A low-altitude economic drone dispatching method and system based on big data

Abstract

The present invention relates to the field of data processing technology, and in particular to a low-altitude, economical drone scheduling method and system based on big data. The method comprises the following steps: determining a collision risk assessment function between the starting node and the destination node based on the visibility value at each moment from the starting node to the destination node; determining the drone's load sensitivity coefficient based on the drone's load offset and critical load offset; using the product of the drone's wind direction potential energy value and the load sensitivity coefficient at each moment as a resistance risk assessment function between the starting node and the destination node at each moment; and determining the selection probability of the starting node and destination node paths in an ant colony algorithm based on the distance between the starting node and the destination node, the resistance risk assessment function at each moment, and the collision risk assessment function, thereby achieving drone scheduling and effectively improving the accuracy of drone scheduling.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a low-altitude economical drone scheduling method and system based on big data. Background Art

[0002] Drone dispatch is a key component of the low-altitude economy. It allows drones to be assigned to different mission areas based on diverse mission requirements, such as logistics and food delivery. To ensure efficient and safe execution of drone missions in these different mission areas, optimal flight paths must be planned to effectively reduce flight time and energy consumption while also effectively avoiding potential collision risks.

[0003] In various delivery contexts, the primary goal of drone scheduling is to efficiently and safely deliver items to various nodes. Nodes are target locations that drones need to visit, and edges represent possible paths a drone can take from one node to another. Existing path planning algorithms typically use the path distance between nodes as a cost weight to plan the optimal path for the drone.

[0004] Existing path planning algorithms have the limitation of static environment assumptions. They usually assume that weather conditions (such as visibility, wind speed, wind direction, etc.) are constant during mission execution. However, when visibility decreases, wind direction or wind speed suddenly changes, or the drone load shifts on the path, it will cause interference with the drone operation, ultimately affecting the path cost, reducing the accuracy of path planning, and posing a greater risk to the safe operation of the drone.

[0005] Therefore, when dynamically planning the operation path of a drone, how to comprehensively consider the static path length and the dynamically changing cost factors on the path, so as to accurately plan the operation route of the drone, is a problem that needs to be solved urgently. Summary of the Invention

[0006] In order to solve the technical problem of how to comprehensively consider the static path length and the dynamically changing cost factors on the path when dynamically planning the operation path of the drone, so as to accurately plan the operation route of the drone, the present invention provides a low-altitude economical drone scheduling method and system based on big data.

[0007] In a first aspect, the present invention provides a low-altitude economical drone dispatching method based on big data, which adopts the following technical solutions:

[0008] A low-altitude economical drone dispatching method based on big data, comprising the following steps:

[0009] Obtain the distance between drone nodes, obtain the visibility value, the angle between the drone flight direction and the wind direction, the wind speed, and the drone load offset and critical load offset at each moment; determine the collision risk assessment function between the starting node and the destination node based on the visibility value at each moment from the starting node to the destination node; determine the drone load sensitivity coefficient based on the drone load offset and critical load offset; calculate the drone's load sensitivity coefficient at the first Wind direction potential energy value at a moment :

[0010] ;

[0011] 、 Respectively The wind speed between the starting node and the destination node at each moment, and the angle between the UAV’s flight direction and the wind direction, is the cosine function, is the absolute value symbol; the product of the wind direction potential energy value and the load sensitivity coefficient of the UAV at each moment is used as the resistance risk assessment function between the starting node and the destination node at each moment; according to the distance between the starting node and the destination node, the resistance risk assessment function and the collision risk assessment function at each moment, the selection probability of the starting node and the destination node path in the ant colony algorithm is determined to realize UAV scheduling.

[0012] The present invention pre-plans drone routes using an ant colony algorithm, effectively improving the accuracy and efficiency of drones in achieving low-altitude economy, thereby increasing resource utilization. During the pre-planning process, the present invention considers that the probability of the path between each node being selected as the optimal path by the ant colony algorithm depends on the path length and does not account for the dynamically changing meteorological conditions along the path. Therefore, by analyzing direct and indirect meteorological conditions, the present invention constructs a dynamic collision risk assessment function and a resistance risk assessment function, which can accurately assess the degree to which drones are affected by meteorological conditions. Based on the dynamic collision risk assessment function and the resistance risk assessment function, the distance heuristic term used to determine selection probability in the traditional ant colony algorithm is expanded into a multi-objective weighted function, effectively improving the accuracy of drone scheduling paths and thereby effectively reducing the collision risk of drones in complex weather conditions.

