Intelligent flight route planning method and system for low-altitude logistics aircraft

Abstract

The invention discloses an intelligent flight route planning method and system for a low-altitude logistics aircraft, and the method comprises the steps: obtaining the three-dimensional space map data of the current position of the aircraft, including the coordinate information of a building and a telegraph pole, and the boundary range of a no-fly zone; acquiring wind speed, wind direction and rainfall data of the current area from a meteorological data source, and calculating meteorological impact factors by combining aircraft performance parameters; dynamically adjusting the candidate flight routes according to the maneuvering characteristic parameters of the aircraft, and calculating the flight time, the energy consumption cost and the safety risk score of each route; meteorological changes in the flight process are monitored in real time, and if the meteorological conditions exceed a preset threshold value, the meteorological data are obtained again, and the meteorological influence factors are updated; and carrying out local optimization adjustment on the current flight route according to the updated meteorological influence factor, generating a new flight route segment and replacing the original segment.

Description

Technical Field

[0001] The present invention belongs to the field of information technology, and in particular relates to a flight route intelligent planning method and system for low-altitude logistics aircraft. Background Art

[0002] Flight route planning for low-altitude logistics aircraft presents a technical dilemma: how to optimize flight time and cost while ensuring flight safety. Specifically, aircraft must avoid obstacles and no-fly zones in three-dimensional space while adjusting altitude and speed based on real-time weather data. However, while traditional path-finding algorithms can generate multiple possible flight routes, they often struggle to balance multi-objective optimization requirements when faced with complex three-dimensional environments and dynamically changing weather conditions.

[0003] For example, when an aircraft needs to fly over densely populated urban areas, obstacles such as buildings and utility poles can significantly restrict its flight path. In this case, the path-finding algorithm must precisely calculate the aircraft's maneuverability to ensure it can flexibly turn, ascend, and descend within the confined space. However, this precise calculation often increases flight time, thereby increasing flight costs. Alternatively, to shorten flight time, the aircraft may choose a higher altitude, but this increases the risk of encountering inclement weather.

[0004] Furthermore, the introduction of real-time weather data further exacerbates this conflict. Aircraft must adjust their flight paths in real time based on weather conditions such as wind speed, direction, and rainfall to ensure flight safety. However, frequent route adjustments not only increase computational complexity but can also cause aircraft to deviate from their optimal routes, extending flight time and increasing costs. Therefore, rapidly generating safe and cost-effective flight routes in a dynamically changing environment has become a pressing technical challenge. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for intelligent flight route planning for low-altitude logistics aircraft.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A method for intelligent flight route planning for low-altitude logistics aircraft, comprising:

[0008] Get the three-dimensional spatial map data of the aircraft's current location;

[0009] Obtain wind speed, wind direction and rainfall data for the current area from meteorological data sources and calculate meteorological impact factors;

[0010] Obstacle coordinate information, no-fly zone boundaries, and meteorological factors are input into a deep reinforcement learning path search algorithm to generate multiple candidate flight routes.

[0011] Dynamically adjust candidate flight routes based on aircraft maneuverability parameters, and calculate the flight time, energy cost, and safety risk score for each route;

[0012] Evaluate the adjusted flight routes and select the optimal flight route with the shortest flight time, lowest energy consumption and minimum safety risk;

[0013] Real-time monitoring of weather changes during flight. If weather conditions exceed preset thresholds, weather data will be re-acquired and weather impact factors will be updated.

[0014] Perform local optimization and adjustment on the current flight route based on the updated meteorological impact factors, generate new flight route segments and replace the original segments;

[0015] The aircraft's position is continuously monitored relative to obstacles and no-fly zones. If a potential collision risk is detected, the emergency obstacle avoidance algorithm is triggered to generate an obstacle avoidance path.

[0016] Preferably, the three-dimensional space map data includes: coordinate information of buildings and utility poles and the boundary range of the no-fly zone.

[0017] Preferably, the wind speed, wind direction and rainfall data of the current area are obtained from the meteorological data source, and the meteorological impact factor is calculated in combination with the aircraft performance parameters.

[0018] As a preference, a multi-objective optimization algorithm is used to evaluate the adjusted flight route to screen out the optimal flight route with the shortest flight time, lowest energy consumption and lowest safety risk.

[0019] The present invention also provides a flight route intelligent planning system for low-altitude logistics aircraft, comprising:

[0020] Position information acquisition module, used to obtain the three-dimensional space map data of the current location of the aircraft;

[0021] Meteorological data processing module, used to obtain wind speed, wind direction and rainfall data of the current area from the meteorological data source and calculate meteorological impact factors;

[0022] The path generation module is used to input obstacle coordinate information, no-fly zone boundary range and meteorological influencing factors into the deep reinforcement learning path search algorithm to generate multiple candidate flight routes;

[0023] The path adjustment module is used to dynamically adjust candidate flight routes based on the aircraft's maneuverability parameters and calculate the flight time, energy cost, and safety risk score of each route;

[0024] The path evaluation module is used to evaluate the adjusted flight route and select the optimal flight route with the shortest flight time, lowest energy consumption and minimum safety risk;

[0025] The weather monitoring module is used to monitor weather changes during flight in real time. If weather conditions exceed preset thresholds, weather data is retrieved and weather impact factors are updated.

