the operation scene into levels, determines the probability distribution analysis of the abnormal impact of the real-time motion trajectory prediction simulation generated route on different levels of space in the safe operation range of the operation scene, and constructs a drone linkage model based on the analysis results to determine the decision.

[0010] Furthermore, the specific steps of predicting and simulating the target's motion trajectory based on the target's real-time motion trajectory are as follows: By collecting positioning data for the target, recording its movement trajectory, and using the target spatial positioning point recorded at the current real-time time point as the starting node of the predicted route, the movement trajectory of the target within the subsequent prediction period is predicted; the prediction period is the time required for the drone with the shortest relative distance to the target spatial position at the current time point to reach the target spatial position node; The target's moving trajectory is obtained by obtaining the corresponding spatial node tangents and recording the corresponding spatial node tangent angles respectively; the maximum value A of the tangent angle corresponding to each spatial node on the target's moving trajectory is determined; the tangent angle is the angle between the tangent of the corresponding spatial node and the horizontal line; the extreme points of the target's moving trajectory are determined, and the vertical distance D between the maximum maximum point and the minimum minimum point is calculated, with D being the fluctuation distance of the moving trajectory; at the starting node of the predicted route, [0, A] is used as the tangent angle change interval, and [0, D] is used as the fluctuation distance interval of the predicted motion trajectory, and a full traversal is performed to generate a random motion trajectory route; wherein the full traversal generation of the random route means that, at the starting node of the predicted route facing the direction of the work scene, the tangent angle change interval [0, A] is fully traversed, and routes are generated for each tangent angle in the interval. On this basis, during the route generation process for each tangent angle, the fluctuation distance of the generated route is limited to be fully traversed between [0, D].

[0011] Furthermore, the safe operation range of the operation scene is hierarchically divided into the following steps: determining the center of the range within the safe operation range in the operation scene, and planning the first safety distance, the second safety distance, and the third safety distance as the sphere radius with the center of the range as the sphere center, and hierarchically dividing the hemispherical space of the safe operation range into the first safety layer, the second safety distance, and the third safety layer; wherein the first safety distance is greater than the second safety distance, and the second safety distance is greater than the third safety distance; Combined with the hierarchical division results of the safe operation range of the operation scene, the probability distribution analysis of abnormal impact situations of different levels of space in the safe operation range of the operation scene is carried out based on the random motion trajectory routes generated by the starting node of the predicted route where the real-time target is located; by coordinating the total number of random motion trajectory routes generated by the starting node of the predicted route where the real-time target is located, and coordinating the distribution of different levels of space in the safe operation range of the operation scene involved in each generated random motion trajectory route; respectively calculate the proportion of the number of random motion trajectory routes involving the third safety layer, the proportion of the number of random motion trajectory routes involving the second safety layer, and the proportion of the number of random motion trajectory routes involving the first safety layer at the current time point; respectively compare the proportion of each level. Corresponding to the proportion threshold, the UAV linkage model is determined, and the decisions are determined as conventional decision-making, first-level decision-making, second-level decision-making and third-level decision-making respectively; among them, according to the level comparison priority, the third level is greater than the second level, and the second level is greater than the first level; then the corresponding is that if the proportion of the number of random motion trajectory routes involving the third safety layer is greater than the corresponding level proportion threshold, the third level decision is directly executed; otherwise, the proportion of the number of random motion trajectory routes involving the second safety layer is judged, and if it is greater than the threshold, the second level decision is directly executed; otherwise, the proportion of the number of random motion trajectory routes involving the first safety layer is judged, and if it is greater than the threshold, the first level decision is directly executed; otherwise, the conventional decision is executed; The conventional decision-making is to track the UAV with the shortest relative distance to the target spatial positioning point at the current time point in real time and drive the target away from the safe operating range of the operation scene; The first-level decision-making is to build a drone linkage model by linking two drones, in which the drone with the shortest relative distance to the target spatial distance positioning point at the current time point tracks and drives the target in real time; the first-level spatial point with the shortest vertical distance to the target spatial positioning point at the current time point is determined, and the drone with the shortest distance to the first-level spatial point at the current moment is mobilized to fly parallel to the target on the first level, and to drive the target away when the target enters its expulsion range; the first level is the outer boundary surface on the first safety layer; The second-level decision-making is to build a drone linkage model by linking three drones. Based on the first-level decision-making, the second-level spatial point with the shortest vertical distance to the target spatial positioning point at the current time point is determined, and the drone with the shortest distance to the second-level spatial point at the current moment is mobilized to fly parallel to the target on the second level, and to expel the target when it enters its expulsion range; the second level is the outer boundary surface on the second safety layer; The third-level decision is to build a drone linkage model by linking four drones. Based on the second-level decision, the third-level spatial point with the shortest vertical distance to the target spatial positioning point at the current time point is determined, and the drone with the shortest distance from the third-level spatial point at the current moment is mobilized, and ordered to fly parallel to the target on the third level, and to expel the target when it enters its expulsion range; the third level is the outer boundary surface on the third safety layer.