[0013] According to a low-altitude economic drone scheduling method based on big data provided by the present invention, the method of obtaining the distance between drone nodes also includes: drawing a drone scheduling diagram based on the position coordinates of the drone's starting point, transfer point and destination, recording the departure position of the drone as the starting node, and the transfer point or destination to be reached as the destination node, and collecting the wind direction, wind speed, visibility and building density between the nodes at each moment within a preset time.

[0014] By comprehensively acquiring various parameters in the meteorological parameters, the present invention can accurately evaluate the meteorological interference that the UAV may be subject to during flight from all aspects, thereby achieving pre-emptive avoidance and reducing safety hazards.

[0015] According to a low-altitude economic drone scheduling method based on big data provided by the present invention, the collision risk assessment function between the starting node and the destination node is determined, including: if there are no other nodes between the starting node and the destination node, the collision risk assessment function between the starting node and the destination node is determined based on the visibility value at each moment from the starting node to the destination node and the maximum visibility between all nodes; if there are other nodes between the starting node and the destination node, the collision risk assessment function between the starting node and the destination node is calculated based on the building density between the starting node and the destination node, the visibility value at each moment, the maximum visibility between all nodes, the distance between the intersecting nodes on the perpendicular bisector of the drone and the tangent of its running trajectory, and the visibility value.

[0016] The present invention takes into account that the collision risk generated when there are other nodes between the starting node and the destination node is different, so the relevant parameters are analyzed separately according to the actual situation to more accurately evaluate the collision risk between the starting node and the destination node.

[0017] According to a big data-based low-altitude economical drone scheduling method provided by the present invention, the collision risk assessment function between the starting node and the destination node is determined based on the visibility value at each moment from the starting node to the destination node and the maximum visibility between all nodes, including:

[0018] ;

[0019] is the collision risk assessment function between the starting node and the destination node, is the total number of moments, 、 They are the maximum visibility value between the starting node and the destination node within the preset time, The visibility value at a moment, is the building density within the preset range from the starting node to the destination node. is the normalization function, is an exponential function with base e.

[0020] According to a low-altitude economical drone scheduling method based on big data provided by the present invention, the calculation of the collision risk assessment function between the starting node and the destination node includes:

[0021] ; is the collision risk assessment function between the starting node and the destination node, 、 They are the first The distance and visibility value between the UAV and the i-th intersection node on the perpendicular bisector of the tangent of its trajectory at the moment, 、 are the maximum distance and visibility value between the UAV and all its intersecting nodes at all times between the starting node and the destination node, respectively. 、 are the number of nodes and building density within the preset range from the starting node to the destination node, is the number of intersecting nodes, is an exponential function with base e.

[0022] The present invention takes into account the increased risk of collision when other nodes exist between the starting node and the destination node. Therefore, it provides a method for calculating the collision risk assessment function between the starting node and the destination node when other nodes exist. By accurately analyzing the distance and visibility between the drone and its intersecting nodes, and combining the surrounding building density, the collision risk assessment function can be accurately obtained.

[0023] According to a low-altitude economic UAV dispatching method based on big data provided by the present invention, the UAV load sensitivity coefficient is determined according to the UAV load offset and critical load offset, including:

[0024] ;

[0025] 、 Respectively The load sensitivity coefficient and load offset of the UAV at each moment, is the critical load offset, The preset load index.

[0026] According to a low-altitude economical drone scheduling method based on big data provided by the present invention, the method of determining the selection probability of the path of the starting node and the destination node in the ant colony algorithm includes: taking the maximum value of the resistance risk assessment function at all times between the starting node and the destination node as the target resistance risk assessment function between the starting node and the destination node; and calculating the selection probability of the path of the starting node and the destination node:

[0027] ;

[0028] is the selection probability of the path from the i-th starting node to the j-th destination node, 、 、 are the distance from the i-th starting node to the j-th destination node, the collision risk assessment function, and the target resistance risk assessment function, respectively.