[0026] The path optimization module is used to locally optimize the current flight route based on the updated meteorological factors, generate new flight route segments and replace the original segments;

[0027] The obstacle avoidance module continuously monitors the relative distance between the aircraft and obstacles and no-fly zones. If a potential collision risk is detected, an emergency obstacle avoidance algorithm is triggered to generate an obstacle avoidance path.

[0028] Preferably, the three-dimensional space map data includes: coordinate information of buildings and utility poles and the boundary range of the no-fly zone.

[0029] Preferably, the meteorological data processing module obtains the wind speed, wind direction and rainfall data of the current area from the meteorological data source, and calculates the meteorological impact factor in combination with the aircraft performance parameters.

[0030] Preferably, the path evaluation module uses a multi-objective optimization algorithm to evaluate the adjusted flight route and screen out the optimal flight route with the shortest flight time, lowest energy consumption and lowest safety risk.

[0031] The present invention obtains three-dimensional spatial map data of the aircraft's location, including the coordinates of buildings, utility poles, and no-fly zone boundaries, combines real-time meteorological data and aircraft performance parameters, and uses a deep reinforcement learning algorithm to generate multiple candidate flight routes. Subsequently, the present invention dynamically adjusts the route based on the aircraft's maneuverability and uses a multi-objective optimization algorithm to evaluate and select the optimal flight route. During the flight, the present invention continuously monitors meteorological changes and potential collision risks, updates meteorological influencing factors in a timely manner, optimizes local paths, and triggers emergency obstacle avoidance algorithms when necessary. This method can plan a safe, efficient, and energy-saving optimal flight path for aircraft in complex environments, significantly improving the execution efficiency and safety of flight missions. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 The figure is a flow chart of the intelligent flight route planning method for low-altitude logistics aircraft according to the present invention. DETAILED DESCRIPTION

[0033] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0034] Example 1:

[0035] like Figure 1 As shown, an embodiment of the present invention provides a flight route intelligent planning method for a low-altitude logistics aircraft, including:

[0036] S101: Acquire three-dimensional spatial map data of the current location of the aircraft, including coordinate information of buildings and utility poles and the boundary range of the no-fly zone.

[0037] The aircraft's three-dimensional position data is acquired through a geographic information system (GIS) and combined with spatial information to generate a three-dimensional spatial map. Image recognition algorithms are used to extract the coordinates of buildings and utility poles from the spatial map. Based on pre-set no-fly zone rules, the no-fly zone boundaries are delineated and regional demarcation data is generated. If the aircraft's position data overlaps with the no-fly zone boundary data, a no-fly warning mechanism is triggered. Based on the coordinates of buildings and utility poles, flight path planning data is generated. Machine learning algorithms are used to optimize the flight path to avoid collisions with buildings and utility poles. The spatial map and optimized flight path data are used to update the aircraft's navigation information in real time.

[0038] Specifically, the geographic information system (GIS) acquires the real-time location of aircraft using multi-source data acquisition devices such as satellite positioning and lidar, simultaneously recording longitude, latitude, and altitude data. For example, within an urban building complex, an aircraft equipped with a laser scanner can scan and capture 3D point cloud data of surrounding buildings. This data, combined with satellite positioning, forms a high-precision 3D map. Once the spatial map is generated, semantic segmentation algorithms from deep learning are used to identify building outlines. For example, by extracting building features using a convolutional neural network, it can be determined that the spatial coordinates of a high-rise building are at a certain degree and minute east longitude, a certain degree and minute north latitude, and a certain height of meters. For utility poles, object detection algorithms are used to extract their location information, generating coordinate point array data for the pole. Regarding no-fly zone demarcation, circular or polygonal no-fly zones of varying radii are established based on the spatial distribution of sensitive areas such as airports, military facilities, and critical infrastructure. For example, a circular no-fly zone with a radius of 500 meters can be designated around a substation. If an aircraft's real-time location enters this zone, the system immediately issues an alarm and requires the aircraft to adjust its course. Path planning utilizes an improved artificial potential field method, treating buildings and utility poles as sources of repulsive fields. For example, if a tall building is detected 300 meters ahead, a detour is automatically planned while maintaining a safe distance. Using reinforcement learning, the system continuously optimizes its path selection strategy based on historical flight data, selecting the optimal route while ensuring safety. Navigation information updates utilize a Kalman filter algorithm, integrating multi-dimensional data such as position, velocity, and attitude. For example, when the aircraft is over a commercial area, the system receives real-time position feedback and dynamically adjusts altitude and speed based on the distribution of buildings. Simultaneously, the optimized path information is converted into flight instructions to ensure the aircraft remains on a safe course. In complex scenarios, the system must coordinate multiple tasks. For example, suppose the aircraft needs to traverse a densely populated area. First, the building distribution is determined through 3D modeling. Then, all possible obstacles, including tall buildings and power facilities, are identified. Finally, a safe and efficient flight path is planned while avoiding no-fly zones. Throughout this process, data from each subsystem is continuously exchanged to ensure flight safety.

