A method and system for earth task scheduling

Abstract

The application discloses a kind of ground task scheduling method and system, it is related to unmanned aerial vehicle scheduling technical field, the method includes: according to task demand, construct multiple functional units;According to task information and the performance parameter of each unmanned aerial vehicle, unmanned aerial vehicle is grouped, and it is allocated task;For unmanned aerial vehicle group, plan task air route;Definition state node, determine the conversion condition and execution condition between state node;Based on the conversion condition and scheduling relationship of state node, corresponding functional unit and unmanned aerial vehicle group are scheduled;The functional unit of being scheduled controls unmanned aerial vehicle group according to the execution condition of state node, executes task air route to implement corresponding task.Solve the task allocation rule of existing unmanned aerial vehicle scheduling method Single, the collaborative mechanism between different tasks is not perfect Problem.Can improve the flexibility and adaptability of task planning, ensure that unmanned aerial vehicle always executes task with optimal state, improve the coherence and high efficiency of different task transition.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) scheduling technology, and in particular to a method and system for scheduling ground missions. Background Technology

[0002] With the rapid development of drone technology, its applications are becoming increasingly widespread in various fields. Drones can undertake tasks such as information collection, surveillance, target tracking, geographic mapping, power line inspection, and agricultural plant protection. However, drones face many challenges when performing ground-based missions.

[0003] Traditional mission planning often relies on pre-set, fixed rules and simple experience-based judgments. When grouping drones, classification may be based solely on model or quantity, without fully considering actual performance differences such as endurance and payload accuracy. This leads to inefficient drone resource allocation in complex mission scenarios, impacting mission execution efficiency. Furthermore, the algorithms involved in different mission phases lack effective coordination mechanisms during drone mission execution. For example, in information gathering and target tracking missions, if the information gathering task fails to promptly and accurately transmit target information to the target tracking task, delays or errors may occur. This is because existing systems have inadequate data interaction and synchronization mechanisms. When drones encounter unexpected situations, such as communication interruptions due to electronic interference or sudden equipment failures, existing systems often struggle to respond quickly and effectively. The inability to adjust mission plans and flight paths in a short time can result in mission failure or drone loss. Summary of the Invention

[0004] This application provides a method and system for scheduling ground-based tasks, which solves the problems of existing UAV scheduling methods having simple task allocation rules and imperfect coordination mechanisms between different tasks.

[0005] In a first aspect, embodiments of this application provide a ground mission scheduling method, comprising: constructing multiple functional units according to mission requirements and setting an identification identifier for each functional unit; grouping UAVs according to mission information and performance parameters of each UAV and assigning them tasks; planning mission routes for the corresponding UAV groups according to the assigned tasks; defining state nodes according to the operating status of the flight control system and determining the transition conditions and execution conditions between each state node; wherein, the state nodes and the corresponding functional units establish a scheduling relationship through the identification identifier; scheduling the corresponding functional units and UAV groups based on the transition conditions of the state nodes in the current operating state and the scheduling relationship; and the scheduled functional units control the UAV groups to execute the mission routes to implement the corresponding tasks according to the execution conditions of the state nodes.

[0006] In conjunction with the first aspect, in one possible implementation, the step of grouping UAVs according to task information and the performance parameters of each UAV, and assigning tasks to them, includes: normalizing the requirements of each type of task in the task information and the performance parameters of each UAV to obtain requirement standard values ​​and performance standard values; assigning performance weights to the performance standard values ​​according to the task type; determining the matching degree between different types of tasks and each UAV based on the performance weights, the requirement standard values, and the performance standard values; grouping each UAV according to the matching degree and the matching thresholds for different task types; and determining the tasks assigned to each UAV group based on the grouping results of the UAVs.

[0007] In conjunction with the first aspect, in one possible implementation, the step of planning mission routes for the corresponding UAV group based on the assigned task includes: dividing the threat probability of the target into a threat heatmap based on prior information; wherein the prior information is updated based on real-time data collected by the UAV during the mission; gridding the mission area, determining the threat level of each grid based on the threat heatmap, and determining the terrain information of each grid; clustering based on the threat level and the terrain information to obtain multiple sub-regions and regional constraints for different sub-regions; performing 3D modeling and rasterization of the mission area to obtain an environment model; converting the performance parameters of the UAV group into performance constraints, determining the actual cost of each grid in the environment model based on the regional constraints and the performance constraints, embedding the actual cost into the corresponding grid of the environment model to obtain a cost map; and planning mission routes for the UAV group based on the cost map and the target distribution.

[0008] In conjunction with the first aspect, in one possible implementation, the step of planning a mission route for the UAV group based on the cost map and target distribution includes: determining the heuristic value of each node on the cost map based on the actual cost; performing a route exploration step with the UAV's position as the current node until a stopping condition is met, backtracking the explored node list to obtain the mission route; wherein, the route exploration step includes: establishing a search window centered on the current node, and periodically updating the actual cost of the grid within the search window at a first frequency; determining the reachable neighbor nodes of the current node, and determining the comprehensive cost of each reachable neighbor node, and evaluating each reachable neighbor node: if the reachable neighbor node of the current node is not in the node list to be explored or has already been explored... In the node table, the current node is set as the parent node of the reachable neighbor node, and the reachable neighbor node is added to the node list to be explored. If the reachable neighbor node is already in the node list to be explored, its cumulative cost is updated. If the updated cumulative cost is less than the original cumulative cost, the combined cost of the reachable neighbor node and its parent node are updated. If the reachable neighbor node is already in the explored node table, its cumulative cost is updated. If the updated cumulative cost is less than the original cumulative cost, the reachable neighbor node is moved to the node list to be explored. After evaluating all reachable neighbor nodes of the current node, the current node is removed from the node list to be explored, and the node with the lowest combined cost in the node list to be explored is taken as the current node, and the route exploration step is executed.

[0009] In conjunction with the first aspect, one possible implementation also includes: real-time monitoring of the current status of the drones; if any drone in the drone group malfunctions, the mission route is replanned; if a drone triggers the return-to-home condition, the drone plans its return-to-home route based on its current status.

[0010] In conjunction with the first aspect, in one possible implementation, if the UAV triggers the return-to-home condition, the UAV plans a return-to-home route based on its current state, including: if the UAV triggers the return-to-home condition, assessing the accessibility and safety of alternate landing points in the alternate landing point database based on the UAV's current battery level / remaining fuel, and determining a first number of candidate landing points; wherein, the return-to-home condition includes the current battery level / remaining fuel level being below a safety threshold, the UAV malfunctioning, or receiving a return-to-home command; planning a return-to-home route from the UAV's current position to the optimal candidate landing point, and alternative return-to-home routes from the current position to other candidate landing points; smoothing the return-to-home route and breaking it down into a waypoint list, controlling the UAV's return-to-home through the waypoint list, and adjusting the waypoint list in real time during the return-to-home process.