[0012] Furthermore, the data feedback module includes an alert unit and an event recording unit; When the warning unit makes a decision on the execution of the UAV in the operation scene, it sends a warning prompt to the monitoring terminal; The event recording unit logs the drone decision-making behavior events that exist in the operation scene.

[0013] Compared with the prior art, the present invention has the following beneficial effects: The present invention realizes accurate target recognition and multi-level bird-repelling drone decision generation in the operation scene by combining the visual recognition model and the linkage model; it identifies the target and records its continuous time motion trajectory to predict the subsequent route; based on this, the predicted route is analyzed for abnormal distribution of the operation scene, so as to construct a linkage model to determine the bird-repelling decision; the present invention accurately identifies the target and performs effective path prediction, thereby realizing the scene bird-repelling behavior through efficient linkage with the bird-repelling drone; the present invention effectively improves the low efficiency and low precision of traditional bird-repelling methods and the single bird-repelling mode of current bird-repelling drones, and improves the bird-repelling efficiency of the operation scene and the linkage of drones. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The figure is a structural diagram of a real-time monitoring system for an intelligent UAV according to the present invention. DETAILED DESCRIPTION

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0016] Example: Figure 1 As shown, the present invention provides a technical solution: Provided is a real-time monitoring system for intelligent bird-repelling drones, whose detection targets are bird targets; The system includes an environmental perception module, an intelligent positioning control module, a linkage decision module and a data feedback module: The environmental perception module obtains real-time environmental data in the operation scene through the environmental acquisition drone, and formulates the flight trajectory route of the bird-repelling drone based on the operation scene environmental data; the intelligent positioning control module collects corresponding real-time spatial image data on the flight trajectory route of the bird-repelling drone according to the operation flight of the bird-repelling drone, identifies and retrieves bird targets from the spatial image data by constructing a visual recognition model, and performs real-time positioning analysis of the bird targets based on the recognition results; the linkage decision module identifies the bird target positioning data according to the visual recognition model, records the real-time bird target movement trajectory and performs real-time prediction and simulation of the bird target movement trajectory; performs abnormal probability distribution analysis based on the real-time prediction and simulation of the bird target movement trajectory, and determines the drone linkage model based on the analysis results; and plans the drone bird-repelling decision according to the drone linkage model.

[0017] Furthermore, the environment perception module includes an operation scene environment data acquisition unit and a flight trajectory formulation unit; The operation scene environment data acquisition unit obtains the safe operation range in the operation scene and collects real-time environment data based on the safe operation range; it collects environment data by combining an environment acquisition drone with a sensor device; the real-time environment data includes temperature data, wind speed data, humidity data, pressure data, etc. The flight trajectory formulation unit determines the operating parameter settings of the bird-repelling UAV based on the real-time operating scene environment data, and formulates the operating inspection route of the bird-repelling UAV; the operating parameter settings of the bird-repelling UAV refer to determining the flight parameters of the bird-repelling UAV according to the real-time environment, and its parameters include speed parameters, power parameters, steering angle, etc.

[0018] Furthermore, the safe operating range is the safe hemispherical space when equipment or personnel are operating in the operating scene; the operating inspection route refers to the inspection flight route of the bird-scaring drone within the safe space in the operating scene; and the bird-scaring drone is used to interfere with birds on the inspection flight route for detection.