[0039] S102: Obtain wind speed, wind direction, and rainfall data for the current area from a meteorological data source, and calculate meteorological impact factors in combination with aircraft performance parameters.

[0040] The system obtains wind speed, wind direction, and rainfall data for the current area from meteorological data sources and matches this data with the wind and rain resistance capabilities in the aircraft's performance parameters. Using a preset formula, the system uses wind speed, wind direction, and rainfall data as input variables, combined with the wind and rain resistance coefficients in the aircraft's performance parameters, to calculate the meteorological impact factor. Based on the numerical range of the meteorological impact factor, it determines whether the meteorological conditions in the current area are suitable for flight. If the meteorological impact factor exceeds the preset threshold, a meteorological warning signal is generated. A path optimization algorithm is used to combine the meteorological warning signal with the aircraft's current position data to generate a new flight path. The aircraft's heading and altitude are adjusted by updating the flight path data in real time. The aircraft's navigation system data is updated based on the adjusted flight path.

[0041] Specifically, meteorological data collection plays a crucial role in drone flight safety. Wind speed data is collected through weather stations and radiosondes. For example, in a certain city's commercial district, the wind speed is five meters per second, the wind direction is 45 degrees northeast, and the rainfall is five millimeters per hour. This data needs to be matched with the drone's own parameters, such as a maximum wind resistance rating of seven and a rain resistance rating of moderate rain. A typical scenario for calculating meteorological impact factors is logistics delivery. In a city with high-rise buildings, at a wind speed of eight meters per second, the drone's wind resistance coefficient is 0.7, resulting in a meteorological impact factor of 5.6, exceeding the safety threshold of 5.0. The system immediately generates a warning signal, indicating the need to adjust the flight altitude or reroute. Route optimization involves multiple considerations. For example, in a delivery scenario in a residential area, the original planned flight altitude was 30 meters. However, encountering strong winds, the system automatically adjusted the route to a lower altitude of 20 meters, avoiding the higher wind speeds at higher altitudes. Furthermore, due to rainfall, the system deviated from the original trajectory and chose to fly within areas shaded by buildings to minimize the impact of rain on onboard equipment. The practical application of weather warning mechanisms is evident in emergency response. For example, during a delivery mission in a suburban area, a sudden thunderstorm occurred, and the weather impact factor rapidly climbed to 8.3. The system immediately generated data for alternative landing sites, guiding the drone to a nearby safe area. This warning mechanism effectively prevents flight accidents caused by inclement weather. Navigation system data updates are the last line of defense for flight safety. For example, during a drone inspection at a scenic spot, when encountering crosswind gusts, the system adjusts the aircraft's attitude in real time based on wind direction data. For example, if the wind direction is 60 degrees, the aircraft's yaw angle is adjusted to offset the wind's impact and maintain a stable flight trajectory. This dynamic adjustment ensures mission continuity and safety. The comprehensive application of meteorological data is particularly important in densely built-up urban areas. When encountering turbulent airflow, the system predicts possible areas of airflow disturbance based on building distribution and meteorological data. By adjusting flight altitude and speed in real time, the drone avoids these unstable airflow areas and ensures flight safety. This predictive path planning significantly enhances the drone's adaptability in complex environments.

[0042] S103: Input obstacle coordinate information, no-fly zone boundary range, and meteorological influencing factors into a deep reinforcement learning path search algorithm to generate multiple candidate flight routes.

[0043] Obstacle coordinate information, no-fly zone boundaries, and meteorological factors are obtained and input into the deep reinforcement learning path search algorithm. The deep reinforcement learning path search algorithm combines the obstacle coordinate information, no-fly zone boundaries, and meteorological factors to generate multiple candidate flight routes. A path evaluation model is used to score the multiple candidate flight routes, generating a comprehensive score for each route. Based on the comprehensive score of the path evaluation model, the candidate flight route with the highest score is selected as the final flight route. The final flight route is input into the flight control system, and the aircraft's heading and altitude are adjusted. The aircraft's position and meteorological data are monitored in real time. If the meteorological factor exceeds a preset threshold, the deep reinforcement learning path search algorithm is restarted. The updated deep reinforcement learning path search algorithm generates new candidate flight routes, and the path evaluation and selection process repeats.