[0011] Secondly, embodiments of this application provide a ground mission scheduling system, characterized in that it includes: a scheduling module configured to: establish a shared memory based on collected prior information, mission information, and performance parameters of each UAV; receive real-time data from the UAV during the mission and update the shared memory; define state nodes according to the operating status of the flight control system and determine the transition conditions and execution conditions between each state node; and perform full-process scheduling of the UAV's mission cycle through the transition conditions and the execution conditions; and a structural design module configured to: analyze mission requirements, construct multiple mutually cooperating functional units according to the mission requirements, and set an identification identifier for each functional unit; and establish a scheduling relationship between the identification identifier and the corresponding state node. The system includes a pre-planning module configured to: group drones according to task information and the performance parameters of each drone; assign tasks to drone groups based on the matching degree between the drone groups and task requirements; plan task routes for each drone group based on prior information; and send the drone grouping information, task assignment information, and task routes to the task module. The task module is configured to: determine the tasks for different drone groups and, based on the transition conditions of the current state node and the scheduling relationship, schedule the corresponding functional units and drone groups; control the drone groups through the scheduled functional units to execute the task routes according to the execution conditions of the current state node to implement the corresponding tasks; and send a replanning request to the pre-planning module if any drone in the drone group malfunctions.

[0012] In conjunction with the second aspect, in one possible implementation, the functional units include a low-altitude calibration unit, an avoidance unit, an information acquisition unit, and a target tracking unit. The low-altitude calibration unit is configured to: acquire real-time terrain data from different UAV groups during the mission, and fuse the real-time terrain data with a pre-stored high-precision digital elevation model for calibration; control the UAV groups to fly close to the ground based on the calibrated high-precision digital elevation model. The avoidance unit is configured to: analyze the target's monitoring attitude in real time and trigger an avoidance algorithm accordingly. The information acquisition unit is configured to: load prior information about the mission area from the pre-planning module, determine the search range for each UAV based on the prior information; determine a search strategy based on the search range of each UAV and target distribution characteristics; perform a search based on the search strategy and periodically update the search coverage; adjust the search strategy according to the search coverage. The target tracking unit is configured to: determine the comprehensive priority of each target, determine the mission target based on the comprehensive priority; determine mission parameters based on the UAV's performance parameters and the real-time information of the mission target, and lock onto the mission target.

[0013] In conjunction with the second aspect, in one possible implementation, the pre-planning module includes a grouping unit, a task allocation unit, a region division unit, a flight path planning unit, and a communication unit. The grouping unit is configured to: determine the matching degree between different types of tasks and each UAV based on the requirements of each type of task in the task information and the performance parameters of each UAV; and group the UAVs according to the matching degree and matching thresholds for different task types. The task allocation unit is configured to: determine the tasks assigned to each UAV group based on the grouping results. The region division unit is configured to: divide the task area into multiple sub-regions using prior information and determine the regional constraints for different sub-regions. The flight path planning unit is configured to: generate the task flight path for each UAV group using a path planning algorithm based on the performance parameters of each UAV and the regional constraints. The communication unit is configured to: send the grouping information from the grouping unit, the task allocation information from the task allocation unit, and the flight path information from the flight path planning unit to the task module.

[0014] In conjunction with the second aspect, one possible implementation also includes a return-to-home module; the return-to-home module is configured to: monitor in real time whether the UAV triggers a return-to-home condition; when the return-to-home condition is triggered, generate a return-to-home route based on the current state of the UAV; and synchronize the UAV's return-to-home information to the ground station; wherein, the return-to-home module includes a monitoring unit, a planning unit, and a coordination unit; the monitoring unit is configured to: monitor the current state of the UAV in real time; if the current state triggers a return-to-home condition, send the current state to the planning unit; wherein, the return-to-home condition includes the current battery / remaining fuel being below a safety threshold, the UAV malfunctioning, or receiving a return-to-home command. The planning unit is configured to: assess the accessibility and safety of alternate landing points in the alternate landing point database based on the UAV's current battery level / remaining fuel, and determine a first number of candidate landing points; plan the return route from the UAV's current position to the optimal candidate landing point, and alternative return routes from the current position to other candidate landing points; after smoothing the return route, decompose it into a waypoint list, control the UAV's return via the waypoint list, and adjust the waypoint list in real time during the return process; the coordination unit is configured to: synchronize the UAV's return information to the ground station; wherein, the return information includes the return type, landing point, and estimated landing time.

[0015] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0016] This application's embodiments construct multiple functional units based on task requirements, facilitating individual maintenance and upgrades of each unit, avoiding difficulties in fault location or high upgrade risks, and improving development efficiency. Grouping drones according to task information and performance parameters, and assigning tasks to them, avoids resource waste and improves task execution efficiency and drone safety. Determining the transition and execution conditions between various state nodes clarifies an efficient collaborative scheduling mechanism between different tasks, ensuring the continuity and efficiency of task execution. This allows tasks such as information collection and tracking to work closely together, reducing command latency, improving data flow efficiency within the system, and preventing disconnections and delays during task execution. This effectively solves the problems of existing drone scheduling methods, such as simplistic task allocation rules and imperfect collaborative mechanisms between different tasks. Furthermore, it improves the flexibility and adaptability of task planning, ensuring drones always execute tasks in optimal condition and enhancing the continuity and efficiency of transitions between different tasks. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart of a ground mission scheduling method provided in this application embodiment;

[0019] Figure 2 This is a system architecture diagram of a ground mission scheduling system provided in an embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0021] The following description of some technologies involved in the embodiments of this application is provided to aid understanding and should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, some descriptions of well-known functions and structures are omitted in the following description.

[0022] Figure 1This is a flowchart of a ground mission scheduling method provided in an embodiment of this application, including steps 101 to 106. Figure 1 This is merely one execution order shown in the embodiments of this application and does not represent the only execution order of a ground-based task scheduling method. The execution order can be adjusted to achieve the desired final result. Figure 1 The steps shown can be performed in parallel or in reverse order.

[0023] Step 101: Construct multiple functional units according to task requirements and assign identification tags to each functional unit. In this embodiment, the task types include low-altitude calibration tasks, information acquisition tasks, target tracking tasks, and composite tasks. Multiple functional units are constructed for different types of tasks, including low-altitude calibration units, avoidance units, information acquisition units, target tracking units, etc.

[0024] Specifically, the low-altitude calibration unit is configured to: collect real-time terrain data from different UAV groups during the mission, and fuse and calibrate the real-time terrain data with a pre-stored high-precision digital elevation model. The UAV groups are then controlled to fly close to the ground based on the calibrated high-precision digital elevation model.

[0025] The avoidance unit is configured to analyze the target's monitored attitude in real time and trigger the avoidance algorithm accordingly.

[0026] The information acquisition unit is configured to: load prior information about the task area from the pre-planning module; determine the search range for each UAV based on the prior information; determine a search strategy based on the search range and target distribution characteristics of each UAV; perform a search based on the search strategy and periodically update the search coverage; and adjust the search strategy according to the search coverage.

[0027] The target tracking unit is configured to: determine the overall priority of each target, and determine the mission target based on the overall priority; determine mission parameters based on the UAV's performance parameters and real-time information of the mission target, and lock onto the mission target.

[0028] Furthermore, an identification identifier is assigned to each functional unit. For example, the identification identifier for the low-altitude calibration unit is R_DK, and the identification identifier for the information acquisition unit is R_CJ.

[0029] It should be noted that those skilled in the art can also construct more detailed functional units based on the inventive concept of this application and the needs of the mission process. For example, a data receiving unit can also be constructed, configured to receive real-time data collected during the UAV mission and flight control data from the flight control system.