[0019] Furthermore, the intelligent positioning control module includes a visual retrieval unit and a positioning analysis unit; The visual retrieval unit collects real-time spatial image data on the inspection route of the bird-repelling drone and transmits it remotely to the monitoring end; the monitoring end performs continuous-time bird target recognition on the spatial image data by building a visual recognition model; The positioning analysis unit performs positioning analysis on the bird target recognition results of the real-time spatial image data on the bird-repelling UAV's operation inspection route according to the visual recognition model; if the recognition result is that there is no bird target, the flight positioning data of the bird-repelling UAV is continuously recorded and transmitted to the monitoring end; if the recognition result is that there is a bird target, the bird-repelling UAV and the bird target are positioned and recorded at the same time, and the spatial relative position between the bird-repelling UAV and the bird target is determined; wherein, the spatial relative position is determined by spatially dotting the bird-repelling UAV and the bird target respectively, taking the spatial circumscribed spheres of the bird-repelling UAV and the bird target respectively, taking the centers of the spheres as the spatial positioning nodes of the bird-repelling UAV and the bird target respectively, and connecting the two nodes to obtain the spatial relative position data between the bird-repelling UAV and the bird target; the positioning record of the bird-repelling UAV and the bird target and the spatial relative position data are transmitted to the monitoring end.

[0020] Furthermore, the specific steps of performing continuous-time bird target recognition on spatial image data by constructing a visual recognition model are as follows: By filtering the scene background pixels of the real-time spatial image data collected by the bird-repelling drone, dynamic flying targets in the spatial image data are extracted; By performing pixel decomposition on the dynamic flying target image data, the contour pixel points of the dynamic flying target image are obtained, and by performing fusion processing on the contour pixel points, adjacent contour pixel points are connected respectively to determine the slope value of each pixel point connection segment; by performing difference calculation on the slopes between adjacent pixel point connection segments, by setting a slope judgment threshold, if the absolute value of the slope difference between adjacent pixel point connection segments is less than or equal to the slope judgment threshold, the pixel points on the adjacent pixel point connection segment are screened out; otherwise, the pixel points are retained; thereby obtaining the contour posture image data of the dynamic flying target; According to the fusion processing results of the contour pixels of the dynamic flight target image, the remaining contour pixels of the dynamic flight target image data are labeled; by mapping the posture image data of the dynamic flight target in the continuous time space image data to the same coordinate system, the spatial position change data of the contour pixels with the same label in the continuous time space image data are recorded, and the motion trajectory of the pixels with the same label in the continuous time is obtained by line processing; by comparing the curvature values ​​of the continuous time motion trajectory of adjacent labeled contour pixels, combined with the contour posture image data of the dynamic flight target in the continuous time, matching retrieval is performed in the database to determine the bird target; when performing matching retrieval, the initial retrieval is performed through the contour posture image data of the dynamic flight target in the continuous time; secondly, re-retrieval is performed based on the curvature values ​​of the continuous time motion trajectory of the adjacent labeled contour pixels, and the curvature threshold is set. The database retrieval targets with a curvature value less than the curvature threshold are screened out, and the type distribution of the retained database retrieval targets after screening is calculated. By determining the proportion weight of bird targets, if the proportion weight is higher than the set threshold, it is judged as a bird target; otherwise, it is fed back to the monitoring end for manual judgment by the staff; wherein, the spatial position change data of the same-label contour pixel points in the continuous time-space image data record at least one complete trajectory movement, which is expressed as when the spatial position change data of the same-label contour pixel points begin to have a trajectory repetition, the recording is stopped; otherwise, if the spatial position change data of the same-label contour pixel points are all at one point in the continuous time, the response time is set, and the recording is continued within the response time until a complete trajectory movement is taken; if the correct data is still not recorded within the response time, the recording is stopped, and the feedback is fed back to the monitoring end for manual judgment.

[0021] Furthermore, the linkage decision module includes a target trajectory tracking and analysis unit and a linkage decision unit; The target trajectory tracking and analysis unit determines the real-time motion trajectory of the bird target based on the real-time positioning data of the bird target and the recorded positioning data of the bird target; and predicts and simulates the motion trajectory of the bird target based on the real-time motion trajectory of the bird target; The linked bird-repellent decision-making unit performs an abnormal analysis of the operation scene based on the real-time bird motion trajectory prediction simulation, divides the operation scene safety operation range into levels, determines the probability distribution analysis of the abnormal impact of the route generated by the real-time bird motion trajectory prediction simulation on different levels of space in the safety operation range of the operation scene, and constructs a drone linked bird-repellent model based on the analysis results to determine the bird-repellent decision.

