A flying robot route planning method based on weather diagnosis technology

Abstract

The invention belongs to the technical field of unmanned aerial vehicle (UAV) trajectory planning, and specifically discloses a flight robot route planning method based on weather diagnosis technology, comprising: inputting three-dimensional terrain and meteorological parameters of a mission area into a matlab simulation environment, taking terrain data as a basis and performing grid processing, and then initializing pheromones at all nodes on a flight area grid to form an initial pheromone matrix, and using an ant colony algorithm and acellular automata to plan a flight robot route, wherein the meteorological parameters are obtained based on a Unidata IDV weather diagnosis system, and when the method is used to plan the flight robot route, atmospheric phenomena can be accurately captured, extreme weather events can be identified, and the development process of extreme weather events can be analyzed, thereby meeting the needs of complex flight trajectory planning for the flight robot.

Description

Technical Field

[0001] The invention relates to a flight robot trajectory planning method, in particular to a flight robot trajectory planning method based on meteorological parameters obtained by a Unidata IDV weather diagnosis system. Background Art

[0002] The meteorological conditions faced by drones at different altitudes vary greatly. The atmospheric density and pressure decrease exponentially with increasing altitude. At an altitude of 20 km, the atmospheric density is about 7% of the surface, and the atmospheric pressure is about 5.3% of the surface. In the troposphere, as the altitude increases, the ground reflects the radiation heating and the temperature decreases; in the stratosphere, the temperature gradually increases with increasing altitude because ozone absorbs solar radiation to heat the atmosphere and ultraviolet radiation. The existence of inversion layers and jet streams in the atmosphere causes the wind speed to increase first and then decrease with increasing altitude.

[0003] Driven by the mission requirements of flying robots and the development of science and technology, the airspace range of flying robots to perform tasks has gradually expanded from low altitude to high altitude and even near space. First of all, flying robots are constantly fighting against their own gravity. The higher the flying robots fly, the greater the technical difficulty and the higher the requirements for structure, materials, communication, navigation and control technology. Secondly, flying robots are constantly fighting against extreme environmental conditions. Although there is no need to consider the human body limit of the pilot, the environment at different working altitudes of flying robots is very different, and the weather resistance requirements of various components are also very high. For example, the temperature near the top of the troposphere reaches -50℃. In addition, under the action of solar radiation, the flying robots are heated and heated, which brings certain tests to the fuselage materials and airborne sensors. For flying robots that stay in the air for a long time, ultraviolet radiation exacerbates the aging of materials, etc. Finally, flying robots are constantly fighting against complex meteorological conditions. The most common thing is that flying robots need to reduce the interference of wind on flight missions. For example, how to eliminate the deviation caused by wind speed on track tracking through the control system, overcome the influence of atmospheric turbulence on the flight stability of flying robots, and avoid stalling when encountering wind shear during landing. With the continuous development and progress of science and technology, the ability of flying robots to resist the natural environment and complete the tasks assigned by users is constantly improving. For example, the combined effects of wind speed, terrain and obstacles are often simplified or ignored in the comprehensive meteorological environment simulation, resulting in the inability of the comprehensive meteorological environment simulation to fully reflect the real situation. CN 117808985 A discloses a solar unmanned aerial vehicle simulation system based on three-dimensional GIS, including: a map server hardware environment for storing and accessing map data; map preset geographic data, which is a database designed using the data model and organization model of geographic data; map data processing and management software for processing and managing map data; comprehensive meteorological environment simulation for establishing a weather model based on real weather information; a track planning algorithm library, including a search algorithm combined with graph theory, an artificial potential field method, an intelligent method and a planning method based on mathematics; a flight simulation synchronization client, which provides a visual interface and visual operation for the simulation system. The present invention uses basic geographic data to model the real geographical environment, and real meteorological data to simulate the environment. The simulation scene is closer to the real environment, laying a foundation for the development of intelligent task planning functions and virtual flight tests. However, in order to improve efficiency, this comprehensive meteorological environment simulation system will simplify the modeling of weather and atmospheric conditions. For example, the turbulent effects of airflow or extreme climates, such as sandstorms and hurricanes, may affect the flight stability of the flying robot. These complex physical phenomena are represented by simple parameter models in the comprehensive meteorological environment simulation, which cannot fully reproduce the complex fluid dynamics effects in reality. Therefore, when the flying robot flies under complex meteorological conditions, the comprehensive meteorological environment simulation results may underestimate the actual impact of the environment on the performance of the flying robot. Summary of the invention

[0004] The technical problem to be solved by the invention is: to overcome the above-mentioned defects of the prior art and to provide a flying robot path planning method based on weather diagnosis technology. When planning the trajectory of the flying robot, this method can accurately capture atmospheric phenomena, identify extreme weather events, and analyze the development process of extreme weather events, which can meet the needs of complex flight trajectory planning of the flying robot.

[0005] Technical solution to solve technical problems: A flying robot route planning method based on weather diagnosis technology, including:

[0006] Step 1: Input the three-dimensional terrain and meteorological parameters of the mission area into the MATLAB simulation environment, perform grid processing based on the terrain data, and then initialize the pheromone at all nodes on the flight area grid to form an initial pheromone matrix;

[0007] Step 2, place multiple ants at the starting point and wait for departure;

[0008] Step 3: Each ant selects the next point of the grid according to the state transition rule and eventually reaches the target point, forming a feasible route;

[0009] Step 4, calculate the objective function of the feasible route obtained by each ant to obtain the optimal route solution;

[0010] Step 5, according to the definition of cellular neighbors, determine the cellular neighbors of the optimal solution and calculate the objective function of the cellular neighbors;

[0011] Step 6: According to the cellular evolution rule, the current optimal solution is updated to replace the current optimal route solution;

[0012] Step 7, adjusting the pheromone of each point according to the objective function and the pheromone adjustment criterion;

[0013] Step 8, check the optimal solution and determine whether the pheromone evaporation factor needs to be adjusted. If necessary, adjust the pheromone evaporation factor according to the pheromone update formula;

[0014] Step 9, determine whether the iteration condition is met. If so, the search is completed; if not, return to step 2) and re-execute until the algorithm termination condition is met to obtain the flight robot route.

