Unmanned aerial vehicle flight path optimization method and system based on artificial intelligence

Abstract

The invention discloses an unmanned aerial vehicle flight path optimization method and system based on artificial intelligence, and relates to the field of unmanned aerial vehicle flight control, and the method comprises a planning module which is used for uploading a landform model of an unmanned aerial vehicle flight area, and selecting a passing point of an unmanned aerial vehicle flight task in the landform model, so as to determine a flight task reference path; the construction module is used for receiving the flight mission reference path determined in the planning module and translating based on the flight mission reference path to construct a flight mission application path prototype; the flight reference path is determined according to the landform model of the flight area, the application path is formed through optimization, the flight path of the unmanned aerial vehicle better meets the actual environment requirement, in the flight process, the system can sense obstacles in real time, once the obstacles are found, the unmanned aerial vehicle can be accurately controlled to avoid in time, and the unmanned aerial vehicle can rapidly return to the original flight path after avoiding; and the collision risk is greatly reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle flight control, in particular to an unmanned aerial vehicle flight path optimization method and system based on artificial intelligence. BACKGROUND

[0002] Unmanned aerial vehicle flight path optimization is a technology that uses algorithms and techniques to plan efficient and safe flight routes for unmanned aerial vehicles in complex environments. It takes into account factors such as terrain, obstacles, and mission objectives, and uses algorithms such as AI and reinforcement learning to calculate the optimal path and dynamically adjust it based on real-time environmental changes. It is widely used in fields such as logistics distribution and surveying and mapping to improve the efficiency and safety of unmanned aerial vehicle operations.

[0003] The invention patent application with application number 202410716222.7 discloses an unmanned aerial vehicle flight path optimization method based on an improved grey wolf optimization algorithm. The method includes initializing the population size, population dimension, and maximum iteration number, and randomly generating an initial population according to the flight planning space and the central position of different mountain bodies. The grey wolf is initialized through the grey wolf initial model. The fitness value of each grey wolf is calculated according to the unmanned aerial vehicle path cost model function. The alpha wolf, beta wolf, and delta wolf are determined according to the fitness value of each grey wolf. The position of the grey wolf is updated according to the position of the alpha wolf, beta wolf, and delta wolf. The fitness value of the grey wolf after position update is calculated. The alpha wolf, beta wolf, and delta wolf are re-determined according to the fitness value of the updated grey wolf. If the current round reaches the maximum iteration number, the iteration is ended, the optimal individual is output, and the final unmanned aerial vehicle flight trajectory with the minimum fitness value is obtained. If the current round does not reach the maximum iteration number, the step of updating the position of the grey wolf according to the position of the alpha wolf, beta wolf, and delta wolf is performed. The application aims to solve the problem that the optimization of unmanned aerial vehicle path planning has been proven to be a NP-complete problem. The optimization algorithm is a widely used method to solve NP problems. The optimization algorithm can handle complex real-world scenarios that traditional modeling usually cannot solve. However, the optimization algorithm in the related art cannot cope with complex terrain, and has problems such as poor convergence precision and slow convergence speed.

[0004] In the prior art, the optimization focus of unmanned aerial vehicle flight path optimization technology is usually placed on path optimization or in-flight optimization. Currently, there are few applications of unmanned aerial vehicle flight path optimization technology that combines the two.

[0005] Therefore, we propose an unmanned aerial vehicle flight path optimization method and system based on artificial intelligence. SUMMARY

[0006] In view of the above-mentioned shortcomings of the prior art, the present application provides an unmanned aerial vehicle flight path optimization method and system based on artificial intelligence, which can effectively solve the problems of the prior art.

[0007] To achieve the above object, the present application is realized by the following technical solutions. The application discloses an unmanned aerial vehicle flight path optimization system based on artificial intelligence, which comprises: A planning module is configured to upload a topographic model of a flight area of the unmanned aerial vehicle, select passing points of a flight task of the unmanned aerial vehicle in the topographic model, and determine a reference path of the flight task; a construction module is configured to receive the reference path of the flight task determined by the planning module, translate based on the reference path of the flight task, and construct a rough application path of the flight task; a coordination module is configured to obtain the rough application path of the flight task constructed by the construction module, coordinate the rough application path of the flight task, and obtain an application path of the flight task; a driving module is configured to receive the application path of the flight task obtained by the coordination module, feed back the application path of the flight task to the unmanned aerial vehicle, and drive the unmanned aerial vehicle to perform the flight task based on the application path of the flight task; a sensing module is configured to sense obstacles in the process of performing the flight task by the unmanned aerial vehicle based on the application path of the flight task; and an avoidance module is configured to obtain a sensing result of the obstacles in the sensing module in real time, control the unmanned aerial vehicle to avoid the obstacles by leaving the application path of the flight task when the obstacles are sensed by the sensing module, and return to the application path of the flight task.