[0029] According to a low-altitude economic drone scheduling method based on big data provided by the present invention, the selection probability of the starting node and destination node paths in the ant colony algorithm is determined to achieve drone scheduling, including: taking the destination node as the new starting node, and obtaining the destination node of the new starting node in the drone scheduling graph, calculating the selection probability, and finally obtaining the selection probability among all nodes; determining the target path of the drone scheduling according to the selection probability among all nodes to achieve drone scheduling.

[0030] According to a low-altitude economic drone scheduling method based on big data provided by the present invention, the drone scheduling is implemented, and then further includes: allocating resources to each node according to the frequency of each node being selected as the target path for drone scheduling.

[0031] The present invention takes into account that the more frequently a node is used as a node on the target path of each drone, the more times it will be used, and the greater the possibility that the drone will rest here. Therefore, when allocating resources, more resources are allocated to such nodes, thereby effectively improving resource utilization.

[0032] In a second aspect, the present invention provides a low-altitude economical drone dispatching system based on big data, which adopts the following technical solutions:

[0033] A low-altitude economic drone scheduling system based on big data includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned low-altitude economic drone scheduling method based on big data is implemented.

[0034] By adopting the above technical solution, the above-mentioned low-altitude economic drone scheduling method based on big data is generated into a computer program and stored in the memory to be loaded and executed by the processor, so that a terminal device is made according to the memory and the processor for easy use.

[0035] The present invention has the following technical effects:

[0036] Based on the above technical solution, the present invention provides a method and system for scheduling drones for low-altitude economy based on big data. By pre-planning drone routes using an ant colony algorithm, the method can effectively improve the accuracy and efficiency of drones in achieving low-altitude economy, thereby improving resource utilization. During the pre-planning of routes, the present invention considers that the probability of the path between each node being selected as the optimal path by the ant colony algorithm depends on the path length and does not take into account the dynamically changing meteorological conditions along the path. Therefore, the present invention constructs a dynamic collision risk assessment function and a resistance risk assessment function by analyzing direct and indirect meteorological conditions, which can accurately assess the degree to which drones are affected by meteorological conditions. Based on the dynamic collision risk assessment function and the resistance risk assessment function, the distance heuristic term used to determine selection probability in the traditional ant colony algorithm is expanded into a multi-objective weighted function, which can effectively improve the accuracy of drone scheduling routes and thus effectively reduce the collision risk of drones in complex weather conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A flowchart of a low-altitude economical drone dispatching method based on big data provided by an embodiment of the present invention;

[0038] Figure 2 This is an exploded schematic diagram of a downwind wind direction provided by an embodiment of the present invention;

[0039] Figure 3 It is a decomposed schematic diagram of an upwind wind direction provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0041] An embodiment of the present invention discloses a low-altitude economical drone scheduling method based on big data. Before the drone performs delivery, the method seeks the optimal path for drone scheduling through an ant colony algorithm. According to the optimal path, the accuracy of drone scheduling can be effectively improved, thereby effectively improving resource utilization.

[0042] For details, please see Figure 1 As shown, Figure 1 A flowchart of a method for dispatching low-altitude economic drones based on big data provided by an embodiment of the present invention is provided. The method specifically includes the following steps:

[0043] S1: Obtain the distance between drone nodes and building density, and obtain the visibility value, the angle between the drone and the wind direction, the wind speed, and the drone load offset and critical load offset at each moment.

[0044] It's important to note that when planning a drone's route using the ant colony algorithm, in addition to route length, a comprehensive assessment of meteorological conditions along the route is required. Heavy fog or rain can degrade the performance of a drone's visual sensors and reduce visibility, minimizing the likelihood of selecting such a route. The wind angle is the angle between the drone's flight direction and the wind direction. Headwinds can increase the drone's flight resistance and significantly increase energy consumption. Strong winds with excessive speeds can shift the center of gravity of a loaded drone, affecting its flight balance. Excessive payload, combined with wind direction and speed, can affect a drone's flight stability and increase resistance. The greater the building density, the higher the risk of collisions under various meteorological conditions.

[0045] Therefore, when obtaining the probability of each node path being selected as the optimal path, the embodiment of the present invention can obtain the visibility value, distance, building density, wind direction, wind speed, and drone load offset of the path to evaluate the risk level of the path, among which wind direction, wind speed, and visibility value belong to direct meteorological conditions, and distance, building density, and drone load offset belong to indirect meteorological conditions.