[0030] Step 102: Group the drones according to the task information and the performance parameters of each drone, and assign tasks to them. In this embodiment, the requirements of each type of task in the task information and the performance parameters of each drone are normalized to obtain standard requirement values ​​and standard performance values. Performance weights are assigned to the standard performance values ​​according to the task type. Based on the performance weights, standard requirement values, and standard performance values, the matching degree between different types of tasks and each drone is determined. The drones are grouped according to the matching degree and the matching thresholds for different task types. Based on the grouping results of the drones, the tasks assigned to each drone group are determined.

[0031] Specifically, mission information includes mission type and target distribution, target type, mission radius, mission area, mission priority, and mission requirements for different types of missions. The UAV's performance parameters include endurance, flight speed, payload type, data acquisition resolution, detection range, lock-on range, and lock-on accuracy. Mission requirements indicate the performance parameter requirements of different types of missions for the UAV. For example, for a data acquisition mission with a mission radius of 50km, calculated at a minimum speed of 50km / h, the required endurance is at least 1 hour, the minimum speed requirement is 50km / h, the payload type requirement is 1, the required data acquisition resolution is a maximum resolution of 0.5m (the closer the distance, the higher the spatial resolution), the required detection range is 2km, the required lock-on range is 0, and the required lock-on accuracy is 0.

[0032] It should be noted that this application determines the payload type requirements based on the types of payloads required for different types of missions. Composite missions require the most payload types, with a payload type requirement of 4. Next are tracking missions, with a payload type requirement of 3. Following that are low-altitude calibration missions, with a payload type requirement of 2. Finally, information acquisition missions have a payload type requirement of 1.

[0033] In the payload type requirements, 0 indicates the basic payload that each UAV must be equipped with, including: navigation and guidance payloads, such as magnetic compasses, barometers, gyroscopes + accelerometers, and GNSS receivers; communication payloads, such as data radios or satellite communication terminals; and energy payloads, such as battery packs / fuel tanks and generators. 1 indicates that a specific type of payload is carried for the mission. For example, an information gathering mission requires a specific type of sensor for data collection, including electro-optical pods, synthetic aperture radar, and laser rangefinders. 2 indicates that two types of payloads are carried for the mission. For example, a low-altitude calibration mission requires a payload specifically for jamming (such as electronic jamming equipment and decoy systems) and a sensor specifically for data collection. The payload type requirements for the other values ​​are similar and will not be elaborated here.

[0034] Specifically, the task requirements of different types of tasks in the task information are normalized to [0,1] using the max-min normalization method to obtain the standard value of the requirements, and the performance parameters of each UAV are normalized to [0,1] to obtain the standard value of the performance.

[0035] Furthermore, taking the normalization of performance parameters as an example, we can use the max-min formula: This normalizes the performance parameters of different drones. Where, n i N represents the normalized standard performance value of the i-th performance parameter. i N represents the value of the i-th performance parameter. i-min N represents the minimum value of the i-th performance parameter. i-max This represents the maximum value of the i-th performance parameter.

[0036] For example, the performance parameters of different UAVs in this application are as follows: the maximum endurance is 10 hours, and the minimum is 2 hours; the maximum flight speed is 100 km / h, and the minimum is 50 km / h; the maximum payload type is 4, and the minimum is 1; the maximum information acquisition resolution is 0.5 m, and the minimum is 2 m; the maximum detection distance is 20 km, and the minimum is 5 km; the maximum lock distance is 15 km, and the minimum is 0 km; and the maximum lock accuracy is 10 km, and the minimum is 0 km.

[0037] Table 1 below shows the performance parameters of three different UAVs listed in this application and their normalized performance standard values.

[0038] Table 1 Performance parameters and standard performance values ​​of different UAVs

[0039]

[0040] Furthermore, the performance standard values ​​of the different UAVs mentioned above can also be represented in vector form, denoted as the performance vector, namely [performance standard value of endurance time, performance standard value of flight speed, performance standard value of payload type, performance standard value of information acquisition resolution, performance standard value of detection distance, performance standard value of locking distance, performance standard value of locking accuracy]. Taking UAV 1 as an example, its performance vector can be represented as: [1.0,0.6,0,1.0,0.6667,0,0].

[0041] Similarly, for an information collection task (task radius 50km, task area 100km²),... 2Taking the example of a data acquisition task, its endurance requirement is greater than 1 hour, which, after normalization, results in -0.125, a negative value that is unreasonable. Therefore, the task radius of the data acquisition task is used instead to normalize the endurance requirement. The maximum task radius is 100km, and the minimum is 0km, so the normalized task radius of the data acquisition task is 0.5. The flight speed requirement is 50km / h, which is normalized to 0. The payload type requirement is 1, which is normalized to 0. The data acquisition resolution requirement is the highest resolution of 0.5m, which is normalized to 1.0. The detection distance requirement is 2km, but the normalized result is negative, which is unreasonable. Therefore, the task area of ​​the data acquisition task is used instead to normalize the detection distance requirement, with a maximum task area of ​​200km. 2 The minimum value is 0, so the normalized area of ​​the task region is 0.5. The locking distance requirement is 0, which is 0 after normalization. The locking accuracy requirement is 0, which is 0 after normalization.

[0042] In addition, the demand standard values ​​for each type of task can be represented as a vector, denoted as the demand vector. The demand vector for the information collection task can be represented as: [0.5,0,0,1.0,0.5,0,0].

[0043] By normalizing the various task requirements in the task information with the performance parameters of each UAV, the matching degree between UAV performance and task requirements can be quantified. This allows each UAV to perform the task it is best at, avoiding the waste of resources caused by traditional experience-based grouping and improving task reliability.

[0044] Furthermore, for the information acquisition task, based on the importance of different performance parameters to completing the task, the performance weights assigned to the performance standard values ​​of endurance, flight speed, payload type, information acquisition resolution, detection range, lock-on range, and lock-on accuracy are 0.25, 0.15, 0.2, 0.2, 0.2, 0, 0, respectively. Representing these performance weights as a vector, denoted as the weight vector, the weight vector for the information acquisition task can be expressed as: [0.25, 0.15, 0.2, 0.2, 0.2, 0, 0].

[0045] The matching degree between different types of tasks and various drones is calculated using a suitable formula. Taking information collection tasks as an example, the suitable formula for calculating the matching degree is as follows:

[0046] ,

[0047] In the formula, M K,ZC U represents the degree of matching between the information collection task and the drone k. k W represents the performance vector of drone k. ZC T represents the weight vector of the information collection task.ZC This represents the demand vector for the information collection task.

[0048] For example, the matching threshold for the information collection task is 0.8. Therefore, drones with a matching degree greater than or equal to 0.8 can be grouped together (denoted as the information collection group) to perform the information collection task. Based on the above appropriate formula, the matching degrees of drone 1, drone 2, and drone 3 relative to the information collection task are calculated to be 0.921, 0.546, and 0.790, respectively. Therefore, drone 1 is assigned to the information collection group to perform the information collection task.

[0049] Furthermore, if a drone has a match score greater than 0.8 with multiple task types, it will be assigned to the group corresponding to the task with the highest match score. If a drone has an equal match score with multiple task types, and all of them are greater than 0.8, then the drone will be assigned to the group corresponding to the task with the highest task priority, based on task priority.