[0008] Further, in the planning module, the number of the selected passing points is not less than two, and each passing point is located on the surface of the topographic model; after the selection of the passing points is completed, the passing points are sequentially connected along the surface of the topographic model based on the passing points and the selection order of the passing points, so that the reference path of the flight task of the unmanned aerial vehicle is obtained, and the path segment between any two adjacent points on the reference path of the flight task of the unmanned aerial vehicle is a straight line in the overhead view of the topographic model. Further, the reference path of the flight task of the unmanned aerial vehicle is stored in the unmanned aerial vehicle synchronously after being obtained.

[0009] Further, in the construction module, the uploaded topographic model and the reference path of the flight task in the planning module are placed in an adaptive three-dimensional space, so that the topographic model is located at the center position in the three-dimensional space, and the coordinates of any position on the topographic model and the reference path of the flight task are known based on the topographic model and the reference path of the flight task represented in the three-dimensional space. The construction module obtains the coordinates of the highest point and the lowest point on the reference path of the flight task, determines the height difference between the two coordinates, uses the determined height difference as the translation distance of the reference path of the flight task on the surface of the topographic model placed in the three-dimensional space, performs the translation operation, and sets the translation direction to be always vertically upward, so as to complete the construction of the rough application path of the flight task. Before the translation operation is performed, the flight mission reference path is accumulated into the translation distance by user-defined values ​​on the system side. During the translation operation, the translation operation is performed based on the path points of each UAV flight mission in the flight mission reference path as translation operation points.

[0010] Furthermore, the coordination module is equipped with a capture unit and a reduction unit. The capture unit is used to capture a road segment that is straight from a top-down perspective and consists of no less than three waypoints in the preliminary flight mission application path. The reduction unit is used to obtain the road segment captured by the capture unit, identify whether there are any coordinateable waypoints on the road segment, and reduce the coordinateable waypoints on the road segment when the identification result is yes. When the capture unit captures an empty set of road segments, the coordination module ends its operation, reduces the module's operation phase, and identifies whether the two ends of a sub-road segment composed of three or more adjacent waypoints on the acquired road segment are higher than other waypoints between the two ends. If all other waypoints are lower than the two ends, the two ends of the sub-road segment are connected by a straight line to obtain a new road segment. The corresponding sub-road segment is iterated with the new road segment, and a perpendicular line is drawn upward through the waypoints on the atomic road segment. The intersection of the perpendicular line and the new road segment is recorded as the mapping point of the waypoint where the perpendicular line is located.

[0011] Furthermore, during the operation phase of the drive module, after receiving the flight mission application path, it synchronously obtains the scale of the terrain model and the actual terrain, as well as the actual position coordinates of any point on the flight mission application path, in the terrain model. Based on the actual position coordinates and the scale of the terrain model and the actual terrain, it obtains the actual position coordinates of any point on the flight mission application path. The drive module drives the UAV to fly and execute the flight mission based on the actual coordinates corresponding to each point along the flight mission application path.

[0012] Furthermore, the sensing module is integrated with an infrared ranging module. When the UAV is flying between any two waypoints during the analysis task based on the flight mission application path, the ranging end of the infrared ranging module of the sensing module always points to the nearest waypoint in the flight direction and performs ranging operations in real time. The sensing distance of the infrared ranging module meets the following conditions: ; In the formula: V represents the sensing distance of the infrared ranging module; V represents the preset flight speed of the UAV when performing its flight mission; and t represents the value of the UAV's flight distance in V after converting the unit of time to seconds.

[0013] Furthermore, the logic for the infrared ranging module to perceive obstacles is as follows: ; In the formula: is the real-time running distance measurement result of the infrared distance measurement module; The infrared distance measurement module is set to have a running frequency of no less than 1 ms. is true, indicating that the sensing module senses an obstacle.

