Intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion

Abstract

The invention specifically relates to the technical field of aircraft control and navigation, and discloses an intelligent unmanned aircraft trajectory control and navigation system based on multi-source data fusion, which comprises an airborne data acquisition module, a ground cloud data management module, a flight action recognition module, an energy consumption decomposition calculation module, a trajectory calculation calibration module and a trajectory correction control module, flight associated data, flight meteorological data and flight plan data are fused through an airborne data acquisition module and a ground cloud data management module, a flight associated data set is constructed, the trajectory length is calculated in a backstepping manner in an environment without GNSS signals, a high-precision trajectory measurement result of a region without GNSS signals is continuously output, an associated energy consumption coefficient is corrected, and a high-precision trajectory measurement result of a region without GNSS signals is obtained. It is ensured that the mapping relation between the level flight equivalent energy consumption and the flight distance is accurate and reliable, real-time feedback is provided for autonomous navigation and path control of the unmanned aerial vehicle, and the adaptability of the system in a complex environment is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of aircraft control and navigation technology, and more specifically, to an intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion. Background Technology

[0002] Drones are widely used in logistics and power line inspection. In logistics billing, accurate measurement of flight distance is required for reasonable billing. In power line inspection, accurate recording of inspection mileage is required for workload assessment and route planning. With the development of drone technology, the demand for high-precision trajectory measurement and drone navigation control in the absence of GNSS signals is also increasing.

[0003] Traditional UAV trajectory measurement and control mainly relies on GNSS signals to acquire positioning data to determine the flight trajectory and uses inertial measurement units to calculate displacement to assist in measurement. However, it still has some drawbacks in practical use. First, it is too dependent on GNSS signals. The core of traditional UAV trajectory measurement and control relies on positioning data provided by the Global Navigation Satellite System (GNSS) and generates the flight trajectory by continuously collecting satellite positioning coordinates. This has defects in scenarios where signals are limited or missing, and it is easily affected by complex environmental factors, resulting in a decrease in trajectory measurement capability. Second, the accumulation of measurement errors is serious. Traditional UAV trajectory measurement and control usually uses inertial measurement units to assist in trajectory measurement. Displacement is calculated by integrating three-axis acceleration and angular velocity data, which leads to the accumulation of errors over time. The measurement error of the inertial measurement unit increases continuously in the integration calculation. As the flight time increases, the deviation between the calculated displacement and the actual trajectory will become larger and larger, making it impossible to achieve long-term accurate measurement, which in turn affects the navigation and control accuracy of the aircraft. Third, ignoring environmental and maneuver factors, traditional UAV trajectory measurement and control does not consider the impact of key factors such as wind resistance and turning maneuvers on UAV energy consumption. It cannot deduce flight distance from energy consumption, resulting in poor measurement capabilities in complex environments. Furthermore, it cannot quantify the additional energy consumption of wind resistance and turning, cannot distinguish between level flight energy consumption and non-level flight energy consumption, and cannot establish a precise correlation between motor energy consumption and flight distance, leading to insufficient adaptability of trajectory measurement and aircraft navigation control. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide an intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion. By constructing a flight-related dataset through an airborne data acquisition module and a ground cloud data management module, an energy measurement and closed-loop correction mechanism is established to perform energy decomposition of maneuvers and online wind resistance compensation based on wind tunnel tests. The system controls the trajectory of the aircraft based on the error between the measured trajectory and the planned route, effectively solving the problems of excessive GNSS signal dependence, serious accumulation of measurement errors, and neglect of environmental and maneuver factors mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion, comprising an airborne data acquisition module, a ground cloud data management module, a flight action recognition module, an energy consumption decomposition and calculation module, a trajectory calculation and calibration module, and a trajectory correction and control module. Airborne data acquisition module: including motor sensing unit, inertial measurement unit and airspeed meter unit, which collects flight-related physical data in real time and transmits it to ground cloud data management module; Ground cloud data management module: includes meteorological data access unit and flight plan management unit, which acquires flight meteorological data and flight plan data in real time and builds flight-related datasets; Flight motion recognition module: Based on the flight association dataset, it identifies the flight status data of the UAV and transmits the flight status data to the energy consumption decomposition calculation module; Energy consumption decomposition calculation module: includes a component energy consumption calculation model, calculates component energy consumption and total energy consumption based on flight association dataset and flight status data, and transmits it to the trajectory calculation and calibration module; Trajectory calculation and calibration module: includes a trajectory length calculation unit and a model parameter calibration unit, calculates the trajectory length based on the energy consumption of each component, and calibrates the energy consumption calculation model parameters. The trajectory correction control module includes a trajectory deviation analysis unit and a control command generation unit. It continuously receives the trajectory length and performs real-time trajectory control.

[0006] The technical effects and advantages of this invention are as follows: 1. This invention integrates flight-related data, flight meteorological data, and flight plan data through an airborne data acquisition module and a ground cloud data management module to construct a flight-related dataset. In an environment without GNSS signals, the trajectory length is calculated by back-calculating based on a component energy consumption calculation model. The system can continuously output high-precision trajectory measurement results in areas without GNSS signals, improve the adaptability to complex application environments, and provide reliable trajectory feedback for the autonomous control of the aircraft. 2. This invention constructs an energy measurement and closed-loop correction mechanism, and obtains flight-related datasets in real time based on the energy consumption decomposition calculation module and the trajectory calculation calibration module. This reflects the actual energy consumption of the UAV, eliminates the problem of cumulative integral error, and corrects the associated energy consumption coefficient by comparing the calculated trajectory length with the actual trajectory length. This continuously optimizes the model parameters, achieves long-term accurate measurement, and supports continuous navigation and path tracking of the aircraft in the absence of GNSS. 3. This invention decomposes the energy of maneuvering actions and uses online drag compensation based on wind tunnel testing. The energy consumption decomposition calculation module breaks down the flight energy consumption into level flight, turning, climbing and hovering states. Real-time wind field data is introduced, and the drag coefficient table corresponding to the UAV's aerodynamic shape is used to calculate the energy consumption of each state. This ensures that the mapping relationship between the equivalent energy consumption of level flight and the flight distance is accurate and reliable, significantly improving the system's adaptability in complex environments and enhancing the control stability and navigation accuracy of the aircraft under wind disturbance, maneuvering and other conditions. Attached Figure Description