[0046] Because low-altitude, economical drones typically fly at high speeds, analyzing the building density around them can construct a preset range with a radius of one kilometer, centered around the node path, to obtain the building density and number of nodes within the preset range. The preset range can be set based on the drone's flight speed and is not subject to further limitations in this embodiment.

[0047] For example, in an embodiment of the present invention, a drone scheduling diagram can be drawn based on the location coordinates of the drone's starting point, transfer point, and destination. The location where the drone departs is recorded as the starting node, and the transfer point or destination to be reached is recorded as the destination node. The wind direction, wind speed, visibility, and building density between the nodes are collected at each moment within a preset time period.

[0048] The preset duration can be set to 30 minutes, and the collection frequency can be set to every 10 seconds. After collecting all parameters, the parameters of each dimension can be normalized to eliminate the dimensional differences between data of different dimensions.

[0049] It can be understood that a drone node is a location node that a drone can select during drone scheduling, and the entire drone scheduling map is the range within which the drone can operate. GPS can be used to obtain the coordinates of each location, thereby obtaining the distance between nodes and the building density to draw a drone scheduling map for each drone. In a drone scheduling map, the straight-line distance between two nodes can be used as the drone path between the two nodes. This is because drones typically fly long distances and need to avoid buildings as much as possible. Furthermore, within a shorter range, the differences in environmental meteorological conditions are relatively small. Therefore, this embodiment of the present invention abstracts the path between nodes as a straight line to analyze its likelihood of being selected as the optimal path.

[0050] Specifically, the API interface can be used to access the meteorological bureau's website to obtain wind direction, wind speed, and visibility values ​​at different times between nodes. Based on the drone model, its baseline payload and critical payload offset are obtained, and the difference between the drone's actual payload and the baseline payload is used as the drone's payload offset.

[0051] It is understood that among the various parameters obtained between drone nodes, visibility and building density affect the drone's dynamic collision risk, while wind speed, wind direction, and payload offset affect the drone's drag risk. Therefore, when analyzing the impact of meteorological characteristics between nodes on path selection probability, embodiments of the present invention can assess dynamic collision risk using visibility and analyze drag risk using wind speed, wind direction, and payload offset, i.e., perform the following steps S2 and S3, respectively.

[0052] S2: Determine a collision risk assessment function between the starting node and the destination node based on the visibility value at each moment from the starting node to the destination node.

[0053] It is understandable that the path from any node to the next node can be considered the starting node and destination node. During the process of a drone traveling from the starting node to the destination node, other nodes, such as transfer stations, may exist in the drone scheduling graph. These transfer station nodes are more likely to contain drones, meaning that the number of drones flying simultaneously in the same area may exceed one, increasing the risk of collision. Therefore, when evaluating the collision risk between the starting node and the destination node, embodiments of the present invention can perform separate analyses based on whether other nodes exist between the starting node and the destination node.

[0054] For example, in an embodiment of the present invention, determining a collision risk assessment function between a start node and a destination node specifically includes the following two possible situations:

[0055] In one possible scenario, there are no other nodes between the starting node and the destination node. In this case, a collision risk assessment function between the starting node and the destination node can be determined based on the visibility value at each moment in the distance between the starting node and the destination node and the maximum visibility value among all nodes.

[0056] Specifically, in an embodiment of the present invention, a collision risk assessment function between a starting node and a destination node is determined based on the visibility value at each moment from the starting node to the destination node and the maximum visibility value between all nodes, including:

[0057] ;

[0058] is the collision risk assessment function between the starting node and the destination node, is the total number of moments, is the maximum visibility value between the starting node and the destination node within the preset time. The first node between the starting node and the destination node within the preset time The visibility value at a moment, is the building density within the preset range from the starting node to the destination node. is the normalization function, is an exponential function with base e.

[0059] In the above formula, is the average difference between the maximum visibility value between the starting node and the destination node and the visibility value at each moment within the preset time period, The smaller it is, the closer the visibility value at that moment is to the maximum visibility value, the higher the visibility between the starting node and the destination node, and the lower the risk of collision.