[0022] Furthermore, the specific steps of predicting and simulating the motion trajectory of the bird target based on the real-time motion trajectory of the bird target are as follows: By collecting positioning data of bird targets, recording their movement trajectory routes, and using the bird target spatial positioning point recorded at the current real-time time point as the starting node of the predicted route, the movement trajectory of the bird target in the subsequent prediction period is predicted; the prediction period is the time required for the bird-repelling drone with the shortest relative distance to the bird target spatial position node at the current time point to reach the bird target spatial position node; The tangent of the corresponding spatial node of the bird target's moving trajectory is obtained, and the tangent angle of the corresponding spatial node is recorded respectively; the maximum value A of the tangent angle corresponding to each spatial node on the bird target's moving trajectory is determined; the tangent angle is the angle between the tangent of the corresponding spatial node and the horizontal line; the extreme point of the bird target's moving trajectory is determined, and the vertical distance D between the maximum maximum point and the minimum minimum point is calculated, and D is used as the fluctuation distance of the moving trajectory route; at the starting node of the predicted route, the tangent angle change interval [0, A] and the fluctuation distance interval of the predicted movement trajectory route are fully traversed to generate a random movement trajectory route; wherein the full traversal generation of the random route means that the tangent angle change interval [0, A] is fully traversed at the starting node of the predicted route towards the work scene, and the route is generated for each tangent angle in the interval. On this basis, during the process of generating the route for each tangent angle, the fluctuation distance of the generated route is limited to be fully traversed between [0, D].

[0023] Furthermore, the safe operation range of the operation scene is hierarchically divided into the following steps: determining the center of the range within the safe operation range in the operation scene, and planning the first safety distance, the second safety distance, and the third safety distance as the sphere radius with the center of the range as the sphere center, and hierarchically dividing the hemispherical space of the safe operation range into the first safety layer, the second safety distance, and the third safety layer; wherein the first safety distance is greater than the second safety distance, and the second safety distance is greater than the third safety distance; Combined with the hierarchical division results of the safe operation range of the operation scene, the probability distribution analysis of abnormal impact situations of different levels of space in the safe operation range of the operation scene is carried out based on the random motion trajectory routes generated by the starting node of the predicted route where the real-time bird target is located; by coordinating the total number of random motion trajectory routes generated by the starting node of the predicted route where the real-time bird target is located, and coordinating the distribution of different levels of space in the safe operation range of the operation scene involved in each generated random motion trajectory route; respectively calculate the proportion of the number of random motion trajectory routes involving the third safety layer, the proportion of the number of random motion trajectory routes involving the second safety layer, and the proportion of the number of random motion trajectory routes involving the first safety layer at the current time point; respectively compare the corresponding proportion thresholds of each layer to determine A drone-linked bird-repellent model is defined, and the bird-repellent decisions are determined to be conventional bird-repellent decisions, first-level bird-repellent decisions, second-level bird-repellent decisions, and third-level bird-repellent decisions; among them, according to the level comparison priority, the third level is greater than the second level, and the second level is greater than the first level; then the corresponding is that if the proportion of the number of random motion trajectory routes involving the third safety layer is greater than the corresponding level proportion threshold, the third-level bird-repellent decision is directly executed; otherwise, the proportion of the number of random motion trajectory routes involving the second safety layer is judged, and if it is greater than the threshold, the second-level bird-repellent decision is directly executed; otherwise, the proportion of the number of random motion trajectory routes involving the first safety layer is judged, and if it is greater than the threshold, the first-level bird-repellent decision is directly executed; otherwise, the conventional bird-repellent decision is executed; The conventional bird-repelling decision is to determine the bird-repelling drone with the shortest relative distance to the spatial positioning point of the bird target at the current time point to track and drive the bird target away from the safe operating range area of ​​the operation scene in real time; The first-level bird-repelling decision is to build a drone-linked bird-repelling model by linking two bird-repelling drones, in which the bird-repelling drone with the shortest relative distance to the bird target spatial distance positioning point at the current time points performs real-time tracking and repels the bird target; the first-level spatial point with the shortest vertical distance to the bird target spatial positioning point at the current time points is determined, and the bird-repelling drone with the shortest distance to the first-level spatial point at the current moment is mobilized to fly parallel to the bird target on the first level, and to perform repels when the bird target enters its repels range; the first level is the outer boundary surface on the first safety layer; The second-level bird-repelling decision is to build a drone-linked bird-repelling model by linking three bird-repelling drones. Based on the first-level bird-repelling decision, the second-level spatial point with the shortest vertical distance to the bird target spatial positioning point at the current time is determined, and the bird-repelling drone with the shortest distance to the second-level spatial point at the current moment is mobilized to fly parallel to the bird target on the second level, and to expel the bird target when it enters its expulsion range; the second level is the outer boundary surface on the second safety layer; The third-level bird-repelling decision is to build a drone-linked bird-repelling model by linking four bird-repelling drones. Based on the second-level bird-repelling decision, the third-level spatial point with the shortest vertical distance to the bird target spatial positioning point at the current time point is determined, and the bird-repelling drone with the shortest distance from the third-level spatial point at the current moment is mobilized, and ordered to fly parallel to the bird target on the third level, and to expel the bird target when it enters its expulsion range; the third level is the outer boundary surface on the third safety layer.