[0014] Further, in the avoidance module running stage, the position coordinates of the obstacle are determined based on the running distance measurement result of the infrared distance measurement module and the flight task application path, a circle is drawn with the position coordinates of the obstacle as the center and the current distance measurement result of the infrared distance measurement sensor corresponding to the road segment as the radius, the drawn circle is symmetrically bisected with the flight task application path, a semicircle on either side is selected as the avoidance path, and the avoidance path is applied to the flight of the unmanned aerial vehicle to avoid the obstacle. The two ends of the semicircle are connected to the flight task application path, and the road segment defined by the two ends of the semicircle on the flight task application path is replaced by the semicircle.

[0015] Further, the planning module is connected to a construction module and a coordination module through wireless network interaction, the lower-level coordination module is connected to a capture unit and a reduction unit through wireless network interaction, and the coordination module is connected to a driving module, a sensing module, and an avoidance module through wireless network interaction.

[0016] On the other hand, the unmanned aerial vehicle flight path optimization method based on artificial intelligence includes the following steps: uploading an unmanned aerial vehicle flight area topographic model, selecting a way point of a flight task of the unmanned aerial vehicle in the topographic model, creating a flight task reference path based on the selected way point, obtaining the flight task reference path, translating the flight task reference path to construct a flight task application path prototype, obtaining the flight task application path prototype, coordinating the flight task application path prototype to obtain a flight task application path, uploading the flight task application path to the unmanned aerial vehicle, controlling the unmanned aerial vehicle to fly based on the flight task application path, and sensing obstacles in real time during the flight, flying according to the flight task application path when no obstacle is sensed, generating an avoidance path when an obstacle is sensed, applying the avoidance path to avoid the obstacle away from the flight task application path, and then returning to the flight task application path.

[0017] Compared with the known prior art, the technical scheme provided by the present application has the following beneficial effects: The application provides an unmanned aerial vehicle flight path optimization method and system based on artificial intelligence. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0019] Figure 1 FIG. 1 is a structural schematic diagram of an unmanned aerial vehicle flight path optimization system based on artificial intelligence; Figure 2 FIG. 2 is a flowchart of an unmanned aerial vehicle flight path optimization method based on artificial intelligence. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0021] The present application will be further described below in combination with the embodiments.

[0022] Embodiment 1:

[0023] The unmanned aerial vehicle flight path optimization system based on artificial intelligence in this embodiment, as shown in FIG. 1, comprises: Figure 1 a planning module for uploading a topographic model of an unmanned aerial vehicle flight area, selecting a passing point of an unmanned aerial vehicle flight task in the topographic model, and determining a flight task reference path; ​In the planning module, in the stage of selecting the passing points of the UAV flight task, the number of the selected passing points is not less than two, and each passing point is located on the surface of the geomorphic model. After the selection of the passing points is completed, the passing points are sequentially connected along the surface of the geomorphic model based on the passing points and the selection order of the passing points, so as to obtain the reference path of the UAV flight task. In the overhead view of the geomorphic model, the path segment between any two adjacent points on the reference path of the UAV flight task is a straight line. The reference path of the UAV flight task is stored in the UAV synchronously after being obtained. The construction module is configured to receive the reference path of the flight task determined by the planning module, and to translate the reference path of the flight task to construct the embryonic form of the application path of the flight task. In the running stage of the construction module, the geomorphic model and the reference path of the flight task uploaded in the planning module are placed in the adapted three-dimensional space for representation, so that the geomorphic model is located at the center position in the three-dimensional space. The coordinates of any position on the geomorphic model and the reference path of the flight task in the three-dimensional space for representation are known. The construction module obtains the coordinates of the highest point and the lowest point on the reference path of the flight task, determines the height difference between the two coordinates, and uses the determined height difference as the translation distance of the reference path of the flight task placed on the surface of the geomorphic model in the three-dimensional space. The translation operation is performed with the translation direction always being vertically upward, so as to complete the construction of the embryonic form of the application path of the flight task. Before the translation of the reference path of the flight task based on the translation distance, a user of the system side customizes a numerical value and adds the numerical value to the translation distance. Then, the translation operation is performed. During the translation operation, the passing points of each UAV flight task in the reference path of the flight task are used as the translation operation points to perform the translation operation. The coordination module is configured to obtain the embryonic form of the application path of the flight task constructed by the construction module, and to coordinate the embryonic form of the application path of the flight task to obtain the application path of the flight task. The coordination module is provided with a capturing unit and a reduction unit. The capturing unit is configured to capture the path segment composed of at least three passing points that are straight lines in the overhead view in the embryonic form of the application path of the flight task. The reduction unit is configured to obtain the path segment captured by the capturing unit, to identify whether there is a coordinatable passing point on the path segment, and to reduce the coordinatable passing point on the path segment when the identification result is yes. Wherein, when the set of the capture unit running capture sections is empty, the coordination module ends running, the reduction module runs the stage, identifies whether the two end passing points on the sub-section formed by each adjacent three or more passing points on the road section obtained by the recognition module are higher than other passing points between the two end passing points, if all other passing points are lower than the two end passing points, connects the two end passing points with a straight line to obtain a new road section, iterates the corresponding sub-section with the new road section, and marks the intersection point of the perpendicular line and the new road section as the mapping point of the passing point where the perpendicular line is located; The driving module is configured to receive the flight task application path obtained by the coordination module, feed back the flight task application path to the unmanned aerial vehicle, and drive the unmanned aerial vehicle to execute the flight task based on the flight task application path. The driving module runs in the stage, after receiving the flight task application path, synchronously obtains the scale of the topographic model and the real topography and the real position coordinates of any point in the flight task application path in the topographic model, obtains the real position coordinates of any point in the flight task application path based on the real position coordinates and the scale of the topographic model and the real topography, and drives the unmanned aerial vehicle to fly based on the corresponding real coordinates of each passing point in the flight task application path to execute the flight task. The perception module is configured to perceive obstacles in the process that the unmanned aerial vehicle executes the flight task based on the flight task application path. The perception module is integrated by an infrared distance measuring module, when the unmanned aerial vehicle flies between any two passing points in the process that the unmanned aerial vehicle executes the flight task based on the flight task application path, the distance measuring end of the infrared distance measuring module of the perception module always points to the nearest passing point in the flight direction and performs distance measuring operation in real time. The perception distance of the infrared distance measuring module satisfies the following conditions: ; In the formula: is the perception distance of the infrared distance measuring module; V is a preset flight speed of the unmanned aerial vehicle when executing the flight task; t is the value after the unit time of the flight distance of the unmanned aerial vehicle in V is converted into seconds; The logic that the infrared distance measuring module perceives obstacles is: ; In the formula: is the real-time running distance measuring result of the infrared distance measuring module; Wherein, the running frequency of the infrared distance measuring module is not less than 1 ms; is established, which indicates that the perception module perceives obstacles; The avoidance module is configured to obtain the obstacle perception result of the perception module in real time, control the unmanned aerial vehicle to leave the flight task application path to avoid obstacles when the perception module perceives obstacles, and return to the flight task application path. The avoidance module runs in a stage, determines the obstacle position coordinates based on the infrared distance measurement module running distance measurement results and the flight task application path as a reference, draws a circle with the obstacle position coordinates as the center and the current distance measurement results of the infrared distance measurement sensor corresponding to the road section as the radius, makes the drawn circle be symmetrically divided by the flight task application path, selects a semicircle on either side as an avoidance path, and applies it to the flight of the unmanned aerial vehicle to avoid obstacles; Wherein, the two ends of the semicircle are connected with the flight task application path, and the road section defined by the two ends of the semicircle on the flight task application path is replaced by the semicircle; The planning module is connected with the construction module and the coordination module through wireless network interaction, the lower level of the coordination module is connected with the capture unit and the reduction unit through wireless network interaction, and the coordination module is connected with the driving module, the sensing module and the avoidance module through wireless network interaction.

[0024] In this embodiment, the planning module runs to upload the topographic model of the unmanned aerial vehicle flight area, selects the passing points of the flight task of the unmanned aerial vehicle in the topographic model to determine the flight task reference path, the construction module runs behind to receive the flight task reference path determined by the planning module, translates based on the flight task reference path to construct the flight task application path sketch, the coordination module further obtains the flight task application path sketch constructed by the construction module, coordinates the flight task application path sketch to obtain the flight task application path, the capture unit synchronously captures the road section composed of at least three passing points in a straight line in the flight task application path sketch from the overhead perspective, the reduction unit obtains the road section captured by the capture unit in real time, identifies whether there is a coordinatable passing point on the road section, and when the identification result is yes, reduces the coordinatable passing point on the road section, the driving module receives the flight task application path obtained by the coordination module, feeds back the flight task application path to the unmanned aerial vehicle, drives the unmanned aerial vehicle to execute the flight task based on the flight task application path, and perceives the obstacles in the process of the unmanned aerial vehicle executing the flight task based on the flight task application path through the sensing module. The avoidance module obtains the obstacle perception results of the sensing module in real time, controls the unmanned aerial vehicle to avoid the obstacles away from the flight task application path when the obstacles are perceived by the sensing module, and returns to the flight task application path.