[0060] is the credibility of the above difference mean, if Smaller in size, The smaller the value, the smaller the risk of collision. Therefore, the embodiment of the present invention further verifies the accuracy of the collision risk assessment function by setting the credibility of the difference mean, thereby improving the accuracy of the collision risk assessment function.

[0061] In summary, if the maximum visibility value between the starting node and the destination node within the preset time is higher, and the maximum visibility value between the starting node and the destination node within the preset time is higher, the The smaller the difference between the visibility value at each moment and the maximum visibility value, the higher the visibility of the UAV on the path between the starting node and the destination node, and the lower the possibility of collision.

[0062] In another possible scenario, there are other nodes between the starting node and the destination node. In this case, a collision risk assessment function can be calculated between the starting node and the destination node based on the building density between the starting node and the destination node, the visibility value at each moment, the maximum visibility between all nodes, and the distance and visibility value between the intersection nodes on the perpendicular bisector of the tangent of the drone and its trajectory.

[0063] It's understandable that when planning a drone's path, the drone's operating state is simulated. Therefore, the intersection nodes on the perpendicular bisector of the tangent line between the drone and its trajectory are the nodes on the perpendicular bisector of the line between the starting node and the destination node in the drone scheduling diagram that intersect with the perpendicular bisector. A greater number of intersecting nodes indicates a greater number of transit points along the route from the starting node to the destination node, and a higher likelihood of a drone appearing.

[0064] For example, when obtaining the intersection node, the perpendicular bisector of the straight line between the starting node and the destination node can be translated from the starting node to the destination node according to the flight speed of the UAV. In the range of the UAV scheduling diagram, all nodes that intersect with the perpendicular bisector during the translation process are used as the number of intersection nodes between the starting node and the destination node, and the distance from the straight line to the intersection node is used as the distance between the intersection node on the perpendicular bisector of the UAV and the tangent of its running trajectory.

[0065] Specifically, in an embodiment of the present invention, calculating a collision risk assessment function between a starting node and a destination node includes:

[0066] ;

[0067] is the collision risk assessment function between the starting node and the destination node, The number of nodes between the starting node and the destination node The distance between the UAV and the i-th intersection node on the perpendicular bisector of the tangent of its trajectory at the moment, The number of nodes between the starting node and the destination node The visibility value between the UAV and the i-th intersection node on the perpendicular bisector of the tangent of its trajectory at the moment, is the maximum distance between the UAV and all its intersecting nodes at all times between the starting node and the destination node, is the maximum visibility value of the drone and all its intersecting nodes at all times between the starting node and the destination node, is the number of nodes within the preset range from the starting node to the destination node, is the building density within the preset range from the starting node to the destination node. is the number of intersecting nodes, is an exponential function with base e.

[0068] In the above formula, It is the first node between the starting node and the destination node. The normalized value of the distance between the UAV and the i-th intersection node on the perpendicular bisector of the tangent of its trajectory at the moment. The larger the value, the greater the distance between the starting node and the destination node. The greater the distance between the UAV and the i-th intersection node on the perpendicular bisector of the tangent of its trajectory at a certain moment, the smaller the possibility of a UAV collision event.

[0069] It is the first node between the starting node and the destination node. The visibility value between the UAV and the i-th intersection node on the perpendicular bisector of the tangent of its trajectory at each moment is normalized. The larger the value, the higher the visibility value between the UAV and the i-th intersection node, and the machine vision module of the UAV can operate normally, so the possibility of collision risk will be reduced accordingly.

[0070] The higher the number of nodes within the preset range from the starting node to the destination node, the more intersecting nodes, the lower the visibility and the greater the building density, the higher the risk of drone collision, and the higher the corresponding collision risk assessment function.

[0071] The values ​​in brackets are normalized.

[0072] Based on the above steps, a collision risk assessment function between the starting node and the destination node can be obtained. The higher the value, the higher the risk of collision between the starting node and the destination node.

[0073] S3: Determine the load sensitivity coefficient of the UAV based on the load offset and critical load offset of the UAV; use the product of the wind direction potential energy value of the UAV at each moment and the load sensitivity coefficient as the resistance risk assessment function between the starting node and the destination node at each moment.