[0015] The meteorological parameters are obtained based on the Unidata IDV weather diagnostic system according to the following steps:

[0016] A. Meteorological data collection and preprocessing: collect meteorological observation data in the mission area, perform quality control and clean up the data, and perform spatial interpolation and gridding on the data;

[0017] B. Establish a three-dimensional terrain model of the mission area, analyze the three-dimensional terrain structure of the mission area, and display the characteristics of terrain height, mountains, and valleys to diagnose the impact of terrain on local meteorological conditions;

[0018] C. Establish a two-dimensional geostrophic wind field model for the mission area, analyze the horizontal distribution of the geostrophic wind field, and display the wind speed and direction at different heights to diagnose large-scale wind field dynamics;

[0019] D. Establish a three-dimensional wind vector field model for the mission area, analyze the changes in the wind field in three-dimensional space, and display the wind direction, wind speed and vertical movement at different heights for diagnosing the vertical wind field;

[0020] E. Establish a three-dimensional atmospheric boundary layer field model for the mission area, analyze the temperature, humidity, and wind speed parameters of the atmospheric boundary layer, and display the vertical changes within the boundary layer to diagnose weather conditions near the surface;

[0021] F. Establish a two-dimensional temperature field model for the mission area, analyze the horizontal distribution of surface temperature, and display the temperature differences within the mission area for diagnosis of local thermal processes;

[0022] G. Establish a three-dimensional temperature field model for the mission area, analyze the temperature distribution at different heights, and display the temperature variation with height, which is used to diagnose the temperature structure in three-dimensional space;

[0023] H. Establish a two-dimensional atmospheric pressure field model for the mission area, analyze the horizontal distribution of surface atmospheric pressure, and display the pressure changes in the mission area for diagnosis of weather systems;

[0024] I. Establish a three-dimensional atmospheric pressure field model for the mission area, analyze the distribution of atmospheric pressure in three-dimensional space, and display the changes in air pressure at different altitudes to diagnose the three-dimensional distribution of air pressure in the atmosphere;

[0025] J. Establish a three-dimensional temperature stratification slice model for the mission area, analyze the temperature stratification structure of the mission area, and display the temperature variation slices with height; used to diagnose the stability of temperature stratification;

[0026] K. Establish a comprehensive analysis model for meteorological factors, conduct superposition analysis on all the above meteorological factors, comprehensively display the interactions among various factors, evaluate the synergistic effects of multiple meteorological factors through the comprehensive analysis model, generate meteorological diagnostic data, and provide meteorological parameters for the mission area.

[0027] As a further improvement of the present invention: according to the requirements of the flight mission objectives of the flying robot, the meteorological diagnostic data is input into the numerical weather forecast model to simulate the future weather evolution process, thereby generating the meteorological parameters of all the passing points of the flying robot.

[0028] As a further improvement of the present invention: the objective function of the feasible route includes a track cost function, a wind field cost function, a microburst cost function, a thunderstorm cost function, and a turbulence cost function.

[0029] As a further improvement scheme of the present invention: the two-dimensional geostrophic wind field model of the mission area, the three-dimensional wind vector field model of the mission area, the three-dimensional atmospheric boundary layer field model of the mission area, the two-dimensional temperature field model of the mission area, the three-dimensional temperature field model of the mission area, the two-dimensional atmospheric pressure field model of the mission area, the three-dimensional atmospheric pressure field model of the mission area, and the three-dimensional temperature stratification slice model of the mission area are weather diagnosis models constructed on the basis of the weather model constructed using WRF4 and constructed using Unidata IDV secondary development.

[0030] As a further improvement of the present invention: during the flight of the flying robot, meteorological observation data of the mission area is received in real time, and radar and satellite real-time monitoring data are used to track the evolution of the weather system, and the meteorological parameters of the mission area are updated in real time. The future route of the flying robot is corrected according to the current meteorological parameters of the mission area.

[0031] Beneficial effects: The flight robot route planning method based on weather diagnosis technology of the present invention adopts the meteorological parameters provided by Unidata IDV weather diagnosis system for route planning, which has the following advantages over the comprehensive meteorological simulation in the prior art:

[0032] 1. Improve simulation accuracy based on real data: Weather diagnostic analysis technology relies on real atmospheric environment data provided by meteorological satellites, radars, ground observation stations, etc. Introducing these high-precision data into the flight robot simulation system can simulate a realistic flight environment and accurately reflect the impact of weather factors such as wind speed, wind direction, humidity, and air pressure on the flight robot.

[0033] 2. Dynamic weather changes: Different from the static and idealized weather models in traditional simulation systems, weather diagnostic analysis technology can capture real-time weather changes, which is particularly important for flying robot flight simulation. It can simulate the dynamic weather conditions encountered by flying robots during flight, thereby improving simulation accuracy.