[0050] Those skilled in the art can calculate the matching degree between other types of tasks and the performance parameters of other UAVs based on the above method, which will not be described in detail here.

[0051] Step 103: Plan mission routes for the corresponding UAV group based on the assigned tasks. In this embodiment, the threat probability of the target is divided into a threat heatmap based on prior information. The prior information is updated based on real-time data collected by the UAVs during the mission. The mission area is gridded, and the threat level of each grid is determined based on the threat heatmap, as well as the terrain information of each grid. Clustering is performed based on the threat level and terrain information to obtain multiple sub-regions and regional constraints for different sub-regions. A 3D model of the mission area is created and rasterized to obtain an environment model. The performance parameters of the UAV group are converted into performance constraints. Based on the regional constraints and performance constraints, the actual cost of each grid in the environment model is determined, and the actual cost is embedded into the corresponding grid in the environment model to obtain a cost map. Mission routes are planned for the UAV group based on the cost map and target distribution.

[0052] Specifically, a tracking heatmap is created based on the tracking probability of the target's tracking device. Further, at a distance of 0-5km from the tracking device, with a tracking probability of 90%-100%, the area is dark red, representing an absolute no-go zone; at 5-10km, with a tracking probability of 70%-90%, the area is red, representing a high-threat zone; at 10-20km, with a tracking probability of 40%-70%, the area is orange, representing a threat zone; at 20-40km, with a tracking probability of 10%-40%, the area is yellow, representing a warning zone; and at distances above 40km, with a tracking probability of 0%-10%, the area is green, representing a safe zone. The task area is then gridded, and the slope, elevation, and other data for each grid cell are calculated.

[0053] Radar heatmaps are categorized based on the radar's detection probability of the target. Specifically, at distances of 0-50km, the detection probability is 95%-100%, indicated by a deep red color, signifying an extremely high probability of precise tracking; at 50-100km, the probability is 80%-95%, indicated by a red color, signifying a very high probability of detection; at 100-200km, the probability is 50%-80%, indicated by an orange color, signifying a relatively high probability of detection; at 200-400km, the probability is 20%-50%, indicated by a yellow color, signifying a relatively low probability of detection; and above 400km, the probability is 0%-20%, indicated by a green color, signifying that the target is outside the detection range.

[0054] The mean of the tracking probability and the detection probability is used as the threat probability to construct a threat heatmap. In the threat heatmap, dark red represents a threat level of 5, red represents a threat level of 4, orange represents a threat level of 3, yellow represents a threat level of 2, and green represents a threat level of 1.

[0055] For example, the task area is divided into 500m A threat heatmap is overlaid on the grid of the task area using a 500m grid. The threat level of each grid cell is determined by its color. Based on prior environmental information, terrain data such as slope and elevation are determined for each grid cell. The K-means clustering algorithm is used to cluster the grid cells according to "threat level + terrain complexity." Grid cells within the same cluster are then merged to obtain multiple sub-regions. For example, mountainous areas with a threat level of 3-4 and a slope greater than 25° are merged into a high-risk mountainous sub-region. Simultaneously, regional constraints for each sub-region are determined based on the terrain information and threat level of the grid cells.

[0056] Furthermore, a three-dimensional model was constructed based on prior information, and a 10m [model / scale] was used. 10m The 3D model is rasterized using a 5m grid to obtain the environment model. The actual cost of each grid cell is calculated using the following formula:

[0057] ,

[0058] In the formula, C(x,y,z) represents the actual cost at grid (x,y,z). This indicates the threat constraint weight, set to 0.4. This indicates the terrain constraint weight, set to 0.3. The interference constraint weight is set to 0.1. T(x,y) represents the threat level at ground level (x,y), G(x,y,z) represents the terrain elevation and / or canyon width constraint at grid level (x,y,z), E(Z) represents the electromagnetic interference at height z, and B(h, E) represents the performance constraint, and h represents the vertical displacement between adjacent grid cells. This represents the horizontal rotation angle between adjacent grid cells, and E represents the reference energy consumption of a single grid cell. The performance constraint weight is set to 0.2, where x represents longitude, y represents latitude, and z represents altitude.

[0059] For example, for one of the drones, the performance constraints include: the maximum climb rate is less than or equal to 5° / s, that is, the vertical displacement between adjacent grids is less than or equal to 0.87m; the minimum turning radius is greater than or equal to 500m, that is, the horizontal turning angle between adjacent grids is less than or equal to 1.15°; and the endurance constraint is that the total energy consumption is less than or equal to 70% of the battery capacity, that is, the baseline energy consumption of a single grid is equal to the product of the drone's energy consumption coefficient, the drone's total weight, the flight speed and the horizontal distance of the grid.

[0060] The actual cost at each grid cell is embedded into the corresponding grid cell to obtain a cost map.

[0061] In this embodiment, planning a mission route for an unmanned aerial vehicle (UAV) group based on a cost map and target distribution includes: determining the heuristic value of each node on the cost map based on the actual cost; performing route exploration steps with the UAV's position as the current node until a stopping condition is met, and backtracking the explored node list to obtain the mission route. The route exploration steps include: establishing a search window centered on the current node and periodically updating the actual cost of the grid within the search window at a first frequency; determining the reachable neighbor nodes of the current node and the comprehensive cost of each reachable neighbor node; evaluating each reachable neighbor node: if the reachable neighbor node of the current node is not in the node list to be explored or the node list to be explored, then setting the current node as the parent node of that reachable neighbor node and adding that reachable neighbor node to the node list to be explored; if the reachable neighbor node is already in the node list to be explored, then updating its cumulative cost; if the updated cumulative cost is less than the original cumulative cost, then updating the comprehensive cost and parent node of the reachable neighbor node. If a reachable neighbor node is already in the explored node list, its cumulative cost is updated. If the updated cumulative cost is less than the original cumulative cost, the reachable neighbor node is moved to the unexplored node list. After evaluating all reachable neighbors of the current node, the current node is removed from the unexplored node list, and the node with the lowest overall cost in the unexplored node list is selected as the current node. The route exploration step is then executed.

[0062] Specifically, the heuristic value for each node is calculated using a multi-objective heuristic function. The formula for calculating the heuristic value is as follows:

[0063] ,

[0064] In the formula, h(n) represents the heuristic value, D(n,goal) represents the Euclidean distance from the current node n to the target point goal, and ds represents the integral element of the route. This represents the cumulative energy consumption from the current node n to the target point (goal), which is the sum of the baseline energy consumption of each grid cell. E(s) represents the energy consumption per unit length along the flight path s, calculated based on the product of the UAV's energy consumption coefficient, total UAV weight, flight speed, and flight path distance. The energy consumption coefficient is the reciprocal of the air density at the UAV's altitude. C(x,y,z) represents the actual cost at grid cell (x,y,z). 1. 2. 3 indicates dynamic weight, whose value changes dynamically according to different scenarios.

[0065] For example: when C(x,y,z) At 50 degrees Celsius, you are in a danger zone and need to prioritize avoiding danger. 1. 2. 3 represents 0.3, 0.5, and 0.2 respectively; when C(x,y,z) At 30:00, we are in a safe zone; prioritize improving task efficiency. 1. 2. The values ​​are 0.6, 0.2, and 0.2 respectively. When the remaining battery power is less than or equal to 30% of the battery capacity, it is in a low-power state and energy saving should be prioritized. 1. 2. The values ​​for 3 are 0.2, 0.2, and 0.6 respectively.