[0024] Furthermore, the data feedback module includes an alert unit and an event recording unit; When the warning unit executes a bird-repelling decision on the bird-repelling drone in the operation scene, it sends a warning prompt to the monitoring terminal; The event recording unit logs the bird-repelling behavior events that occur in the operation scene; The implementation example for bird-repelling decision-making is as follows: If, in the current operation scene, the random motion trajectory routes generated by the starting nodes of the predicted routes of the real-time bird targets are combined with the safe operation range levels of the operation scene, the number of random motion trajectory routes generated by the starting nodes of the predicted routes of the real-time bird targets in the safe operation range levels of different operation scenes are used to respectively determine the number of random motion trajectory routes A3 of the third safety layer, the number of random motion trajectory routes A2 of the second safety layer, and the number of random motion trajectory routes A1 of the first safety layer; the number of random motion trajectory routes of the corresponding levels are compared with the corresponding level thresholds, and the bird-repelling decision to be executed in the current operation scene is determined according to the comparison results; If A3<a3, A2<a2, and A1<a1, then the conventional bird-repelling decision is executed, which determines the bird-repelling drone with the shortest relative distance to the spatial positioning point of the bird target at the current time point to track and drive the bird target away from the safe operating range of the operation scene in real time; where a1 corresponds to the threshold value of the number of random motion trajectory routes in the first safety layer; a2 corresponds to the threshold value of the number of random motion trajectory routes in the second safety layer; and a3 corresponds to the threshold value of the number of random motion trajectory routes in the third safety layer. If A3<a3, A2<a2, A1≥a1, the first-level bird-repelling decision is executed, which builds a drone-linked bird-repelling model by linking two bird-repelling drones, in which the bird-repelling drone with the shortest relative distance to the bird target spatial distance positioning point at the current time point tracks and repels the bird target in real time; determines the first-level spatial point with the shortest vertical distance to the bird target spatial positioning point at the current time point, mobilizes the bird-repelling drone with the shortest distance to the first-level spatial point at the current moment, and orders it to fly parallel to the bird target on the first level, and repels the bird target when it enters its repeal range; If A3<a3, A2≥a2, then according to the hierarchical comparison priority, the third level is greater than the second level, and the second level is greater than the first level. Therefore, regardless of the comparison results of A1 and a1, the second-level bird-repelling decision is executed. It constructs a drone-linked bird-repelling model by linking three bird-repelling drones. Then, on the basis of the first-level bird-repelling decision, the second-level spatial point with the shortest vertical distance to the bird target spatial positioning point at the current time point is determined, and the bird-repelling drone with the shortest distance from the second-level spatial point at the current moment is mobilized, and ordered to fly parallel to the bird target on the second level, and to expel the bird target when it enters its expulsion range; wherein, since the bird target expulsion behavior is carried out on the second level, the first level will be involved in the expulsion process, so the second-level bird-repelling decision and the first-level bird-repelling decision need to be executed successively in this process; If A3≥a3, regardless of the comparison results of A2 and a2, or A1 and a1, the third-level bird-repelling decision is executed. The drone-linked bird-repelling model is constructed by linking four bird-repelling drones. Then, on the basis of the second-level bird-repelling decision, the third-level spatial point with the shortest vertical distance to the bird target spatial positioning point at the current time point is determined, and the bird-repelling drone with the shortest distance to the third-level spatial point at the current moment is mobilized, and ordered to fly parallel to the bird target on the third level, and to expel the bird target when it enters its expulsion range; since the bird target expulsion behavior is carried out on the third level, the second level and the lower level will be involved in the expulsion process. Therefore, in this process, it is necessary to execute the third-level bird-repelling decision, the second-level bird-repelling decision, and the first-level bird-repelling decision in sequence.