[0034] 3. Enhanced simulation capabilities for extreme weather: Weather diagnostic analysis technology is good at analyzing and identifying extreme weather events, such as strong winds, heavy rains, thunderstorms, etc. These extreme weather events often have a significant impact on the flight of flying robots. Applying this technology in the simulation system can accurately simulate the flight performance of flying robots under extreme weather conditions and test the wind resistance, stability and response strategies of flying robots in harsh environments.

[0035] 4. Early warning and response mechanism simulation: By analyzing the formation and development of extreme weather, the simulation system can simulate the decision-making and adjustment process of the flying robot in response to sudden weather events in advance. For example, by simulating thunderstorms or sudden changes in wind speed during the simulation process, the system can analyze the decision-making ability of the flying robot to avoid, adjust the flight path or return to the base through simulation.

[0036] 5. Improve the reliability of flight mission planning: By using real-time weather diagnostic data, the simulation system can more accurately plan the flight path of the flying robot. For example, weather diagnostic analysis can identify wind field changes and meteorological conditions in different regions. The simulation system adjusts the flight altitude, speed and route of the flying robot accordingly to avoid bad weather and improve the success rate of the flight mission.

[0037] 6. Optimize energy management and endurance: The flight performance of the flying robot is greatly affected by weather factors. By diagnosing and analyzing the impact of weather on the energy consumption of the flying robot (such as wind resistance, air humidity, etc.), the simulation system can optimize energy management and endurance strategies, helping the flying robot to reasonably allocate energy under different meteorological conditions, thereby extending the flight time.

[0038] 7. Improve the applicability of multi-scenario simulation: The introduction of weather diagnosis and analysis technology enables the flight robot simulation system to adapt to a variety of different meteorological scenarios, such as weather changes in complex environments such as cities, mountains, and oceans. In this way, the simulation system can flexibly simulate the flight performance of the flying robot in different regions and under different meteorological conditions according to the needs of the flight mission, providing more accurate simulation support for multi-scenario flight missions.

[0039] 8. Dealing with seasonal meteorological changes: Weather diagnostic analysis can monitor and predict seasonal changes in atmospheric conditions over a long period of time, such as typhoon season, winter storms, etc. The flying robot simulation system can use these forecast data to simulate flight conditions in different seasons in advance, helping flying robot operators to better cope with seasonal extreme weather.

[0040] 9. Improve the decision-making support capabilities of the simulation system: By combining weather diagnosis and analysis technology with flight robot simulation, the system can obtain weather forecasts in real time and adjust the flight strategy of the flying robot accordingly. For example, based on the strong wind warning or heavy rainfall forecast in the weather forecast, the system can suggest that the flying robot adjust its flight altitude or return home early to avoid the risks brought by dangerous weather.

[0041] 10. Improve the comprehensiveness of flying robot performance evaluation: By introducing weather diagnostic analysis data, the simulation system can more comprehensively evaluate the performance of flying robots under different meteorological conditions. Unlike idealized environmental simulation, simulation based on real weather data can test the stability, anti-interference and endurance of flying robots in real environments such as strong winds, rainfall, and high humidity. The simulation system can use multiple meteorological parameters (such as temperature, humidity, wind speed, precipitation, etc.) generated by weather diagnostic technology to compare and analyze multiple performances of flying robots. For example, the system can test how the battery life of flying robots changes under high humidity and high temperature conditions, providing data support for optimizing the design and use strategies of flying robots. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The following is a further detailed description of the flying robot route planning method based on weather diagnosis technology of the present invention in conjunction with the accompanying drawings.

[0043] Figure 1 It is the rendering of the three-dimensional underlying surface modeling of the flight mission area of ​​the flying robot;

[0044] Figure 2 It is a two-dimensional modeling effect diagram of the temperature field in the flight mission area of ​​the flying robot;

[0045] Figure 3 It is a three-dimensional modeling effect diagram of the temperature field in the flight mission area of ​​the flying robot;

[0046] Figure 4 It is a three-dimensional modeling effect diagram of the pressure field of the flying robot's flight mission;

[0047] Figure 5 This is a slice effect diagram of the temperature stratification in the flight mission area of ​​the flying robot;

[0048] Figure 6 It is a time series effect diagram based on the comprehensive analysis of meteorological factors of IDV. DETAILED DESCRIPTION

[0049] A flying robot route planning method based on weather diagnosis technology, comprising:

[0050] Step 1: Input the three-dimensional terrain and meteorological parameters of the mission area into the MATLAB simulation environment, perform grid processing based on the terrain data, and then initialize the pheromone at all nodes on the flight area grid to form an initial pheromone matrix;

[0051] Step 2, place multiple ants at the starting point and wait for departure;

[0052] Step 3: Each ant selects the next point of the grid according to the state transition rule and eventually reaches the target point, forming a feasible route;

[0053] Step 4, calculate the objective function of the feasible route obtained by each ant to obtain the optimal route solution;

[0054] Step 5, according to the definition of cellular neighbors, determine the cellular neighbors of the optimal solution and calculate the objective function of the cellular neighbors;

[0055] Step 6: According to the cellular evolution rule, the current optimal solution is updated to replace the current optimal route solution;

[0056] Step 7, adjusting the pheromone of each point according to the objective function and the pheromone adjustment criterion;

[0057] Step 8, check the optimal solution and determine whether the pheromone evaporation factor needs to be adjusted. If necessary, adjust the pheromone evaporation factor according to the pheromone update formula;

[0058] Step 9, determine whether the iteration condition is met. If so, the search is completed; if not, return to step 2) and re-execute until the algorithm termination condition is met to obtain the flight robot route.