[0074] It should be noted that the direction of the drone from the starting node to the destination node is the flight direction of the drone. The wind direction encountered by the drone during flight mainly includes two types, namely tailwind and headwind. The wind direction data can be decomposed. Among them, the tailwind can be decomposed into the tailwind component and crosswind components, see Figure 2 As shown, Figure 2 This is a schematic diagram of a downwind wind direction provided by an embodiment of the present invention, Figure 2 It can be seen that the tailwind blows from the rear of the drone, forming an angle with the flight direction of the drone. , , can be decomposed into tailwind component in the horizontal direction and crosswind component in the vertical direction. The tailwind component can reduce the resistance of the UAV and improve the operation efficiency. The headwind can be decomposed into headwind component and crosswind component. For details, please refer to Figure 3 As shown, Figure 3 This is a schematic diagram of an upwind wind direction provided by an embodiment of the present invention. Figure 3 It can be seen that the headwind blows from the front of the drone, forming an angle with the flight direction of the drone. , , which is decomposed into headwind components in the horizontal direction and crosswind components in the vertical direction. The headwind components can increase the resistance of the UAV and reduce the operating efficiency of the UAV.

[0075] Therefore, the embodiment of the present invention can obtain the horizontal component of the angle between the drone's flight direction and the wind direction, combined with the wind speed and the drone's load, to jointly evaluate the resistance risk assessment function between the starting node and the destination node at each moment.

[0076] For example, in an embodiment of the present invention, determining the load sensitivity coefficient of the drone based on the load offset and the critical load offset of the drone includes:

[0077] ;

[0078] For the The load sensitivity coefficient of the UAV at this moment, For the The load offset of the UAV at a certain moment, is the critical load offset, The preset load index.

[0079] The load index is used to adjust the influence of the load on the wind direction assessment and can be set to 1.2. It can be set according to actual needs.

[0080] In the above formula, if The higher the UAV's load offset is above the critical load offset at a certain moment, the higher the load of the UAV is, which seriously affects the normal operation of the UAV. In this case, the flight speed of the UAV will be severely reduced, and the corresponding resistance risk assessment function will also be larger.

[0081] For example, in an embodiment of the present invention, calculating the wind direction potential energy value of the drone at each moment includes:

[0082] ;

[0083] For drones The wind direction potential energy value at a moment, For the The wind speed between the starting node and the destination node at the moment, For the The angle between the UAV’s flight direction and the wind direction from the starting node to the destination node at that moment, is the cosine function, To belong to the symbol, in is a closed interval, is an open interval, is the absolute value symbol.

[0084] In the above formula, when , explain the The moment is in the downwind direction, Reflects the strength of the tailwind component. Close to The greater the tailwind component, the The closer to the minimum value, the lower the corresponding resistance risk.

[0085] when , explain the The moment is in the upwind direction, Reflects the strength of the headwind component. The closer it is to 1, the more headwind begins to affect flight. The larger the value, the greater the impact of headwind on the drone. The bigger it is, the greater the risk of resistance.

[0086] According to the above steps, the resistance risk assessment function between the starting node and the destination node at each moment can be obtained based on the product of the wind direction potential energy value and the load sensitivity coefficient at each moment. By combining the collision risk assessment function and the resistance risk assessment function, the selection probability of the starting node and destination node path can be accurately obtained, that is, continue to execute the following steps.

[0087] S4: According to the distance between the starting node and the destination node, the resistance risk assessment function and the collision risk assessment function at each moment, the selection probability of the starting node and the destination node path in the ant colony algorithm is determined to realize drone scheduling.

[0088] It can be understood that there are resistance risk assessment functions at multiple moments between the starting node and the destination node. The embodiment of the present invention can obtain the maximum value of the resistance risk assessment function as the resistance risk assessment function of the path. The larger the maximum value of the resistance risk assessment function, the lower the possibility that this path can be used as the flight path of the drone.

[0089] For example, in an embodiment of the present invention, determining the selection probability of the path of the starting node and the destination node in the ant colony algorithm includes: taking the maximum value of the resistance risk assessment function at all times between the starting node and the destination node as the target resistance risk assessment function between the starting node and the destination node; and calculating the selection probability of the path of the starting node and the destination node.