[0066] The calculation formula for the heuristic value mentioned above can take into account short distance, low threat, and low energy consumption, thereby improving the mission success rate and the safety of the drone.

[0067] Specifically, based on the actual cost and heuristic value of each grid, a route exploration step is performed to explore the optimal mission route from the starting point (the location of the UAV) to the target point.

[0068] Furthermore, the prior information includes information about fixed targets (tracking equipment, radar stations) (coordinates, tracking radius, detection range), while the real-time data collected by the UAV during the mission may include information about temporary targets (newly added temporary tracking equipment, mobile radar). These temporary targets also threaten the safety of our UAV and the mission completion rate. Therefore, this application establishes a search window centered on the current node and periodically updates the actual cost of the grids within the search window at a first frequency. The tracking and detection probabilities of temporary targets are also periodically updated into the cost map. The actual cost of each grid is recalculated using the formula for calculating the actual cost of the grid, thus completing the update. Furthermore, updating only the actual cost of the grids within the search window each time improves computational efficiency, shortens update time, and ensures the timeliness of data updates.

[0069] For example, this application establishes a 500m [node / location] centered on the current node (the location of the drone). A search window of 500m is used. The first frequency is exemplarily set to 200ms. Those skilled in the art will recognize that the above-described search window size and first frequency value are merely one embodiment of this application; the search window size for high-threat sub-regions can also be set to 300m. 300m, while reducing the value of the first frequency to improve security, and setting the search window size for low-threat sub-regions to 800m. 800m, and at the same time increase the value of the first frequency to improve task efficiency.

[0070] Furthermore, regional characteristics are fully considered when conducting route exploration, and different exploration strategies are adopted. Specifically, if in a high-risk mountainous sub-region of the cost map (slope greater than 25°, threat level 3-4), small-step exploration is conducted, that is, exploring one grid at a time, while using local spiral search to find hidden passages to avoid terrain collisions. If in a low-threat open sub-region of the cost map (slope less than 5°, threat level less than 2), large-step exploration is conducted, that is, exploring 10 grids at a time, to accelerate progress towards the target.

[0071] Initialization: Create two tables: the node to be explored table and the node already explored table. Start exploring from the starting point (the current node where the first route exploration step is executed, i.e., the drone's position). The cumulative cost of the starting point (the cumulative cost equals the sum of the actual costs from the starting point to the current node) is 0. At this time, the node to be explored table contains only the starting point node, and the node already explored table is empty.

[0072] Based on the performance and regional constraints of the UAV, the reachability of the current node's eight neighboring nodes is determined to identify reachable neighborhood nodes. Reachable neighboring nodes satisfy the grid conditions. Topographic elevation and / or canyon width constraints , grid The actual cost of the place Nodes with a horizontal rotation angle between adjacent grid cells less than or equal to 1.15° and a vertical displacement between adjacent grid cells less than or equal to 0.87m are considered as neighboring nodes. The comprehensive cost of each reachable neighbor node is calculated (the comprehensive cost equals the sum of the corresponding cumulative cost and the heuristic value). Then, each reachable neighbor node is evaluated as follows: If the current node's reachable neighbor node is not in the list of nodes to be explored or the list of explored nodes, the current node is set as the parent node of that reachable neighbor node, and the reachable neighbor node is added to the list of nodes to be explored. If the reachable neighbor node is already in the list of nodes to be explored, its cumulative cost is updated. If the updated cumulative cost is less than the original cumulative cost, the comprehensive cost of the reachable neighbor node and its parent node are updated. If the reachable neighbor node is already in the list of explored nodes, its cumulative cost is updated. If the updated cumulative cost is less than the original cumulative cost, the reachable neighbor node is moved to the list of nodes to be explored.

[0073] After evaluating all reachable neighboring nodes of the current node, remove the current node from the list of nodes to be explored, and select the node with the lowest overall cost in the list of nodes to be explored as the current node, and execute the route exploration step.

[0074] The route exploration steps are iteratively executed according to the above steps until a stopping condition is met. For example, the stopping condition is that the current node coincides with the target point. After the stopping condition is met, the list of explored nodes is backtracked to obtain a mission route. Sharp turns can also be eliminated through B-spline curve fitting to ensure that all segments meet the UAV's performance constraints, resulting in the final mission route.

[0075] In one embodiment of this application, the exploration progress of the route can be monitored in real time. If no new node with a lower overall cost than the current path is found after 50 consecutive route exploration steps, a "deadlock" occurs (i.e., the current path is trapped in a local optimum). At this time, the current node three steps back can be traced back, and the nodes after the traced-back node (denoted as the traced-back node) are removed from the list of explored nodes. Three branches with deflections of 15°, 30°, and 45° are extended from the traced-back node, with the length of the branch being the step size of one exploration. The cumulative cost, heuristic value, and overall cost of the initial node of each branch are calculated. Then, the initial node with the minimum overall cost is taken as the current starting point, and the route exploration steps are iteratively executed again until the stopping condition is met.

[0076] In addition, ant colony algorithms can be used to plan multiple backup mission routes for drones.

[0077] Step 104: Define state nodes based on the operating status of the flight control system, and determine the transition conditions and execution conditions between each state node. In this embodiment, the state nodes and their corresponding functional units establish a scheduling relationship through identification identifiers.

[0078] Specifically, the flight control system's operational states include: initialization state (initializing system variables, threads, collecting prior information, mission information, UAV performance parameters, etc., and storing them in shared memory), standby state (UAVs are grouped, tasks are assigned to each UAV group, initial mission information is obtained, and mission routes are planned), takeoff state (controlling UAV groups to take off according to the mission route), low-altitude flight state (performing low-altitude calibration tasks), data acquisition state (performing information acquisition tasks), tracking state (performing target tracking tasks), return-to-home state (triggering return-to-home conditions, generating return-to-home routes, and performing the return), and landing state (the UAV arrives above the landing point, i.e., altitude ≤ 100m, position error ≤ 200m, calls the landing algorithm to land, transmits mission data back after landing, and releases resources). The defined state nodes include initialization state nodes, standby state nodes, takeoff state nodes, low-altitude flight state nodes, data acquisition state nodes, tracking state nodes, return-to-home state nodes, and landing state nodes.

[0079] For example, the transition conditions and execution conditions between state nodes can be defined by those skilled in the art according to the actual situation. This application lists the following transition conditions and execution conditions for state nodes.

[0080] The transition from a standby node to a takeoff node is conditional upon receiving mission instructions and completing mission route planning. The execution condition is that the flight control system schedules the takeoff unit, which then sends takeoff instructions to the flight control computer of the corresponding UAV (or UAV group).

[0081] The transition from a low-altitude flight status node to a data acquisition status node is conditional upon reaching a position within 500m of the mission area boundary. The execution conditions are as follows: the flight control system schedules the information acquisition unit, which sends information acquisition commands to the flight control computer of the corresponding UAV (or UAV group), initiates the search algorithm, divides the mission area into multiple units, and assigns these units to different UAVs for searching according to the search strategy.