[0059] The meteorological parameters are obtained based on the Unidata IDV weather diagnostic system according to the following steps:

[0060] A. Meteorological data collection and preprocessing: collect meteorological observation data in the mission area, perform quality control and clean up the data, and perform spatial interpolation and gridding on the data;

[0061] B. Establish a three-dimensional terrain model of the mission area, analyze the three-dimensional terrain structure of the mission area, and display the characteristics of terrain height, mountains, and valleys to diagnose the impact of terrain on local meteorological conditions;

[0062] C. Establish a two-dimensional geostrophic wind field model for the mission area, analyze the horizontal distribution of the geostrophic wind field, and display the wind speed and direction at different heights to diagnose large-scale wind field dynamics;

[0063] D. Establish a three-dimensional wind vector field model for the mission area, analyze the changes in the wind field in three-dimensional space, and display the wind direction, wind speed and vertical movement at different heights for diagnosing the vertical wind field;

[0064] E. Establish a three-dimensional atmospheric boundary layer field model for the mission area, analyze the temperature, humidity, and wind speed parameters of the atmospheric boundary layer, and display the vertical changes within the boundary layer to diagnose weather conditions near the surface;

[0065] F. Establish a two-dimensional temperature field model for the mission area, analyze the horizontal distribution of surface temperature, and display the temperature differences within the mission area for diagnosis of local thermal processes;

[0066] G. Establish a three-dimensional temperature field model for the mission area, analyze the temperature distribution at different heights, and display the temperature variation with height, which is used to diagnose the temperature structure in three-dimensional space;

[0067] H. Establish a two-dimensional atmospheric pressure field model for the mission area, analyze the horizontal distribution of surface atmospheric pressure, and display the pressure changes in the mission area for diagnosis of weather systems;

[0068] I. Establish a three-dimensional atmospheric pressure field model for the mission area, analyze the distribution of atmospheric pressure in three-dimensional space, and display the changes in air pressure at different altitudes to diagnose the three-dimensional distribution of air pressure in the atmosphere;

[0069] J. Establish a three-dimensional temperature stratification slice model for the mission area, analyze the temperature stratification structure of the mission area, and display the temperature variation slices with height; used to diagnose the stability of temperature stratification;

[0070] K. Establish a comprehensive analysis model for meteorological factors, conduct superposition analysis on all the above meteorological factors, comprehensively display the interactions among various factors, evaluate the synergistic effects of multiple meteorological factors through the comprehensive analysis model, generate meteorological diagnostic data, and provide meteorological parameters for the mission area.

[0071] Preferably, according to the requirements of the flight mission objectives of the flying robot, the meteorological diagnostic data is input into the numerical weather forecast model to simulate the future weather evolution process, thereby generating the meteorological parameters of all the waypoints of the flying robot.

[0072] Preferably, the objective function of the feasible route includes a track cost function, a wind field cost function, a microburst cost function, a thunderstorm cost function, and a turbulence cost function.

[0073] Preferably, the two-dimensional geostrophic wind field model of the mission area, the three-dimensional wind vector field model of the mission area, the three-dimensional atmospheric boundary layer field model of the mission area, the two-dimensional temperature field model of the mission area, the three-dimensional temperature field model of the mission area, the two-dimensional atmospheric pressure field model of the mission area, the three-dimensional atmospheric pressure field model of the mission area, and the three-dimensional temperature stratification slice model of the mission area are weather diagnosis models constructed based on the weather model constructed using WRF4 and constructed using Unidata IDV secondary development.

[0074] Preferably, during the flight of the flying robot, meteorological observation data of the mission area is received in real time, and radar and satellite real-time monitoring data are used to track the evolution of the weather system, update the meteorological parameters of the mission area in real time, and correct the future route of the flying robot based on the current meteorological parameters of the mission area. Example

[0075] Flying robot path planning algorithm based on cellular ant colony algorithm:

[0076] In the ant algorithm search process, after each iteration, the cellular automaton principle is applied to perform a cell update operation on the optimal solution to make it jump out of the local extreme value; after the cell update operation is completed, the ant search continues. That is, in the ant algorithm, a cell update operation is added to the optimal solution searched each time. In the cellular ant algorithm, each feasible solution to the optimization problem is regarded as a cell. Therefore, the set of feasible solutions to the problem constitutes the cellular space.

[0077] For the single UAV (flying robot) route planning problem, a feasible route is a cell, so the set of feasible routes is the entire cellular space. From the representation of the route, it can be seen that the route is represented by a combination of points on each of the m-1 vertical lines in the flight area grid, that is, the route can be represented as:

[0078]

[0079] Therefore, the cell space C can be defined as:

[0080] The set of feasible flight paths for UAV is C = (path1, path2, ..., pathn). Among them, the combination pathi formed by taking a point on each vertical line is a cell, that is, each feasible path is a cell.

[0081] The cell neighbors use the extended Moore neighbor type. The extended Moore neighbor type is:

[0082]

[0083] Where:

[0084] diff(pathi, pathj)≤r is the difference between the two combined sortings, that is, the difference between the waypoints of the two feasible routes. If there is no difference, diff(pathi, pathj) is 0;

[0085] When there is a difference, diff(pathi, pathj) is the number of different points in the two combinations, that is, the number of different waypoints;

[0086] r is the degree of difference.

[0087] Since the route is composed of a series of waypoints, different routes can be obtained by the differences between the combinations of waypoints that make up the route, and the range of the cell neighbors can be adjusted by adjusting the size of the difference degree r.