[0044] Specifically, the deep reinforcement learning path-finding algorithm learns the optimal strategy through the interaction between the agent and the environment. This algorithm uses a neural network as a value function approximator and learns the optimal path through trial and error. In specific applications, the flight environment can be defined as a state space, containing information such as obstacle coordinates and no-fly zone locations. The aircraft's maneuvers, such as turning, acceleration, and deceleration, are then defined as the action space. The reward function is designed based on flight safety and efficiency. To process obstacle coordinate information, a gridding approach can be used to divide the flight space into several cells, each marked with its occupancy status. For example, the airspace above a city can be divided into a 100-meter grid. Obstacles such as buildings and towers can be 3D modeled and their locations within the grid determined. The boundaries of a no-fly zone can be represented by a polygonal area. For example, a five-kilometer no-fly zone around a military base can be converted into grid occupancy information. The input of meteorological factors should consider the combined effects of wind speed, direction, and rainfall on the aircraft. Assume a certain area has a wind speed of 8 meters per second from the northeast, and moderate rainfall. Considering the aircraft's wind and rain resistance, the calculated weather impact factor for this area is 0.7. This value indicates that current weather conditions have some impact on flight, but are still manageable. When generating candidate flight routes, the algorithm simultaneously considers multiple objectives: minimizing route length, avoiding obstacles and no-fly zones, and minimizing the number of sections affected by weather. For example, five different candidate routes might be generated from the starting point to the destination, each meeting basic safety requirements. The path evaluation model scores these candidate routes on multiple dimensions, including flight distance, energy consumption, and safety margin. For example, the first route, while shortest, passes through high winds and has a lower overall score. The third route, while longer, avoids all adverse weather conditions and achieves the highest overall score. The real-time monitoring system continuously collects aircraft position and weather data. When severe convective weather is detected in the area ahead and the weather impact factor rises to 0.9, route replanning is triggered. The algorithm regenerates candidate routes based on updated environmental information to ensure flight safety. This dynamic adjustment mechanism enables the aircraft to flexibly respond to complex and changing flight environments.

[0045] S104. Dynamically adjust the candidate flight routes based on the aircraft's maneuverability performance parameters, and calculate the flight time, energy consumption cost, and safety risk score of each route.

[0046] Obtain the aircraft's maneuverability performance parameters and a set of candidate flight routes, and use the dynamic adjustment method to modify each route. Calculate the flight time, energy consumption, and risk value of each modified route using the performance parameter route set to generate a score. Use a preset scoring threshold to determine the score of each route, and filter out the candidate route set that meets the threshold. Sort the filtered candidate route set according to the score, and determine the optimal solution as the final flight route. Input the final flight route into the flight control system, and monitor the aircraft's maneuverability and environmental data in real time. If the monitored flight time or energy consumption value exceeds the preset range, restart the dynamic adjustment method to modify the route set. By recalculating the score value of the modified route, update the optimal solution and adjust the execution parameters of the flight control system.

[0047] Specifically, the dynamic adjustment method is a route optimization method based on aircraft maneuverability, achieving route correction by adjusting flight parameters. Specifically, this method modifies candidate routes based on parameters such as the aircraft's maximum climb rate and turning radius. For example, if the aircraft's climb rate is 10 meters per second and its minimum turning radius is 50 meters, and the original route contains maneuvers that exceed these limits, the dynamic adjustment method will adjust the route to meet the maneuverability constraints. In calculating the score, the system comprehensively considers factors such as flight time, energy consumption, and risk. Flight time can be calculated based on route length and average speed, while energy consumption is related to parameters such as engine power and flight altitude. The risk value primarily considers factors such as distance to obstacles and weather conditions. For example, if a route is 10 kilometers long, has an average speed of 100 kilometers per hour, fuel consumption of 0.5 liters per kilometer, and a minimum distance to obstacles of 200 meters, these parameters are weighted to produce a comprehensive score. Setting the scoring threshold requires balancing multiple performance indicators. For example, flight time must not exceed 20% of the scheduled time, energy consumption must not increase by more than 15% of the standard value, and the distance to obstacles must be at least 150% of the safe clearance. These thresholds can screen candidate routes that meet basic requirements. The optimal route is determined using a multi-objective optimization approach. For example, suppose an aircraft needs to complete a city logistics delivery mission. The system generates three candidate routes: Route 1 offers shorter flight time but higher energy consumption; Route 2 offers lower energy consumption but longer flight time; and Route 3 strikes a balance between the two. The route that best suits the mission requirements is selected based on a ranking of scores. A real-time monitoring system continuously tracks the aircraft's status. If flight time or energy consumption exceeds expectations, such as a 20% increase in fuel consumption due to headwind, the system initiates a route correction. This correction process re-evaluates all available routes from the current location to the destination and adjusts the route based on updated maneuverability parameters. Parameter adjustments involve multiple control steps. For example, if energy consumption is detected to be excessively high, the system may lower the flight altitude to reduce energy consumption. This adjustment must consider safety and efficiency at the new altitude to ensure that the adjusted execution parameters still meet mission requirements. This dynamic optimization mechanism ensures that the aircraft always maintains optimal flight status.