[0082] Furthermore, the takeoff, low-altitude flight, data acquisition, tracking, return-to-home, and landing status nodes can all be switched in emergencies. Specifically, when a threat level suddenly increases by 30%, the UAV is damaged, the deviation from the waypoint exceeds 500 meters, or communication is interrupted, the system can switch to an emergency status node. This emergency status node is an independent node, unrelated to the flight control system's operational status. It can be embedded in any status node except the initialization node, enabling rapid response to unforeseen circumstances, ensuring mission success, minimizing the impact of emergencies, and guaranteeing mission integrity and UAV safety.

[0083] Step 105: Based on the transition conditions and scheduling relationships of the current operating state's state nodes, schedule the corresponding functional units and UAV groups. In this embodiment, according to the current operating state's state nodes and corresponding scheduling relationships, schedule the corresponding functional units and UAV groups to execute the tasks corresponding to that operating state.

[0084] For example, if the current state is low-altitude flight, the low-altitude flight state node will schedule the corresponding low-altitude calibration unit, and the low-altitude calibration unit will control the UAV group (the UAV group assigned to the low-altitude calibration task) to perform the low-altitude calibration task.

[0085] Step 106: The scheduled functional unit controls the UAV group to execute the mission route according to the execution conditions of the status node to carry out the corresponding mission. In this embodiment, the scheduled functional unit controls the corresponding UAV group to execute the corresponding mission according to the mission route. During the mission, real-time data (flight control data, situational data, terrain data, etc.) is transmitted back. The flight control system updates the shared memory according to the received real-time data and adjusts the mission route according to the real-time data.

[0086] Furthermore, this application also allows for the following during the drone group's mission execution: real-time monitoring of the drones' current status; replanning the mission route if any drone in the group malfunctions; and replanning a return route based on the drone's current status if a drone triggers a return-to-home condition.

[0087] Specifically, if the drone triggers the return-to-home condition, the accessibility and safety of alternative landing points in the database are assessed based on the drone's current battery level and remaining fuel, determining a first set of candidate landing points. The return-to-home conditions include the current battery level / remaining fuel falling below a safety threshold, drone malfunction, and receiving a return-to-home command. A return route is planned from the drone's current position to the optimal candidate landing point, as well as alternative return routes from the current position to other candidate landing points. The return routes are smoothed and then broken down into a waypoint list. The drone's return is controlled through this waypoint list, which is adjusted in real-time during the return process.

[0088] In addition, the drone's return information can be synchronized to the ground station during the return process. This information includes the return type, landing point, and estimated landing time, so that technicians can handle the situation promptly.

[0089] For example, the return route can be implemented using ant colony optimization. The algorithm uses path planning algorithms such as [algorithm name missing] to plan the route. The current safe threshold for battery power is less than 30% of the battery capacity, and the safe threshold for remaining fuel is less than 20% of the remaining fuel. The return route and alternative return routes are smoothed using B-spline curves.

[0090] Furthermore, the accessibility and safety of alternate landing points in the alternate landing point database are evaluated using the coverage area of ​​the current battery / remaining fuel and the cumulative cost on the return route. If the coverage area of ​​the current battery / remaining fuel is greater than or equal to 95%, the alternate landing point has high accessibility. The cumulative cost on the return route from the high-accessibility alternate landing point to the UAV's current location is calculated. If the cumulative cost is less than or equal to 100, the alternate landing point is considered safe. The first quantity is exemplarily set to 3, meaning that 3 alternate landing points with the lowest cumulative cost are selected as candidate landing points from the safe alternate landing points.

[0091] In this embodiment, the complete task cycle from initialization to task planning, task execution, and finally recycling is realized through the mutual conversion between different state nodes. Each state node is implemented by a separate functional unit to meet its corresponding requirements, which can reduce the task interruption rate, improve reliability, and reduce maintenance costs.

[0092] The method described in this application can be used in scenarios such as film / advertising shooting, wildlife rescue, and power line maintenance. The task radius, task area, target, task objective, target priority, prior information, tracking heatmap, radar heatmap, grid size, grid size, task type, performance parameters, and requirements for different types of tasks vary depending on the scenario and can be set by those skilled in the art according to actual needs.

[0093] Taking wildlife rescue as an example, the target is a wild animal (population or individual), and the prior information is the environmental information of the wild animal's usual activity range (i.e., the area of ​​the mission area). The overall priority of the target is determined based on the wild animal's endangered status, mobility, and health status. For example, if an injured wild animal needs treatment, then that injured wild animal is the mission target. In the performance parameters of the UAV, the detection range refers to the distance at which the injured wild animal's population can be identified; the lock-on distance refers to the distance at which the injured wild animal can be identified within the population without being detected; and the lock-on accuracy refers to the distance at which the injured wild animal can be accurately tracked for a long time without being detected. The tracking heatmap in this scenario is a heatmap categorized according to the degree of interference to wild animals at different distances, while the radar heatmap is a heatmap based on the distance at which wild animals can perceive the anomalies brought by the UAV (without or with minimal interference to the wild animals). Grid size and raster size can be categorized based on the wild animal's habits, activity range, and environmental information, which will not be specifically explained here.

[0094] While this application provides method operation steps as shown in the embodiments or flowcharts, more or fewer operation steps may be included based on conventional or non-inventive labor. The order of steps listed in this embodiment is merely one possible execution order among many and does not represent the only execution order. In actual device or client product execution, the method can be executed sequentially according to this embodiment or the accompanying drawings, or in parallel (e.g., in a parallel processor or multi-threaded processing environment).

[0095] like Figure 2 As shown in the figure, this application embodiment also provides a ground mission scheduling system, which includes: a scheduling module, a structural design module, a pre-planning module, a mission module and a return module, as detailed below.

[0096] In this embodiment, the scheduling module is configured to: establish a shared memory based on the collected prior information, task information, and performance parameters of each UAV; receive real-time data from the UAV during the mission and update the shared memory; define state nodes based on the operating status of the flight control system and determine the transition conditions and execution conditions between each state node; and perform full-process scheduling of the UAV's mission cycle through the transition conditions and execution conditions.

[0097] Specifically, the scheduling module is the core controller of the system, operating throughout the entire task lifecycle and capable of updating shared memory in real time. This prevents task disconnects caused by disordered interaction between modules. Establishing shared memory eliminates the need for separate communication between modules, reducing data transmission latency to less than 50ms.

[0098] In this embodiment of the application, the structural design module is configured to: analyze task requirements, construct multiple mutually cooperating functional units according to the task requirements, and set an identification identifier for each functional unit; and establish a scheduling relationship between the identification identifier and the corresponding status node.

[0099] Specifically, the functional units include a low-altitude calibration unit, an avoidance unit, an information acquisition unit, and a target tracking unit.

[0100] The low-altitude calibration unit is configured to: collect real-time terrain data of different UAV groups during the mission, and fuse and calibrate the real-time terrain data with the pre-stored high-precision digital elevation model; control the UAV groups to fly close to the ground based on the calibrated high-precision digital elevation model.

[0101] The avoidance unit is configured to analyze the target's monitored attitude in real time and trigger the avoidance algorithm accordingly.

[0102] The information acquisition unit is configured to: load prior information of the task area in the pre-planning module; determine the search range of each UAV based on the prior information; determine the search strategy based on the search range and target distribution characteristics of each UAV; perform the search based on the search strategy and periodically update the search coverage; and adjust the search strategy according to the search coverage.