[0088] The cellular evolution rule is mainly determined based on the performance index function corresponding to the cell, that is, the feasible route. By calculating the objective function value of the cell's neighbors, comparing the objective function values ​​of the cell and its neighbors, the cell corresponding to the optimal function value is selected to update the optimal solution. That is, for the currently searched optimal route, it is used as the current cell, and the corresponding cell neighbor range is determined by expanding the Moore neighbor type, and the objective function value of the cell's neighbors is calculated, and the objective function value of the current cell and its neighbors is compared. If the optimal objective function value in the neighboring cell is better than the objective function value of the currently searched optimal route, the currently searched optimal route is updated, and the currently searched optimal route is replaced with the corresponding optimal route in the cellular neighborhood.

[0089] Selection criteria:

[0090] A selection strategy combining deterministic selection and random selection is adopted, and it is dynamically adjusted during the search process to determine the probability of selection.

[0091] At time t, the ant colony moves to the vertical line segment L i Suppose ant k is at point a(x i ,y g ), then the node that ant k will reach at the next moment is determined by the following formula, that is:

[0092]

[0093] Where:

[0094] r is a random number uniformly distributed in (0,1), r0∈(0,1). As the search proceeds, the value of r0 can be dynamically adjusted.

[0095] r is a random number uniformly distributed in (0, 1), r0∈(0, 1). As the search proceeds, the value of r0 can be dynamically adjusted.

[0096] Pheromone Adjustment Guidelines:

[0097] The ants determine the next point to be passed by the probability of reaching the next node according to the state transition rule, and repeat this process until they find the target point B. In one iteration, when all ants have searched for a feasible route, the pheromone adjustment criterion is applied to update the pheromone on each node of the grid. Pheromone update formula, the update of pheromone includes two parts: volatilization and enhancement.

[0098] Volatilization: After each iteration, the pheromone of all paths will evaporate at a certain rate. The formula of this process is expressed as:

[0099] ,

[0100] in:

[0101] is the pheromone concentration on edge j at time t,

[0102] is the volatility coefficient (0< <1);

[0103] Enhancement: After a round of ant search, the pheromone on the excellent path will increase. The formula of this process is expressed as:

[0104] ,

[0105] in: The amount of pheromone added to path j is usually determined by the quality of the path (such as path length): ,

[0106] m is the number of ants using the path in the current iteration,

[0107] It can be expressed as =Q / L k , Q is a constant (pheromone intensity); L k is the length of the path traversed by ant k.

[0108] Cell update criteria:

[0109] In the ant algorithm search process, after each iteration to obtain the optimal solution, the cellular automaton principle is applied to perform a cell update operation on the optimal solution alone, that is, the current optimal solution is used as the current cell, the extended Moore neighbor type is used, the difference degree of the neighbors is set to 3, the cellular neighbors of the current optimal solution are determined, and the objective function value of the cellular neighbors is calculated. Finally, the cells are updated according to the cellular evolution rules, and the ant search continues after the cell update operation is completed.

[0110] Implementation and process of the algorithm:

[0111] Step 1, input the three-dimensional terrain and meteorological parameters into the MATLAB simulation environment, based on the terrain data and grid processing, and then initialize the pheromone at all nodes on the flight area grid to form an initial pheromone matrix T;

[0112] Step 2: M ants are placed at the starting point A and wait to set off;

[0113] Step 3: Each ant selects the next point on the grid graph according to the state transition rule, and finally reaches the target point, forming a feasible route;

[0114] Step 4, calculate the objective function of the feasible route obtained by each ant to obtain the optimal route solution;

[0115] Step 5, according to the definition of cellular neighbors, determine the cellular neighbors of the optimal solution and calculate the objective function of the cellular neighbors;

[0116] Step 6: According to the cellular evolution rule, the current optimal solution is updated to replace the current optimal route solution;

[0117] Step 7, adjusting the pheromone of each point according to the objective function and the pheromone adjustment criterion;

[0118] Step 8, check the optimal solution and determine whether the pheromone evaporation factor needs to be adjusted. If necessary, adjust the pheromone evaporation factor according to the pheromone update formula;

[0119] Step 9, determine whether the iteration condition is met. If so, the search is completed; if not, return to step B and execute again until the algorithm termination condition is met.

[0120] When planning a single flying robot’s route, it will face various threats (such as three-dimensional terrain, meteorological environment, etc.) and performance limitations. The basic requirements are as follows:

[0121] Minimum track segment length:

[0122] The minimum track length is the shortest distance the aircraft must fly before it starts to change its flight attitude. This limitation depends not only on the maneuverability of the aircraft, but also on the navigation requirements of the aircraft.

[0123] Assume that the flight track consists of and the minimum track segment length is, the constraint can be expressed as:

[0124]

[0125] Maximum turning angle:

[0126] The maximum turning angle restricts the generated trajectory to turn within a range less than or equal to the predetermined maximum turning angle. This constraint depends on the maneuverability of the aircraft and the flight mission.

[0127] Assume that the horizontal projection of track segment i is and the maximum allowable turning angle is, then the constraint can be expressed as:

[0128]

[0129] Where:

[0130] Maximum climb / dive angle:

[0131] This constraint is determined by the maneuverability of the aircraft itself, and it limits the maximum angle of climb and dive of the resulting track in the altitude direction (z).

[0132] Assuming the maximum allowable climb / dive angle is, the constraint can be expressed as:

[0133]

[0134] Maximum track length:

[0135] It constrains the length of the track to be less than or equal to a preset maximum distance, which is determined by the fuel carried by the aircraft and the flight time allowed to reach the target in a specific mission.

[0136] Assume the flight path consists of, the maximum path length is L max , then the constraint can be expressed as:

[0137]

[0138]

[0139] Where:

[0140] Minimum flight altitude:

[0141] If an aircraft flies too low, the probability of colliding with the ground will often increase. Therefore, it is necessary to ensure that the height of the aircraft from the ground is not lower than a certain given height during the flight.