[0048] S105. Use a multi-objective optimization algorithm to evaluate the adjusted flight route and select the optimal flight route with the shortest flight time, lowest energy consumption and lowest safety risk.

[0049] Obtain the aircraft's maneuverability performance parameters and candidate route set, and use the dynamic adjustment method to modify each route. Calculate the flight time, energy consumption, and risk value of each modified route using the performance parameters and route set to generate a score. Use a preset scoring threshold to determine the score of each route, and filter out the candidate route set that meets the threshold. Sort the filtered candidate route set according to the score, and determine the optimal solution as the final flight route. Input the final flight route into the flight control system, and monitor the aircraft's maneuverability and environmental data in real time. If the monitored flight time or energy consumption value exceeds the preset range, restart the dynamic adjustment method to modify the route set. By recalculating the score value of the modified route, update the optimal solution and adjust the execution parameters of the flight control system.

[0050] Specifically, aircraft maneuverability parameters include key indicators such as turning radius, speed range, and climb rate. For example, a certain drone model has a minimum turning radius of 30 meters, an adjustable speed between 50 and 150 kilometers per hour, and a maximum climb rate of 8 meters per second. These parameters directly impact the feasibility of the flight route. When modifying the original route through dynamic adjustment methods, the constraints of the actual flight environment must be taken into account. For example, when encountering densely built-up areas, the turning radius should be adjusted to at least 40 meters to ensure a safety margin. In the scoring system, flight time primarily considers route length and average speed. For a 15-kilometer route with an expected speed of 100 kilometers per hour, the baseline flight time is approximately nine minutes. Energy consumption is evaluated comprehensively based on speed and altitude variations. For example, energy consumption during a climb is approximately twice that of cruising. Risk assessments take into account factors such as terrain complexity and weather conditions. The risk factor for crossing mountainous areas can be set at 1.5 times that of plains. When setting scoring thresholds, flight time is typically kept within 20 percent of the expected value, energy consumption does not exceed 70 percent of full payload capacity, and the risk score is no higher than 80 points. The selected candidate routes are ranked by comprehensive score, and the one with the highest score is selected as the final option. For example, if three feasible routes are selected for a certain mission, with scores of 92, 85, and 78, respectively, the route with a score of 92 will be selected for execution. The flight control system continuously monitors maneuverability and environmental data during execution. If it detects that the flight time is about to exceed expectations, such as when only 70% of the planned nine-minute flight has been completed after eight minutes, the system automatically triggers a route correction. By adjusting flight parameters for the remaining legs, such as increasing speed within a safe range or selecting a shorter alternative route, the mission is completed on schedule. If energy consumption monitoring shows that the remaining battery power is less than 40%, the energy-saving route is prioritized, and the flight altitude and speed are reduced if necessary.

[0051] S106. Real-time monitoring of weather changes during the flight. If weather conditions exceed a preset threshold, reacquire weather data and update weather impact factors.

[0052] The data acquisition module acquires real-time meteorological data from meteorological sensors in the flight area. Data sources include temperature, air pressure, and wind speed. In the meteorological data processing module, the meteorological impact factor is calculated in conjunction with a pre-established meteorological model to determine whether meteorological conditions exceed a preset threshold. If so, the data update module is activated to re-collect meteorological data and update the meteorological impact factor to generate new meteorological data. Based on the updated meteorological data, a meteorological forecasting algorithm is used to calculate the meteorological trend over the next period of time to determine whether meteorological conditions are stable. If the meteorological conditions are stable, a path planning algorithm is used in conjunction with the meteorological impact factor to generate a new flight path and determine whether the path meets safety standards. If the path meets safety standards, a path optimization algorithm is used to optimize the flight path to obtain the final flight path plan. The final flight path plan is input into the flight control system, which monitors meteorological changes during flight in real time to determine whether meteorological data needs to be updated.