[0103] The target tracking unit is configured to: determine the overall priority of each target, determine the mission target based on the overall priority; determine the mission parameters based on the performance parameters of the UAV and the real-time information of the mission target, and lock onto the mission target.

[0104] In this embodiment, the pre-planning module is configured to: group the drones according to the task information and the performance parameters of each drone; assign tasks to the drone groups according to the matching degree between the drone groups and the task requirements; plan task routes for each drone group based on prior information; and send the drone grouping information, task assignment information, and task routes to the task module.

[0105] Specifically, the pre-planning module includes grouping units, task allocation units, area division units, route planning units, and communication units.

[0106] The grouping unit is configured to: determine the matching degree between different types of tasks and each UAV based on the requirements of each type of task in the task information and the performance parameters of each UAV; and group each UAV according to the matching degree and the matching threshold of different task types.

[0107] The task allocation unit is configured to determine the tasks to be assigned to each drone group based on the grouping results of the drones.

[0108] The region partitioning unit is configured to divide the task region into multiple sub-regions using prior information and determine the region constraints for different sub-regions.

[0109] The route planning unit is configured to generate the mission route for each UAV group using a path planning algorithm, based on the performance parameters of each UAV and regional constraints.

[0110] The communication unit is configured to send the grouping information of the grouping unit, the task allocation information of the task allocation unit, and the route planning information of the route planning unit to the task module.

[0111] In this embodiment of the application, the task module is configured to: determine the tasks of different UAV groups, and schedule the corresponding functional units and UAV groups based on the transition conditions and scheduling relationships of the current state node; control the UAV groups to execute the task route to implement the corresponding tasks according to the execution conditions of the current state node through the scheduled functional units; if any UAV in the UAV group malfunctions, send a replanning request to the preplanning module.

[0112] In this embodiment of the application, the return-to-home module is configured to: monitor in real time whether the UAV triggers the return-to-home condition; when the return-to-home condition is triggered, generate a return-to-home route based on the current state of the UAV; and synchronize the UAV's return-to-home information to the ground station.

[0113] Specifically, the return-to-home module includes a monitoring unit, a planning unit, and a coordination unit.

[0114] The monitoring unit is configured to monitor the current status of the drone in real time, and if the current status triggers a return-to-home condition, it will send the current status to the planning unit. The return-to-home conditions include the current battery / remaining fuel level falling below a safety threshold, a drone malfunction, or receiving a return-to-home command.

[0115] The planning unit is configured to: assess the accessibility and safety of alternate landing points in the alternate landing point database based on the drone's current battery level and remaining fuel, and determine a first number of candidate landing points; plan the return route from the drone's current position to the optimal candidate landing point, as well as alternative return routes from the current position to other candidate landing points; after smoothing the return route, decompose it into a waypoint list, control the drone's return via the waypoint list, and adjust the waypoint list in real time during the return process.

[0116] The coordination unit is configured to synchronize the UAV's return-home information to the ground station. This return-home information includes the return-home type, landing point, and estimated landing time.

[0117] Some modules in the apparatus described in this application can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0118] The apparatus or module described in the above embodiments can be implemented by a computer chip or physical entity, or by a product with a certain function. For ease of description, the above apparatus is described by dividing it into various modules according to their functions. When implementing the embodiments of this application, the functions of each module can be implemented in one or more software and / or hardware. Of course, a module that implements a certain function can also be implemented by combining multiple sub-modules or sub-units.

[0119] The methods, apparatus, or modules described in this application can be implemented in a computer-readable program code manner. The controller can be implemented in any suitable manner, such as a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of a memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code manner, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included within it for implementing various functions can also be considered as structures within the hardware component. Alternatively, the device used to implement various functions can be viewed as either a software module that implements the method or a structure within a hardware component.

[0120] This application also provides an apparatus, the apparatus comprising: a processor; a memory for storing processor-executable instructions; wherein, when the processor executes the executable instructions, it implements the method described in this application.

[0121] This application also provides a non-volatile computer-readable storage medium storing a computer program or instructions thereon, which, when executed, enables the method described in this application embodiment to be implemented.

[0122] Furthermore, in the various embodiments of the present invention, each functional module can be integrated into a processing module, or each module can exist independently, or two or more modules can be integrated into a single module.

[0123] The aforementioned storage media include, but are not limited to, Random Access Memory (RAM), Read-Only Memory (ROM), Cache, Hard Disk Drive (HDD), or Memory Card. The memory can be used to store computer program instructions.

[0124] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary hardware. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product, or it can be embodied in the process of data migration. The computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, mobile terminal, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0125] The various embodiments described in this specification are presented in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. All or part of this application can be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, mobile communication terminals, multiprocessor systems, microprocessor-based systems, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.

[0126] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of this application.

Claims

1. A method of earth task scheduling, characterized by, The method comprises the following steps: According to the task requirements, a plurality of functional units are constructed, and an identification mark is set for each functional unit; According to the task information and the performance parameters of each unmanned aerial vehicle, the unmanned aerial vehicles are grouped and assigned tasks; According to the assigned tasks, a task route is planned for the corresponding unmanned aerial vehicle group, including: converting the performance parameters of the unmanned aerial vehicle group into performance constraints, determining the actual cost of each grid in the environment model based on the regional constraint conditions and the performance constraints, and embedding the actual cost in the corresponding grid of the environment model to obtain a cost map; based on the cost map and the target distribution, a task route is planned for the unmanned aerial vehicle group; wherein the regional constraint conditions are determined according to the terrain information and threat level of each grid in the task area, and the threat level is determined according to the threat probability constructed by the mean value of the tracking probability and the detection probability; According to the running state of the flight control system, a state node is defined, and the transition conditions and execution conditions between each state node are determined; wherein the state node and the corresponding functional unit establish a scheduling relationship through the identification mark; Based on the transition conditions of the state node of the current running state and the scheduling relationship, the corresponding functional unit and the unmanned aerial vehicle group are scheduled; The scheduled functional unit controls the unmanned aerial vehicle group to execute the task route according to the execution conditions of the state node to implement the corresponding task.

2. The method of claim 1, wherein, The method of grouping and assigning tasks according to the task information and the performance parameters of each unmanned aerial vehicle comprises: The requirements of each type of task in the task information and the performance parameters of each unmanned aerial vehicle are normalized respectively to obtain demand standard values and performance standard values; According to the task type, a performance weight is assigned to the performance standard value; Based on the performance weight, the demand standard value and the performance standard value, the matching degree between different types of tasks and each unmanned aerial vehicle is determined; According to the matching degree and the matching threshold of different task types, each unmanned aerial vehicle is grouped; According to the grouping result of the unmanned aerial vehicle, the task assigned to each unmanned aerial vehicle group is determined.

3. The method of claim 1, wherein, The construction method of the environment model comprises: According to prior information, the threat probability of the target is divided into a threat heat map; wherein the prior information is updated according to the real-time data collected by the unmanned aerial vehicle during the task; The task area is gridded, the threat level of each grid is determined according to the threat heat map, and the terrain information of each grid is determined; According to the threat level and the terrain information, clustering is performed to obtain a plurality of sub-areas and regional constraint conditions of different sub-areas; The task area is three-dimensionally modeled and gridded to obtain an environment model.