[0142] Assume that the minimum altitude above the ground for the i-th flight path is Hi, and the minimum flight altitude limit is H min , then the constraint can be expressed as:

[0143] Maximum / minimum flight speed:

[0144] The speed of an aircraft on a certain flight trajectory can only be maintained within a limited range, which depends on the aircraft's power system and is also related to the specific environment, spatial position and mission.

[0145] Assume the flight speed of the aircraft is v and the minimum flight speed is v min , the maximum flight speed is v max , then the constraint can be expressed as:

[0146] Track cost function:

[0147]

[0148] According to the above factors that affect the trajectory performance of the aircraft, the trajectory cost function is as follows:

[0149]

[0150] Where:

[0151] represents the length of the i-th track;

[0152] is the altitude of the aircraft;

[0153] is the weight coefficient;

[0154] is the threat index of the i-th track segment.

[0155] Wind farm cost function:

[0156]

[0157] Where:

[0158] The absolute derivative of wind speed with respect to time dWi / dt;

[0159] The superscript l indicates that the local fixed position [xi,yi,xi] T Variation of wind speed with time;

[0160] Assume that the wind field is an approximate frozen wind field δ Wi / δ t= 0, the speed of the flying robot in the wind field is much greater than the local time variation of the wind speed, then:

[0161]

[0162] In the wind field, the inertial velocity vector of the flying robot is Vi can be decomposed into the airspeed vector V and wind speed vector Wi :

[0163]

[0164] Microburst cost function:

[0165] Microburst is the most dangerous form of wind shear. When an aircraft passes through a microburst at low altitude, especially during takeoff and landing, the aircraft speed is very low and the efficiency of each control surface is also low. If a microburst is encountered, it is very dangerous. In the wind shear engineering model of a microburst, the horizontal wind speed and vertical wind speed are expressed as:

[0166]

[0167]

[0168] w xmax Maximum outflow velocity;

[0169] h m and r m Changing the sizes of these three parameters for the corresponding height and radius can yield

[0170] Microbursts of varying intensity and reach;

[0171] h and r are the height and radius of the microburst, respectively.

[0172] Thunderstorm cost function:

[0173] Thunderstorms are common weather phenomena at the turn of spring and summer, and are caused by cumulonimbus clouds with vigorous convection. Thunderstorms include a variety of weather phenomena that endanger flight safety, such as turbulence, turbulence, ice accumulation, lightning strikes and heavy rain, sometimes accompanied by hail, tornadoes, downbursts and low-level wind shear. According to the US Air Force's analysis and statistics on flight accidents caused by meteorological reasons, thunderstorms account for 55% to 60%. The threat level of a thunderstorm can be directly calibrated by the normalization of the impact range in its impact range model. If the threat level assessment factor of a thunderstorm is set to, the threat level of a thunderstorm can be expressed as:

[0174]

[0175] Turbulence cost function:

[0176] Turbulence refers to the rapid and variable airflow in a certain airspace. Its main feature is that there are large differences and sharp changes between the airflow velocity vectors at different locations in a small airspace. Once a flying robot enters such an area, it will not only cause sharp turbulence and difficulty in control, but also the flying robot will be subjected to huge stress at different locations, which may cause damage to the structural strength of the aircraft in severe cases.

[0177]

[0178] Where:

[0179] Ω is the spatial frequency along the x-axis. For a vehicle flying at a speed of v, its relationship with the temporal frequency ω is: Ω = ω / v;

[0180] σ u ,σ v ,σ ｗ is the turbulence intensity in three directions;

[0181] L u ,L v ,L w are the turbulence scales in three directions.

[0182] Aircraft trajectory planning function:

[0183] Assume (x, y, z) is the coordinate of a point in a planning space, where x, y represent longitude and latitude, and z represents altitude. Then the trajectory planning space can be represented as a set:

[0184] Example

[0185] Method flow for weather system feature identification based on Unidata IDV

[0186] A. 3D terrain modeling of the mission area. Use IDV's terrain dataset (such as DEM data) to load the 3D terrain of the mission area. Show the height changes of the surface, which is used to analyze how airflow is affected by terrain. Visualization method: 3D terrain rendering in IDV shows mountains, valleys and other terrain features.

[0187] Unidata IDV's 3D modeling algorithm is as follows:

[0188] make3DMap(MapData,Topography[categories=param:ALTITUDE;param:Elevation;param:Z])

[0189] Where:

[0190] MapData: GIS data of the relevant research area, mainly in *.shp format.

[0191] Topography: Geopotential height@Ground or watersurface in 2D in fnl data

[0192] Modeling effect Figure 1 .

[0193] B. Modeling of the two-dimensional geostrophic wind field in the mission area. Use IDV to load atmospheric data (such as WRF simulated data, GRIB or NetCDF format) to display wind speed and direction. Display the wind field on a two-dimensional map, using a plane diagram to show wind vectors, and arrows to indicate wind speed and direction, usually at a certain altitude or atmospheric level.

[0194] IDV Geostrophic Wind Vector Modeling Algorithm

[0195] The geostrophic wind vector modeling algorithm is as follows:

[0196] geo(D1)

[0197] Where:

[0198] geostrophic_wind_vector (Geostrophic Wind Vectors (from Geopotential_height_isobaric))

[0199] C. Modeling of three-dimensional wind vector field in the mission area. The three-dimensional wind vector field can show the change of wind at different heights. IDV can be displayed by vector streamlines or contour lines.