[0053] Specifically, the real-time meteorological data collection system utilizes a complementary layout of different sensors, forming a multi-layered, three-dimensional monitoring network within the flight area. The temperature sensor utilizes a thermocouple type with a measurement range of -40°C to -80°C and an accuracy of 0.1°C. The pressure sensor utilizes a piezoresistive type with a measurement range of 0 to 100 kilopascals and a resolution of 0.01 kilopascals. The wind speed sensor utilizes an ultrasonic measurement principle with a measurement range of 0 to 70 meters per second and an accuracy of 0.1 meter per second. During meteorological data processing, various meteorological factors are comprehensively considered. The temperature factor primarily affects aircraft engine performance. For example, for every 10°C increase in temperature, engine thrust decreases by approximately 3%. The pressure factor affects lift coefficient: for every 100°C decrease in pressure, lift decreases by approximately 1%. The wind speed factor also impacts flight stability; altitude adjustment is required when crosswind components exceed 15 meters per second. Data updates utilize a sliding time window mechanism, with a ten-minute window length and a one-minute update interval. Data updates are automatically triggered when a temperature change rate exceeding one degree per minute, an air pressure change rate exceeding 0.5 kilopascals per minute, or a wind speed change exceeding five meters per second per minute is detected. The weather forecast algorithm builds a weather change model based on historical data statistics and short-term trend extrapolation. For example, by analyzing wind speed variations over the past six hours and combining them with current measured data, the wind speed trend for the next hour can be predicted with an accuracy of 85%. Path planning prioritizes the spatial distribution of meteorological influences. In areas with large temperature gradients, a higher flight altitude is planned to prevent engine overheating. When strong local airflow exists, the density of waypoints is increased to avoid areas of airflow disturbance. Safety standards are assessed across multiple dimensions, including minimum weather separations and restricted areas. During the path optimization process, meteorological influence factors are comprehensively scored. The temperature impact is weighted 0.3, the pressure impact 0.2, and the wind speed impact 0.5. An iterative optimization method selects the flight path with the optimal score while maintaining safety margins. The flight control system employs an adaptive control strategy, dynamically adjusting control parameters based on real-time weather changes. When the wind speed suddenly changes to more than five meters per second, the system automatically increases the attitude control gain to improve flight stability. A weather data anomaly detection mechanism is also established. If the data exceeds the normal range three times in a row, the weather data re-collection process is triggered.

[0054] S107: Perform local optimization and adjustment on the current flight route according to the updated meteorological impact factors, generate a new flight route segment and replace the original segment.

[0055] The data acquisition module is used to obtain real-time meteorological data, including temperature, air pressure, and wind speed, from meteorological sensors. In the meteorological data processing module, the meteorological impact factor is calculated in combination with a pre-established meteorological model to determine whether it exceeds the preset threshold. If the meteorological conditions exceed the preset threshold, the data update module is activated to re-collect meteorological data and update the meteorological impact factor. Based on the updated meteorological impact factor, the path optimization algorithm is used to locally optimize the current flight path and generate a new path segment. It is determined whether the newly generated path segment meets the preset safety standards. If so, the original path segment is replaced. A meteorological forecast algorithm is used to calculate the meteorological change trend over a period of time in the future to determine whether the meteorological conditions are stable. If the meteorological conditions are stable, the optimized flight path segment is input into the flight control system to monitor meteorological changes during the flight in real time.

[0056] Specifically, for meteorological data collection, a real-time meteorological sensor network typically consists of multiple sensor nodes distributed throughout the flight area. For temperature sensors, for example, thermal resistor sensors can be deployed at monitoring points every 500 meters throughout the flight area, enabling detailed temperature monitoring with a measurement accuracy of 0.1 degrees Celsius. Pressure sensors use piezoresistive sensors with a measurement range of 800 to 1,100 hPa. Wind speed monitoring uses ultrasonic anemometers with a measurement range of 0 to 70 meters per second. The meteorological data processing module, combined with a pre-set meteorological model, calculates the impact factors, taking into account the combined effects of temperature, air pressure, and wind speed on flight. For example, when the temperature exceeds 35 degrees Celsius, the reduced air density affects the lift coefficient; when the air pressure drops below 900 hPa, engine performance is affected; and when the wind speed exceeds 15 meters per second, the flight attitude is significantly affected. These factors are comprehensively evaluated using weighting coefficients to form the meteorological impact factor. When a significant change in meteorological conditions in a particular area is detected, the data update module is activated. For example, if the wind speed in a particular area increases from 8 to 18 meters per second within ten minutes, exceeding a preset threshold, the system will immediately recollect meteorological data for that area and surrounding areas. This newly collected data is used to update the meteorological influencing factors, ensuring that flight decisions are based on the latest weather conditions. Regarding route optimization, the system will make local adjustments to the flight path based on the updated meteorological influencing factors. For example, if the original route requires passing through an area with high wind speeds, the optimization algorithm will plan a new route segment that avoids the strong winds, while ensuring that the flight distance does not increase by more than 10 percent. The newly generated route segment must meet safety standards, such as a minimum distance from terrain obstacles of at least 500 meters and a minimum separation distance from other aircraft of at least 1,000 meters. The weather forecast algorithm analyzes historical data and current weather trends to predict weather changes within the next 30 minutes. For example, by analyzing the rate of change of air pressure and temperature gradients, it can predict the occurrence of localized convective weather. If the forecast indicates relatively stable weather conditions within the next 30 minutes—meaning that the temperature change does not exceed 5 degrees Celsius, the air pressure change does not exceed 20 hPa, and the wind speed change does not exceed 5 meters per second—the optimized flight path is input into the flight control system for execution. During actual flight, the system continuously monitors weather changes along the route. For example, during a cross-regional flight, when an aircraft enters a mountainous area from a plain, topographical shift can significantly alter local weather conditions. The system assesses the impact of these changes on flight safety in real time and, if necessary, triggers a new round of route optimization. This dynamic optimization mechanism effectively responds to complex and changing weather conditions and ensures flight safety.