[0025] Through the system running in the above embodiment, for the unmanned aerial vehicle executing flight task scene, flight path optimization and obstacle avoidance optimization are provided to ensure that the unmanned aerial vehicle stably completes the flight task, the corresponding obstacle avoidance capability is improved, and the flight task path is simpler and more reliable.

[0026] Embodiment 2:

[0027] In a specific implementation level, based on embodiment 1, this embodiment refers to Figure 2 Further specific description is made to the unmanned aerial vehicle flight path optimization system based on artificial intelligence in embodiment 1: An unmanned aerial vehicle flight path optimization method based on artificial intelligence, comprising the following steps: Step 1: uploading an unmanned aerial vehicle flight area topographic model, selecting a way point of an unmanned aerial vehicle flight task in the topographic model, and creating a flight task reference path based on the selected way point; Step 2: obtaining the flight task reference path, and translating the flight task reference path to construct a flight task application path sketch; Step 3: obtaining the flight task application path sketch, and coordinating the flight task application path sketch to obtain a flight task application path; Step 4: uploading the flight task application path to the unmanned aerial vehicle, controlling the unmanned aerial vehicle to fly based on the flight task application path, and sensing obstacles in real time during the flight; Step 5: flying according to the flight task application path when no obstacle is sensed, and generating an avoidance path when an obstacle is sensed; Step 6: avoiding the obstacle by applying the avoidance path to leave the flight task application path, and returning to the flight task application path again.

[0028] In summary, in the above-mentioned embodiments, the flight reference path is determined according to the flight area topographic model, and the application path is optimized to make the unmanned aerial vehicle flight route more suitable for the actual environmental requirements. During the flight, the system can sense obstacles in real time, and once an obstacle is found, the unmanned aerial vehicle can be precisely controlled to avoid the obstacle in time, and can quickly return to the original flight path after avoiding the obstacle, greatly reducing the risk of collision. At the same time, the system can accurately obtain the real coordinates of each point of the flight path according to the scale of the model and the real topography, drive the unmanned aerial vehicle to accurately perform the task, greatly improve the accuracy of the flight, reduce the flight error, and ensure that the unmanned aerial vehicle can stably and efficiently complete various flight tasks, which has important practical value in the fields of surveying and mapping, inspection, etc.

[0029] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements will not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An artificial intelligence based drone flight path optimization system characterized in that, The application relates to a flight task planning method for a UAV (Unmanned Aerial Vehicle), which comprises the following steps: a planning module is used to upload a topographic model of a UAV flight area, select waypoints of a UAV flight task in the topographic model, and determine a flight task reference path; a construction module is used to receive the flight task reference path determined by the planning module, perform translation based on the flight task reference path, and construct a flight task application path sketch; a coordination module is used to obtain the flight task application path sketch constructed by the construction module, coordinate the flight task application path sketch, and obtain a flight task application path; a driving module is used to receive the flight task application path obtained by the coordination module, feed back the flight task application path to the UAV, and drive the UAV to execute the flight task based on the flight task application path; a sensing module is used to sense obstacles in the process that the UAV executes the flight task based on the flight task application path; and an avoidance module is used to obtain the obstacle sensing result of the sensing module in real time, control the UAV to avoid the obstacles by leaving the flight task application path when the obstacles are sensed by the sensing module, and return to the flight task application path. In the planning module, the number of selected waypoints is not less than two, and each waypoint falls on the surface of the topographic model; after the selection of the waypoints is completed, the waypoints are sequentially connected along the surface of the topographic model based on the waypoints and the selection order of the waypoints, so that the flight task reference path of the UAV is obtained, and the path section between any two adjacent points on the flight task reference path of the UAV is a straight line under the overhead perspective of the topographic model. The flight task reference path of the UAV is stored in the UAV in real time after being obtained. In the construction module, the topographic model and the flight task reference path uploaded by the planning module are placed in an adaptive three-dimensional space, so that the topographic model is located at the center position in the three-dimensional space; the coordinates of any position on the topographic model and the flight task reference path are known based on the topographic model and the flight task reference path represented in the three-dimensional space; The construction module obtains the position coordinates of the highest point and the lowest point on the flight task reference path, determines the height difference between the two position coordinates, uses the determined height difference as the translation distance of the flight task reference path on the surface of the topographic model placed in the three-dimensional space, performs the translation operation, and sets the translation direction to be always vertically upward, so as to complete the construction of the flight task application path sketch. The coordination module is provided with a capturing unit and a reduction unit; the capturing unit is used to capture a path section composed of at least three waypoints which are straight lines under the overhead perspective in the flight task application path sketch; the reduction unit is used to obtain the path section captured by the capturing unit, identify whether there are coordinatable waypoints on the path section, and reduce the coordinatable waypoints on the path section when the identification result is yes. ​ 2.The AI-based UAV flight path optimization system of claim 1, wherein, ​ ​ 3.The AI-based UAV flight path optimization system of claim 1, wherein, ​ ​ ​ 4.The AI-based UAV flight path optimization system of claim 1, wherein, ​ Wherein, when the set of capture unit running capture road sections is empty, the coordination module ends running, the reduction module runs the stage, and identifies whether the two end passing points on the sub-road section composed of each adjacent three and above passing points on the road section obtained by the recognition are higher than other passing points between the two end passing points. If all other passing points are lower than the two end passing points, the two end passing points of the sub-road section are connected by a straight line to obtain a new road section. The corresponding sub-road section is iterated with the new road section, and the intersection point of the perpendicular line drawn from the passing point on the original road section is recorded as the mapping point of the passing point on the perpendicular line. 5.The AI-based UAV flight path optimization system of claim 1, wherein, The driving module runs the stage, synchronously obtains the scale of the topography model and the real topography and the real position coordinates of any point in the flight task application path in the topography model after receiving the flight task application path, obtains the real position coordinates of any point in the flight task application path based on the real position coordinates and the scale of the topography model and the real topography, and drives the UAV to fly based on the real coordinates corresponding to each passing point on the flight task application path to execute the flight task. 6.The AI-based UAV flight path optimization system of claim 1, wherein, The perception module is integrated by an infrared distance measuring module. When the UAV flies between any two passing points during the analysis task process based on the flight task application path, the distance measuring end of the infrared distance measuring module of the perception module always points to the nearest passing point in the flight direction and performs distance measuring operation in real time. The sensing distance of the infrared distance measuring module satisfies the following conditions: ； In the formula: is the sensing distance of the infrared distance measuring module; V is a preset flight speed of the UAV when performing a flight task; and t is a value after converting the unit time of the flight distance of the UAV in V into seconds.