[0200] 3D True Wind vector algorithm based on IDV

[0201] The algorithm of True Wind vector is as follows:

[0202] DerivedGridFactory.createTrueFlowVectors(D1, D2, D3)

[0203] Where:

[0204] flowvectors3dT (3D True Wind Vectors (from u-component_of_wind_isobaric&v-component_of_wind_isobaric&Vertical_velocity_pressure_isobaric))

[0205] D. Three-dimensional atmospheric boundary layer field modeling in the mission area. The atmospheric boundary layer can be displayed in IDV through contour lines, contour surfaces, etc.

[0206] The PBLH 3D modeling algorithm based on Unidata IDV is as follows:

[0207] PBLH - Color-Filled Contoured Parameter As Topography

[0208] E. Modeling of the two-dimensional temperature field in the mission area. Use the surface temperature data set in IDV to draw the two-dimensional temperature field. Visualization method: flat color map, temperature gradient is represented by different colors, and isotherms show temperature distribution.

[0209] The temperature two-dimensional modeling algorithm based on Unidata IDV is as follows:

[0210] TEMP - Color-Filled Countour Plan View

[0211] Modeling effect Figure 2 .

[0212] F. Modeling of the three-dimensional temperature field in the mission area. The three-dimensional temperature field can display the temperature changes at different heights and analyze them in combination with the terrain. Visualization method: The temperature isosurface is displayed in three dimensions, which can be superimposed with the wind field to show the relationship between temperature changes and airflow.

[0213] The temperature field 3D modeling algorithm based on Unidata IDV is as follows:

[0214] Temperature - @Isobaric surface-Volum Rendering

[0215] Modeling effect Figure 3 .

[0216] G. Modeling of the two-dimensional atmospheric pressure field in the mission area. Use meteorological datasets to display the two-dimensional atmospheric pressure field.

[0217] The two-dimensional modeling algorithm of pressure field based on Unidata IDV is as follows:

[0218] Pressure reduced to MSL - @Mean sea level-Color-Filled Contour PlanView

[0219] H. Modeling of the three-dimensional atmospheric pressure field in the mission area. The three-dimensional data is used to display the changes in atmospheric pressure in the height direction, which helps to analyze the difference in air pressure between high altitude and low altitude. Visualization method: three-dimensional pressure contour map, combined with color coding or contour line display.

[0220] The pressure field 3D modeling algorithm based on Unidata IDV is as follows:

[0221] Pressure(from Geopotential_height_isobaric)- 3D Surface-Contours OverTopography

[0222] Modeling effect Figure 4 .

[0223] I. Modeling of 3D temperature stratification slices in the mission area. Use IDV's slice tool to perform cross-sectional analysis of the 3D temperature field to show the vertical change of temperature. Visualization method: Analyze the stability of the stratification through temperature slices and show the change of temperature at different heights.

[0224] The atmospheric temperature stratification (Temperature) modeling algorithm based on IDV is: Temperature_isobaric. Where:

[0225] Temperature_isobaric - Color-Filled Contour Cross Section.

[0226] Modeling effect Figure 5 .

[0227] J. Comprehensive analysis model of meteorological factors. Multiple meteorological elements (such as temperature, wind speed, air pressure, etc.) are superimposed for analysis. Visualization method: Multiple layers are superimposed in the same view to show the interaction of multiple factors such as wind, temperature and air pressure.

[0228] The time series effect of comprehensive analysis of meteorological factors based on IDV is shown in Figure 6 .

[0229] The advantages of using Unidata IDV to build a complex weather diagnosis system to obtain meteorological parameters for flight robot trajectory planning are:

[0230] 1. Real-time and accuracy: Weather diagnostic analysis technology relies on actual observation data, such as satellite remote sensing, weather radar, ground weather stations and high-altitude balloons. These observation data can reflect the true state of the atmospheric environment in real time. Therefore, when analyzing the current atmospheric environment, weather changes and extreme meteorological events, they have extremely high accuracy and timeliness.

[0231] 2. Accurately capture atmospheric phenomena: This technology can capture detailed atmospheric phenomena, such as temperature gradients, humidity distribution, wind fields, air pressure changes, etc., and provide high-resolution meteorological parameters. This accuracy is very helpful for analyzing the formation and development process of extreme weather events (such as heavy rains, thunderstorms, hurricanes, etc.).

[0232] 3. Rich diagnostic indicators: Weather diagnostic analysis technology has rich diagnostic indicators and can conduct detailed analysis of weather systems, such as analysis of fronts, low pressure, and high pressure systems, tracking the formation and path of tropical cyclones, and monitoring the structure of the atmosphere and air pollution.

[0233] 4. Identification of extreme weather events: Weather diagnostic analysis technology is good at identifying and tracking extreme weather events, such as hurricanes, tornadoes, rainstorms, etc. By analyzing parameters such as atmospheric circulation, atmospheric instability, and vertical velocity, it can provide early warning of extreme weather and provide decision-making support for disaster prevention and mitigation.

[0234] 5. Analysis of the development process of extreme weather events: It can not only identify extreme weather, but also analyze its development process, for example, it has significant advantages in typhoon path forecasting, thunderstorm probability assessment, etc. This is of great significance for disaster warning and emergency response.

[0235] 6. Data sources are extensive and reliable, and multi-source data is integrated: Weather diagnostic analysis technology uses the global meteorological data network to integrate satellite, radar, ground observation, high-altitude meteorology and other multi-source data to form a comprehensive and accurate atmospheric environment analysis. This fusion of multi-source data makes the analysis results more reliable and comprehensive.