[0057] S108. Continuously monitor the relative distance between the aircraft position and obstacles and no-fly zones. If a potential collision risk is detected, trigger an emergency obstacle avoidance algorithm to generate an obstacle avoidance path.

[0058] The positioning module obtains the real-time position coordinates of the aircraft, and the coordinates of obstacles and no-fly zone boundaries are obtained through the geographic information system. A distance calculation algorithm is used to calculate the relative distance between the aircraft's position coordinates and the coordinates of obstacles and no-fly zone boundaries. The calculated relative distance is compared with the preset safety threshold to determine whether there is a potential collision risk. If the relative distance is less than the preset safety threshold, the obstacle avoidance path generation algorithm is invoked to calculate a new flight path. The aircraft's flight trajectory planning data is updated based on the new path coordinates output by the obstacle avoidance path generation algorithm. A path smoothing algorithm is used to optimize the generated obstacle avoidance path to ensure that it meets the aircraft's kinematic constraints. The optimized obstacle avoidance path data is input into the flight control system to complete the aircraft trajectory adjustment.

[0059] Specifically, the positioning module utilizes a dual-mode Beidou and GPS system, achieving sub-meter accuracy in open areas and meter-level accuracy in densely built-up urban areas. Combined with an inertial navigation system, accurate positioning is maintained even when satellite signals are temporarily obstructed. The geographic information system utilizes a three-dimensional vector database, storing fixed obstacles such as buildings and mountains as polygons or cylinders, and dynamically updating temporary no-fly zones. Relative distance calculation utilizes an improved Euclidean distance algorithm, taking into account the actual dimensions of the aircraft and obstacles. For example, for a 200-meter-long bridge, the obstacle boundary is modeled as multiple key points, and the shortest distance from the aircraft to these points is calculated. The safety threshold is dynamically adjusted based on flight speed, set at 300 meters for low-speed flight and increased to 500 meters at high speeds. When the relative distance approaches the threshold, the system proactively intervenes in obstacle avoidance planning. Obstacle avoidance path generation utilizes a hybrid algorithm combining an artificial potential field method and a rapidly expanding random tree. For example, when traversing an urban building complex, the buildings generate a repulsive field, while the target point generates an attractive field. The combined forces create a safe path. For situations with multiple obstacles, such as multiple drones flying simultaneously within a building complex, the system establishes dynamic avoidance priorities and assigns avoidance tasks based on proximity. Path smoothing utilizes the Bezier curve method to optimize sharp turns in the obstacle avoidance path into smooth curves. Considering the turning radius limitations of aircraft, for example, for quadrotors, the standard turning radius is no less than 50 meters, while the minimum turning radius can be reduced to 30 meters for emergency obstacle avoidance. This smoothing not only improves flight comfort but also reduces energy consumption. The flight control system receives the optimized path data and converts it into attitude control commands. For example, when crossing a crosswind with a wind speed of 10 meters per second, the system calculates the required yaw angle to maintain a constant ground speed over the flight path. Flight trajectory adjustments are implemented through three control loops: position, velocity, and attitude, ensuring a smooth transition to the new path. During this adjustment process, the system continuously monitors the surrounding environment to maintain real-time response to new obstacles. In emergency situations, a hover-hold mechanism can be activated to buy time for rerouting. The entire process operates in a high-frequency dynamic loop, with positioning data updated at 20 Hz, obstacle detection at 10 Hz, and path planning at 5 Hz, ensuring system response times meet real-time obstacle avoidance requirements. In different scenarios, the operating parameters of each module can be dynamically adjusted based on actual conditions to achieve the optimal balance between safety and efficiency.

[0060] Example 2:

[0061] An embodiment of the present invention further provides a flight route intelligent planning system for low-altitude logistics aircraft, comprising:

[0062] The location information acquisition module is used to obtain the three-dimensional spatial map data of the aircraft's current location, including the coordinate information of buildings and utility poles and the boundaries of the no-fly zone;

[0063] The meteorological data processing module is used to obtain the wind speed, wind direction and rainfall data of the current area from the meteorological data source and calculate the meteorological impact factor based on the aircraft performance parameters;

[0064] The path generation module is used to input obstacle coordinate information, no-fly zone boundary range and meteorological influencing factors into the deep reinforcement learning path search algorithm to generate multiple candidate flight routes;

[0065] The path adjustment module is used to dynamically adjust candidate flight routes based on the aircraft's maneuverability parameters and calculate the flight time, energy cost, and safety risk score of each route;

[0066] The path evaluation module is used to evaluate the adjusted flight path using a multi-objective optimization algorithm to select the optimal flight path with the shortest flight time, lowest energy consumption and minimum safety risk;

[0067] The weather monitoring module is used to monitor weather changes during flight in real time. If weather conditions exceed preset thresholds, weather data is retrieved and weather impact factors are updated.