4. The method of claim 3, wherein, The method of planning a task route for the unmanned aerial vehicle group based on the cost map and the target distribution comprises: Based on the actual cost, the heuristic value of each node on the cost map is determined; Taking the position of the unmanned aerial vehicle as the current node, a route exploration step is performed until the stop condition is reached, and the explored node table is backtracked to obtain a task route; The route exploration step comprises: A search window is established around the current node, and the actual cost of the grids in the search window is updated periodically at a first frequency; determining reachable neighborhood nodes of the current node and determining a comprehensive cost of each reachable neighborhood node, evaluating each reachable neighborhood node: if the reachable neighborhood node of the current node is not in the to-be-explored node table or the explored node table, setting the current node as a parent node of the reachable neighborhood node and adding the reachable neighborhood node to the to-be-explored node table; if the reachable neighborhood node is in the to-be-explored node table, updating the cumulative cost of the reachable neighborhood node, and if the updated cumulative cost is less than the updated cumulative cost, updating the comprehensive cost of the reachable neighborhood node and the parent node; if the reachable neighborhood node is in the explored node table, updating the cumulative cost of the reachable neighborhood node, and if the updated cumulative cost is less than the updated cumulative cost, moving the reachable neighborhood node to the to-be-explored node table; after evaluating all reachable neighborhood nodes of the current node, removing the current node from the to-be-explored node table, and taking the node with the minimum comprehensive cost in the to-be-explored node table as the current node, performing the route exploration step.

5. The method of claim 1, wherein, Further comprising: real-time monitoring of the current state of the unmanned aerial vehicle; if an unmanned aerial vehicle in the unmanned aerial vehicle group appears abnormal, re-planning the task route; if the unmanned aerial vehicle triggers a return condition, the unmanned aerial vehicle plans a return route according to the current state.

6. The method of claim 5, wherein, The if the unmanned aerial vehicle triggers a return condition, the unmanned aerial vehicle plans a return route according to the current state, comprising: if the unmanned aerial vehicle triggers a return condition, according to the current power / residual fuel of the unmanned aerial vehicle, evaluating the accessibility and safety of the emergency landing point in the emergency landing point database to determine a first number of candidate landing points; wherein the return condition includes the current power / residual fuel being lower than a safety threshold, the unmanned aerial vehicle malfunctioning, receiving a return instruction; planning a return route between the current position of the unmanned aerial vehicle and the optimal candidate landing point, and a backup return route between the current position and other candidate landing points; after smoothing the return route, disassembling it into a waypoint list to control the unmanned aerial vehicle to return, and adjusting the waypoint list in real time during the return process.

7. A ground mission scheduling system for implementing the method of any one of claims 1 to 6, characterized by Comprising: a scheduling module configured to: establish a shared memory according to the collected prior information, task information and performance parameters of each unmanned aerial vehicle; receive real-time data of the unmanned aerial vehicle during the task process and update the shared memory; define state nodes according to the running state of the flight control system and determine the transition conditions and execution conditions between each state node; and schedule the task cycle of the unmanned aerial vehicle through the transition conditions and execution conditions; a structure design module configured to: analyze task requirements, respectively construct multiple mutually cooperative functional units according to the task requirements, and set an identification mark for each functional unit; and establish a scheduling relationship between the identification mark and the corresponding state node; a pre-planning module configured to: group the unmanned aerial vehicles according to the task information and performance parameters of each unmanned aerial vehicle; assign tasks to the unmanned aerial vehicle group according to the matching degree of the unmanned aerial vehicle group and the task requirements; plan a task route for each unmanned aerial vehicle group based on prior information; and send the grouping information, task assignment information and task route of the unmanned aerial vehicle to the task module; The task module is configured to determine tasks of different groups of unmanned aerial vehicles, and schedule corresponding functional units and groups of unmanned aerial vehicles based on the conversion condition of the current state node and the scheduling relationship. The groups of unmanned aerial vehicles are controlled by the scheduled functional units to perform the task route according to the execution condition of the current state node to implement corresponding tasks, and if an unmanned aerial vehicle in the group of unmanned aerial vehicles is abnormal, a re-planning request is sent to the pre-planning module.

8. The system of claim 7, wherein, The functional units include a low-altitude calibration unit, an avoidance unit, an information acquisition unit and a target tracking unit. The low-altitude calibration unit is configured to acquire real-time terrain data of different groups of unmanned aerial vehicles during a task process, and fuse and calibrate the real-time terrain data with a pre-stored high-precision digital elevation model; and control the groups of unmanned aerial vehicles to fly close to the ground based on the calibrated high-precision digital elevation model. The avoidance unit is configured to analyze a monitoring posture of a target in real time, and timely trigger an avoidance algorithm. The information acquisition unit is configured to load prior information of a task area in the pre-planning module, determine a search range of each unmanned aerial vehicle based on the prior information, determine a search strategy based on the search range of each unmanned aerial vehicle and a target distribution feature; The target tracking unit is configured to determine a comprehensive priority of each target, determine a task target according to the comprehensive priority, determine a task parameter according to a performance parameter of the unmanned aerial vehicle and real-time information of the task target, and lock the task target. The pre-planning module includes a grouping unit, a task allocation unit, a region division unit, a route planning unit and a communication unit.

9. The system of claim 7, wherein, The grouping unit is configured to determine a matching degree between different types of tasks and each unmanned aerial vehicle according to a requirement of each type of task in the task information and the performance parameter of each unmanned aerial vehicle, and group each unmanned aerial vehicle according to the matching degree and a matching threshold of different task types. The task allocation unit is configured to determine a task allocated to each group of unmanned aerial vehicles according to a grouping result of the unmanned aerial vehicles. The region division unit is configured to divide a task region into a plurality of sub-regions by using prior information, and determine a region constraint condition of different sub-regions. The route planning unit is configured to generate a task route of each group of unmanned aerial vehicles by using a path planning algorithm according to the performance parameter of each unmanned aerial vehicle and the region constraint condition. The communication unit is configured to send grouping information of the grouping unit, task allocation information of the task allocation unit and route information of the route planning unit to the task module. The return module is further included.

10. The system of claim 7, wherein, The return module is configured to monitor whether a return condition is triggered in real time, generate a return route according to a current state of the unmanned aerial vehicle when the return condition is triggered, and synchronize return information of the unmanned aerial vehicle to a ground station. The return module includes a monitoring unit, a planning unit and a coordination unit. ​ The monitoring unit is configured to monitor the current state of the UAV in real time, and if the current state triggers a return condition, send the current state to the planning unit; wherein the return condition includes that the current power / residual fuel is lower than a safety threshold, the UAV has a fault, or a return instruction is received; The planning unit is configured to evaluate the accessibility and safety of the emergency landing points in the emergency landing point database according to the current power / residual fuel of the UAV, determine a first number of candidate landing points; plan a return route between the current position of the UAV and an optimal candidate landing point, and alternative return routes between the current position and other candidate landing points; after smoothing the return route, disassemble the return route into a waypoint list, control the UAV to return through the waypoint list, and adjust the waypoint list in real time during the return process; The coordination unit is configured to synchronize the return information of the UAV with the ground station; wherein the return information includes a return type, a landing point, and a predicted landing time.

Images