[0090] For example, in an embodiment of the present invention, calculating the selection probability of the path between the start node and the destination node includes:

[0091] ;

[0092] is the selection probability of the path from the i-th starting node to the j-th destination node, is the distance from the i-th starting node to the j-th destination node, is the collision risk assessment function for the path from the i-th starting node to the j-th destination node, is the target resistance risk assessment function for the path from the i-th starting node to the j-th destination node.

[0093] In the above formula, the longer the path from the i-th starting node to the j-th destination node, and the larger its collision risk assessment function and target resistance risk assessment function, the higher the possibility of collision and large resistance on this long path, which is not suitable for UAV flight scheduling, and therefore the smaller the corresponding selection probability.

[0094] For example, in an embodiment of the present invention, the selection probability of the starting node and the destination node path in the ant colony algorithm is determined to realize drone scheduling, including: taking the destination node as the new starting node, and obtaining the destination node of the new starting node in the drone scheduling graph, calculating the selection probability, and finally obtaining the selection probability between all nodes; determining the target path of the drone scheduling according to the selection probability between all nodes to realize drone scheduling.

[0095] Specifically, the target path of drone scheduling is determined based on the selection probability between all nodes. When implementing drone scheduling, the number of ants in the ant colony algorithm and the initial pheromone on the path can be preset, and the selection probability between all starting nodes and destination nodes can be calculated; after the ants arrive at the starting node, they make random selections based on the selection probability between the starting node and the destination node and the initial pheromone, determine their forward route, and update the pheromone concentration on the route of the starting node and the destination node; subsequent ants choose the forward route based on the pheromone concentration on the route of the starting node and the destination node and the selection probability, and finally obtain the optimal target path.

[0096] The number of ants in the ant colony algorithm and the initial pheromone on the path can be set according to actual needs. The specific steps of the ant colony algorithm determining the target path for drone scheduling based on the selection probability between all nodes can be implemented according to existing technologies and will not be described in detail in this embodiment of the present invention.

[0097] It is understandable that for the nodes selected as the target path, there may be a large number of drones passing through them, and the possibility of drones stopping there is also high. Therefore, more backup resources need to be allocated to such nodes so that drones can replenish relevant materials or energy in a timely manner.

[0098] For example, in an embodiment of the present invention, drone scheduling is implemented, and then the method further includes: allocating resources to each node according to the frequency at which each node is selected as a target path for drone scheduling.

[0099] It can be seen that in the embodiment of the present invention, when implementing the scheduling of drones, the distance between drone nodes can be obtained, and the visibility value, the angle between the drone's flight direction and the wind direction, the wind speed, and the drone's load offset and critical load offset can be obtained at each moment; according to the visibility value at each moment from the starting node to the destination node, the collision risk assessment function between the starting node and the destination node is determined; according to the drone's load offset and critical load offset, the drone's load sensitivity coefficient is determined; the drone's load sensitivity coefficient at the first moment is calculated. Wind direction potential energy value at a moment :

[0100] ;

[0101] 、 Respectively The wind speed between the starting node and the destination node at each moment, and the angle between the UAV’s flight direction and the wind direction, is the cosine function, is the absolute value symbol; the product of the wind direction potential energy value of the UAV at each moment and the load sensitivity coefficient is used as the resistance risk assessment function between the starting node and the destination node at each moment; according to the distance between the starting node and the destination node, the resistance risk assessment function at each moment and the collision risk assessment function, the selection probability of the starting node and the destination node path in the ant colony algorithm is determined to realize UAV scheduling, which effectively improves the accuracy of UAV scheduling.

[0102] An embodiment of the present invention also discloses a low-altitude economic drone scheduling system based on big data, including a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, a low-altitude economic drone scheduling method based on big data provided by the present invention is implemented.

[0103] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0104] In the present invention, the aforementioned memory may be any tangible medium that contains or stores a program, which may be used by or in combination with an instruction execution system, apparatus, or device.