[0025] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A real-time monitoring system for intelligent drones, characterized by: The system includes an environmental perception module, an intelligent positioning control module, a linkage decision module and a data feedback module: The environmental perception module acquires real-time environmental data in the operation scene through the environmental acquisition drone, and formulates the drone operation flight trajectory route based on the operation scene environmental data; the intelligent positioning control module collects the corresponding real-time spatial image data on the drone operation flight trajectory according to the drone operation flight, identifies and retrieves the target from the spatial image data by constructing a visual recognition model, and performs real-time target positioning analysis based on the recognition result; the linkage decision module identifies the target positioning data according to the visual recognition model, records the real-time target motion trajectory and performs real-time prediction and simulation of the target motion trajectory; performs abnormal probability distribution analysis based on the real-time prediction and simulation of the target motion trajectory, and determines the drone linkage model based on the analysis result; and plans drone decisions based on the drone linkage model.

2. The real-time monitoring system for an intelligent drone according to claim 1, characterized in that: The environmental perception module includes an operating scene environmental data acquisition unit and a flight trajectory formulation unit; The operation scene environment data acquisition unit acquires the safe operation range in the operation scene and performs real-time environment data acquisition based on the safe operation range; it realizes environment data acquisition by combining an environment acquisition drone with a sensor device; The flight trajectory formulation unit determines the operating parameter settings of the UAV according to the real-time operating scene environment data, and formulates the operating inspection route of the UAV.

3. The real-time monitoring system for an intelligent drone according to claim 2, characterized in that: The safe operating range is the safe hemispherical space when equipment or personnel are operating in the operating scene; the operating inspection route refers to the inspection flight route of the drone within the safe space in the operating scene; interference detection is performed on the inspection flight route by the drone.

4. The real-time monitoring system for an intelligent drone according to claim 3, characterized in that: The intelligent positioning control module includes a visual retrieval unit and a positioning analysis unit; The visual retrieval unit collects real-time spatial image data along the inspection route of the UAV and transmits it remotely to the monitoring terminal; the monitoring terminal performs continuous time target recognition on the spatial image data by building a visual recognition model; The positioning analysis unit performs positioning analysis on the target recognition results of the real-time spatial image data on the UAV operation inspection route according to the visual recognition model; if the recognition result is that there is no target, the flight positioning data of the UAV is continuously recorded and transmitted to the monitoring end; if the recognition result is that there is a target, the positioning of the UAV and the target are recorded at the same time, and the spatial relative position between the UAV and the target is determined; wherein, the spatial relative position is determined by spatially dotting the UAV and the target respectively, taking the spatial circumscribed spheres of the UAV and the target respectively, taking the centers of the spheres as the spatial positioning nodes of the UAV and the target respectively, and connecting the two nodes to obtain the spatial relative position data between the UAV and the target; the positioning record of the UAV and the target and the spatial relative position data are transmitted to the monitoring end.

5. The real-time monitoring system for an intelligent drone according to claim 4, characterized in that: The specific steps of performing continuous time target recognition on spatial image data by constructing a visual recognition model are as follows: By filtering the scene background pixels of the real-time spatial image data collected by the UAV, dynamic flying targets in the spatial image data are extracted; By performing pixel decomposition on the dynamic flying target image data, the contour pixel points of the dynamic flying target image are obtained, and by performing fusion processing on the contour pixel points, adjacent contour pixel points are connected respectively to determine the slope value of each pixel point connection segment; by performing difference calculation on the slopes between adjacent pixel point connection segments, by setting a slope judgment threshold, if the absolute value of the slope difference between adjacent pixel point connection segments is less than or equal to the slope judgment threshold, the pixel points on the adjacent pixel point connection segment are screened out; otherwise, the pixel points are retained; thereby obtaining the contour posture image data of the dynamic flying target; The remaining contour pixel points of the dynamic flight target image data are labeled according to the fusion processing results of the dynamic flight target image contour pixel points; the posture image data of the dynamic flight target in the continuous time-space image data are mapped to the same coordinate system, the spatial position change data of the contour pixel points with the same label in the continuous time-space image data are recorded, and the motion trajectory of the pixel points with the same label in the continuous time is obtained by connecting the lines; by comparing the curvature values ​​of the continuous time motion trajectories of adjacent labeled contour pixel points, combined with the contour posture image data of the dynamic flight target in the continuous time, a matching search is performed in the database to determine the target.