[0236] Based on the principle of weather diagnosis simulation, the present invention selects the mesoscale atmospheric numerical model WRF4 (Weather Research and Forecasting Model) to construct a complex weather model of the mission area, and uses Unidata IDV to build a complex weather diagnosis system. The system is integrated into the Matlab simulation platform to perform a comprehensive simulation of the flight of the flying robot and realize the flight path planning of the flying robot based on weather diagnosis technology.

Claims

1. A flying robot route planning method based on weather diagnosis technology, comprising: Step 1: Input the three-dimensional terrain and meteorological parameters of the mission area into the MATLAB simulation environment, perform grid processing based on the terrain data, and then initialize the pheromone at all nodes on the flight area grid to form an initial pheromone matrix; Step 2, place multiple ants at the starting point and wait for departure; Step 3: Each ant selects the next point of the grid according to the state transition rule and eventually reaches the target point, forming a feasible route; Step 4, calculate the objective function of the feasible route obtained by each ant to obtain the optimal route solution; Step 5, according to the definition of cellular neighbors, determine the cellular neighbors of the optimal solution and calculate the objective function of the cellular neighbors; Step 6: According to the cellular evolution rule, the current optimal solution is updated to replace the current optimal route solution; Step 7, adjusting the pheromone of each point according to the objective function and the pheromone adjustment criterion; Step 8, check the optimal solution and determine whether the pheromone evaporation factor needs to be adjusted. If necessary, adjust the pheromone evaporation factor according to the pheromone update formula; Step 9, determine whether the iteration condition is met. If so, the search is completed; if not, return to step 2 and re-execute until the algorithm termination condition is met to obtain the flight robot route. The characteristic is that the meteorological parameters are obtained based on the Unidata IDV weather diagnostic system according to the following steps: A. Meteorological data collection and preprocessing: collect meteorological observation data in the mission area, perform quality control and clean up the data, and perform spatial interpolation and gridding on the data; B. Establish a three-dimensional terrain model of the mission area, analyze the three-dimensional terrain structure of the mission area, and display the characteristics of terrain height, mountains, and valleys to diagnose the impact of terrain on local meteorological conditions; C. Establish a two-dimensional geostrophic wind field model for the mission area, analyze the horizontal distribution of the geostrophic wind field, and display the wind speed and direction at different heights to diagnose large-scale wind field dynamics; D. Establish a three-dimensional wind vector field model for the mission area, analyze the changes in the wind field in three-dimensional space, and display the wind direction, wind speed and vertical movement at different heights for diagnosing the vertical wind field; E. Establish a three-dimensional atmospheric boundary layer field model for the mission area, analyze the temperature, humidity, and wind speed parameters of the atmospheric boundary layer, and display the vertical changes within the boundary layer to diagnose weather conditions near the surface; F. Establish a two-dimensional temperature field model for the mission area, analyze the horizontal distribution of surface temperature, and display the temperature differences within the mission area for diagnosis of local thermal processes; G. Establish a three-dimensional temperature field model for the mission area, analyze the temperature distribution at different heights, and display the temperature variation with height, which is used to diagnose the temperature structure in three-dimensional space; H. Establish a two-dimensional atmospheric pressure field model for the mission area, analyze the horizontal distribution of surface atmospheric pressure, and display the pressure changes in the mission area for diagnosis of weather systems; I. Establish a three-dimensional atmospheric pressure field model for the mission area, analyze the distribution of atmospheric pressure in three-dimensional space, and display the changes in air pressure at different altitudes to diagnose the three-dimensional distribution of air pressure in the atmosphere; J. Establish a three-dimensional temperature stratification slice model for the mission area, analyze the temperature stratification structure of the mission area, and display the temperature variation slices with height; used to diagnose the stability of temperature stratification; K. Establish a comprehensive analysis model for meteorological factors, conduct superposition analysis on all the above meteorological factors, comprehensively display the interactions among various factors, evaluate the synergistic effects of multiple meteorological factors through the comprehensive analysis model, generate meteorological diagnostic data, and provide meteorological parameters for the mission area.

2. The flying robot route planning method based on weather diagnosis technology according to claim 1 is characterized by: According to the requirements of the flight mission objectives of the flying robot, the meteorological diagnostic data is input into the numerical weather forecast model to simulate the future weather evolution process, and then generate the meteorological parameters of all the passing points of the flying robot.

3. The flying robot route planning method based on weather diagnosis technology according to claim 1 is characterized by: The objective functions of the feasible route include a track cost function, a wind field cost function, a microburst cost function, a thunderstorm cost function, and a turbulence cost function.

4. The flying robot route planning method based on weather diagnosis technology according to claim 1 is characterized by: The two-dimensional geostrophic wind field model of the mission area, the three-dimensional wind vector field model of the mission area, the three-dimensional atmospheric boundary layer field model of the mission area, the two-dimensional temperature field model of the mission area, the three-dimensional temperature field model of the mission area, the two-dimensional atmospheric pressure field model of the mission area, the three-dimensional atmospheric pressure field model of the mission area, and the three-dimensional temperature stratification slice model of the mission area are weather diagnosis models constructed based on the weather model constructed using WRF4 and are constructed using Unidata IDV secondary development.

5. The method for flight robot route planning based on weather diagnosis technology according to any one of claims 1 to 4, characterized in that: During the flight of the flying robot, it receives real-time meteorological observation data of the mission area, uses radar and satellite real-time monitoring data, tracks the evolution of the weather system, updates the meteorological parameters of the mission area in real time, and corrects the future route of the flying robot based on the current meteorological parameters of the mission area.

Images