[0105] The above are all preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A low-altitude economic drone dispatching method based on big data, characterized in that: include: Obtain the distance between drone nodes, and obtain the visibility value, the angle between the drone's flight direction and the wind direction, the wind speed, and the drone's load offset and critical load offset at each moment; According to the visibility value at each moment from the starting node to the destination node, a collision risk assessment function between the starting node and the destination node is determined, including: If there is no other node between the starting node and the destination node, the collision risk assessment function between the starting node and the destination node is determined according to the visibility value at each moment from the starting node to the destination node and the maximum visibility between all nodes; If there are other nodes between the starting node and the destination node, the collision risk assessment function between the starting node and the destination node is calculated based on the building density between the starting node and the destination node, the visibility value at each moment, the maximum visibility between all nodes, and the distance and visibility value between the intersection nodes on the perpendicular bisector of the tangent of the drone and its trajectory; According to the UAV's load offset and critical load offset, the UAV's load sensitivity coefficient is determined, including: ； 、 Respectively The load sensitivity coefficient and load offset of the UAV at each moment, is the critical load offset, is the preset load index; Computing drones in Wind direction potential energy value at a moment : ； 、 Respectively The wind speed between the starting node and the destination node at each moment, and the angle between the UAV’s flight direction and the wind direction, is the cosine function, is the absolute value symbol; The product of the UAV's wind potential energy value and load sensitivity coefficient at each moment is used as the resistance risk assessment function between the starting node and the destination node at each moment; According to the distance between the starting node and the destination node, the resistance risk assessment function and the collision risk assessment function at each moment, the selection probability of the starting node and the destination node path in the ant colony algorithm is determined to realize drone scheduling.

2. The method for dispatching drones with low-altitude economy based on big data according to claim 1, characterized in that: The method of obtaining the distance between the drone nodes also includes: A drone scheduling diagram is drawn based on the location coordinates of the drone's starting point, transfer point, and destination. The departure location of the drone is recorded as the starting node, and the transfer point or destination to be reached is recorded as the destination node. The wind direction, wind speed, visibility, and building density between the nodes are collected at each moment within the preset time.

3. The method for dispatching drones with low-altitude economy based on big data according to claim 2, characterized in that: The method of determining a collision risk assessment function between the starting node and the destination node based on the visibility value at each moment from the starting node to the destination node and the maximum visibility value between all nodes includes: ； is the collision risk assessment function between the starting node and the destination node, is the total number of moments, 、 They are the maximum visibility value between the starting node and the destination node within the preset time, The visibility value at a moment, is the building density within the preset range from the starting node to the destination node. is the normalization function, is an exponential function with base e.

4. The method for dispatching drones with low-altitude economy based on big data according to claim 2, characterized in that: The calculation of the collision risk assessment function between the starting node and the destination node includes: ; is the collision risk assessment function between the starting node and the destination node, 、 They are the first The distance and visibility value between the UAV and the i-th intersection node on the perpendicular bisector of the tangent of its trajectory at the moment, 、 are the maximum distance and visibility value between the UAV and all its intersecting nodes at all times between the starting node and the destination node, respectively. 、 are the number of nodes and building density within the preset range from the starting node to the destination node, is the number of intersecting nodes, is an exponential function with base e.

5. The method for dispatching drones with low-altitude economy based on big data according to claim 1, characterized in that: The determining of the selection probability of the starting node and the destination node path in the ant colony algorithm includes: The resistance risk assessment function with the maximum value at all times between the starting node and the destination node is used as the target resistance risk assessment function between the starting node and the destination node; the selection probability of the path between the starting node and the destination node is calculated: ； is the selection probability of the path from the i-th starting node to the j-th destination node, 、 、 They are the distance from the i-th starting node to the j-th destination node, the collision risk assessment function, and the target resistance risk assessment function, respectively.

6. The method for dispatching drones with low-altitude economy based on big data according to claim 1, characterized in that: Determining the selection probability of the starting node and the destination node path in the ant colony algorithm to achieve drone scheduling includes: Take the destination node as the new starting node, obtain the destination node of the new starting node in the UAV scheduling graph, calculate the selection probability, and finally obtain the selection probability among all nodes; The target path of UAV scheduling is determined according to the selection probability among all nodes to realize UAV scheduling.

7. The method for dispatching drones with low-altitude economy based on big data according to claim 6, characterized in that: The implementation of drone scheduling further includes: Resources are allocated to each node according to the frequency with which each node is selected as the target path for UAV scheduling.

8. A low-altitude economic drone dispatching system based on big data, characterized by: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a low-altitude economic drone scheduling method based on big data according to any one of claims 1 to 7 is implemented.

Images