6. The real-time monitoring system for an intelligent drone according to claim 5, characterized in that: The linkage decision module includes a target trajectory tracking and analysis unit and a linkage decision unit; The target trajectory tracking and analysis unit determines the target's real-time motion trajectory based on the target's real-time positioning data and the recorded target positioning data; and predicts and simulates the target's motion trajectory based on the target's real-time motion trajectory; The linkage decision-making unit performs an abnormal analysis of the operation scene based on the real-time motion trajectory prediction simulation, divides the safe operation range of the operation scene into levels, determines the probability distribution analysis of the abnormal impact of the target real-time motion trajectory prediction simulation generated route on different levels of space in the safe operation range of the operation scene, and constructs a drone linkage model based on the analysis results to determine the decision.

7. The real-time monitoring system for an intelligent drone according to claim 6, characterized in that: The specific steps of predicting and simulating the target's motion trajectory based on the target's real-time motion trajectory are as follows: By collecting positioning data for the target, recording its movement trajectory, and using the target spatial positioning point recorded at the current real-time time point as the starting node of the predicted route, the movement trajectory of the target within the subsequent prediction period is predicted; the prediction period is the time required for the drone with the shortest relative distance to the target spatial position at the current time point to reach the target spatial position node; Obtain the corresponding spatial node tangent of the target's moving trajectory and record the corresponding spatial node tangent angles respectively; The maximum value A of the tangent angle corresponding to each spatial node on the target's motion path is determined; the tangent angle is the angle between the tangent of the corresponding spatial node and the horizontal line; the extreme points of the target's motion trajectory are determined, and the vertical distance D between the maximum maximum point and the minimum minimum point is calculated, with D being the fluctuation distance of the motion trajectory; at the starting node of the predicted route, [0, A] is used as the tangent angle variation interval, and [0, D] is used as the fluctuation distance interval of the predicted motion trajectory, and a full traversal is performed to generate a random motion trajectory route.

8. The real-time monitoring system for an intelligent drone according to claim 7, characterized in that: The safe operation range of the operation scene is divided into three levels: the center of the range is determined within the safe operation range in the operation scene, and the first safety distance, the second safety distance and the third safety distance are respectively planned as the sphere radius with the center of the range as the center of the sphere, and the hemispherical space of the safe operation range is divided into the first safety layer, the second safety distance and the third safety layer; wherein the first safety distance is greater than the second safety distance, and the second safety distance is greater than the third safety distance; Combined with the results of the hierarchical division of the safe operation range of the operation scene, the probability distribution analysis of the abnormal impact of different levels of space in the safe operation range of the operation scene is carried out based on the random motion trajectory routes generated by the starting node of the predicted route where the real-time target is located; by coordinating the total number of random motion trajectory routes generated by the starting node of the predicted route where the real-time target is located, and coordinating the distribution of different levels of space in the safe operation range of the operation scene involved in each generated random motion trajectory route; calculate the proportion of the number of random motion trajectory routes involving the third safety layer, the proportion of the number of random motion trajectory routes involving the second safety layer, and the proportion of the number of random motion trajectory routes involving the first safety layer at the current time point; compare the corresponding proportion thresholds of each level respectively, determine the UAV linkage model, and determine the decisions as conventional level decision, first level decision, second level decision and third level decision respectively.

9. The real-time monitoring system for an intelligent unmanned aerial vehicle according to claim 8, characterized in that: The data feedback module includes an alert unit and an event recording unit; When the warning unit makes a decision on the drone in the operation scene, it sends a warning prompt to the monitoring terminal; The event recording unit logs the drone decision events that exist in the operation scene.

Images