7. The Artificial Intelligence based UAV flight path optimization system according to claim 6, wherein, The logic of the infrared distance measuring module running to perceive obstacles is as follows: ； In the formula: is the real-time running ranging result of the infrared ranging module; The operation frequency of the infrared distance measuring module is not less than 1ms. is true, indicating that the sensing module senses the obstacle. 8.The AI-based UAV flight path optimization system of claim 1 or 6, wherein, The avoidance module runs the stage, determines the position coordinates of the obstacles based on the distance measuring result of the infrared distance measuring module with reference to the flight task application path, draws a circle with the position coordinates of the obstacles as the center and the current distance measuring result of the infrared distance measuring sensor as the radius, makes the drawn circle be symmetrically divided by the flight task application path, selects a semicircle on one side as the avoidance path, and applies it to the flight of the UAV to avoid obstacles. Wherein, the two ends of the semicircle are connected with the flight task application path, and the road section defined by the two ends of the semicircle on the flight task application path is replaced by the semicircle. 9.The AI-based UAV flight path optimization system of claim 1, wherein, The planning module is connected with the construction module and the coordination module through wireless network interaction. The subordinate of the coordination module is connected with the capture unit and the reduction unit through wireless network interaction. The coordination module is connected with the driving module, the perception module, and the avoidance module through wireless network interaction.

10. The method for optimizing the flight path of a UAV based on artificial intelligence, the method is the implementation method of the system for optimizing the flight path of a UAV based on artificial intelligence according to any one of claims 1-9, characterized in that, The method comprises the following steps: Step 1: uploading a topography model of a UAV flight area, selecting passing points of a UAV flight task in the topography model, and creating a flight task reference path based on the selected passing points; Step 2: obtaining the flight task reference path, translating the flight task reference path, and constructing a flight task application path sketch; Step 3: obtaining the flight task application path sketch, coordinating the flight task application path sketch, and obtaining the flight task application path; Step 4: uploading the flight task application path to the UAV, controlling the UAV to fly based on the flight task application path, and sensing obstacles in real time during the flight process. Step 5: flying on the flight task application path when no obstacle is perceived, and generating an avoidance path when an obstacle is perceived; Step 6: applying the avoidance path to avoid the obstacle after leaving the flight task application path, and returning to the flight task application path again.

Images