[0068] The path optimization module is used to locally optimize the current flight route based on the updated meteorological factors, generate new flight route segments and replace the original segments;

[0069] The obstacle avoidance module continuously monitors the relative distance between the aircraft and obstacles and no-fly zones. If a potential collision risk is detected, an emergency obstacle avoidance algorithm is triggered to generate an obstacle avoidance path.

[0070] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. A flight route intelligent planning method for low-altitude logistics aircraft, characterized in that: include: Get the three-dimensional spatial map data of the aircraft's current location; Obtain wind speed, wind direction and rainfall data for the current area from meteorological data sources and calculate meteorological impact factors; Obstacle coordinate information, no-fly zone boundaries, and meteorological factors are input into a deep reinforcement learning path search algorithm to generate multiple candidate flight routes. Dynamically adjust candidate flight routes based on aircraft maneuverability parameters, and calculate the flight time, energy cost, and safety risk score for each route; Evaluate the adjusted flight routes and select the optimal flight route with the shortest flight time, lowest energy consumption and minimum safety risk; Real-time monitoring of weather changes during flight. If weather conditions exceed preset thresholds, weather data will be re-acquired and weather impact factors will be updated. Perform local optimization and adjustment on the current flight route based on the updated meteorological impact factors, generate new flight route segments and replace the original segments; The aircraft's position is continuously monitored relative to obstacles and no-fly zones. If a potential collision risk is detected, the emergency obstacle avoidance algorithm is triggered to generate an obstacle avoidance path.

2. The method according to claim 1, characterized in that The three-dimensional spatial map data includes the coordinate information of buildings and utility poles and the boundaries of no-fly zones.

3. The flight route intelligent planning method for low-altitude logistics aircraft according to claim 2 is characterized in that: The wind speed, wind direction and rainfall data of the current area are obtained from the meteorological data source, and the meteorological impact factor is calculated in combination with the aircraft performance parameters.

4. The intelligent flight route planning method for low-altitude logistics aircraft according to claim 3 is characterized in that: A multi-objective optimization algorithm is used to evaluate the adjusted flight route and select the optimal flight route with the shortest flight time, lowest energy consumption and minimum safety risk.

5. An intelligent flight route planning system for low-altitude logistics aircraft, characterized in that: include: Position information acquisition module, used to obtain the three-dimensional space map data of the current location of the aircraft; Meteorological data processing module, used to obtain wind speed, wind direction and rainfall data of the current area from the meteorological data source and calculate meteorological impact factors; The path generation module is used to input obstacle coordinate information, no-fly zone boundary range and meteorological influencing factors into the deep reinforcement learning path search algorithm to generate multiple candidate flight routes; The path adjustment module is used to dynamically adjust candidate flight routes based on the aircraft's maneuverability parameters and calculate the flight time, energy cost, and safety risk score of each route; The path evaluation module is used to evaluate the adjusted flight route and select the optimal flight route with the shortest flight time, lowest energy consumption and minimum safety risk; The weather monitoring module is used to monitor weather changes during flight in real time. If weather conditions exceed preset thresholds, weather data is retrieved and weather impact factors are updated. The path optimization module is used to locally optimize the current flight route based on the updated meteorological factors, generate new flight route segments and replace the original segments; The obstacle avoidance processing module is used to continuously monitor the relative distance between the aircraft position and obstacles and no-fly zones. If a potential collision risk is detected, the emergency obstacle avoidance algorithm is triggered to generate an obstacle avoidance path.

6. The flight route intelligent planning system for low-altitude logistics aircraft according to claim 5 is characterized in that: The three-dimensional spatial map data includes the coordinate information of buildings and utility poles and the boundaries of no-fly zones.

7. The flight route intelligent planning system method for low-altitude logistics aircraft according to claim 6 is characterized in that: The meteorological data processing module obtains the wind speed, wind direction and rainfall data of the current area from the meteorological data source, and calculates the meteorological impact factor in combination with the aircraft performance parameters.

8. The flight route intelligent planning system for low-altitude logistics aircraft according to claim 7 is characterized in that: The path evaluation module uses a multi-objective optimization algorithm to evaluate the adjusted flight route and select the optimal flight route with the shortest flight time, lowest energy consumption and minimum safety risk.

Images