[0007] Figure 1 This is a schematic diagram of the overall structure of the present invention.

[0008] Figure 2 This is a schematic diagram illustrating the steps for constructing the flight-related dataset according to the present invention.

[0009] Figure 3 This is a schematic diagram of the parameter calibration steps for the sub-item energy consumption calculation model of the present invention. Detailed Implementation

[0010] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0011] As attached Figure 1 The intelligent unmanned aerial vehicle trajectory control and navigation system shown includes an airborne data acquisition module, a ground cloud data management module, a flight action recognition module, an energy consumption decomposition calculation module, a trajectory calculation and calibration module, and a trajectory correction control module.

[0012] It should be further explained that the output of the airborne data acquisition module is connected to the input of the ground cloud data management module. The output of the ground cloud data management module is connected to the input of the airborne data acquisition module, the flight action recognition module, the energy consumption decomposition calculation module, and the trajectory calculation and calibration module. The output of the flight action recognition module is connected to the input of the energy consumption decomposition calculation module. The output of the energy consumption decomposition calculation module is connected to the input of the trajectory calculation and calibration module. The output of the trajectory calculation and calibration module is connected to the input of the trajectory correction control module. The output of the trajectory correction control module is connected to the input of the ground cloud data management module.

[0013] The specific embodiments of the present invention include the following: Airborne data acquisition module: including motor sensing unit, inertial measurement unit and airspeed meter unit, which collects flight-related physical data in real time and transmits it to ground cloud data management module; Furthermore, the flight-related physical data includes basic energy consumption data, inertial correlation data, and airspeed data. The motor sensing unit is used to collect basic energy consumption data, the inertial measurement unit is used to collect inertial correlation data, and the airspeed unit is used to collect airspeed data. The basic energy consumption data includes voltage U and current I. The inertial correlation data includes yaw axis angular velocity Wz, real-time altitude H, instantaneous angular velocity W during turning, and turning angle θ. The airspeed data refers to airspeed Va.

[0014] In this embodiment, it should be specifically explained that the basic energy consumption data is collected by the motor sensing unit, which connects a current sensor in series in the motor power supply circuit and a voltage sensor in parallel across the motor power supply terminals, continuously collecting voltage U and current I according to a millisecond-level preset frequency; the inertial correlation data is collected by directly acquiring the yaw axis angular velocity Wz through the built-in gyroscope of the inertial measurement unit. When the inertial measurement unit (IMU) determines that the UAV is in a turning state, the instantaneous value of the yaw axis angular velocity during that time period is extracted in real time to obtain the instantaneous turning angular velocity W. The turning angle θ is calculated by integrating the angular velocity over time, and the real-time altitude of the UAV is measured; the airspeed data is collected by acquiring the static pressure of the air around the UAV and the total pressure generated by the airflow impact through the static pressure orifice and total pressure orifice built into the airspeed meter. The difference between the total pressure and the static pressure is calculated through the built-in sensor to obtain the dynamic pressure. The real-time airspeed Va is calculated based on the relationship that the dynamic pressure is proportional to the square of the airspeed.

[0015] Ground cloud data management module: includes meteorological data access unit and flight plan management unit, which acquires flight meteorological data and flight plan data in real time and builds flight-related datasets; Furthermore, the meteorological data access unit is used to acquire flight meteorological data, and the flight plan management unit is used to acquire flight plan data. The meteorological data unit acquires flight meteorological data based on the real-time location information of the UAV through a preset meteorological data interface, specifically including wind speed Vw and wind direction θw. The flight plan management unit receives the flight plan pre-configured by the user, acquires the flight path point sequence, generates the theoretical route based on the flight path point sequence, and obtains flight plan data, specifically including the flight path point sequence and the theoretical route.

[0016] In this embodiment, it is necessary to specifically explain the preset meteorological data interface, such as a public meteorological service platform, a regional meteorological monitoring station, or a dedicated meteorological sensing network for UAVs. The real-time location of the UAV can be obtained through short-term GNSS signals, base station positioning, or flight plan path recommendations. The meteorological data access unit performs timeliness verification on the received flight meteorological data and only retains flight meteorological data whose data generation time and current time difference are less than a preset threshold. This avoids calculation errors caused by data lag and ensures that the wind speed and wind direction data truly reflect the wind field status of the current environment of the UAV.

[0017] It should be specifically noted that each waypoint in the flight path sequence [WPT1, WPT2, ..., WPTn] includes specific geographic coordinates, planned transit time, and expected maneuvers for that waypoint. For example, a 90° turn is planned at WPT2, and a 50-meter climb is planned at WPT4. The flight path sequence data should be imported into the local database to ensure that the data can be accessed in real time. The generation of the theoretical flight path needs to be based on the waypoint sequence. The straight-line distance and azimuth between adjacent waypoints are calculated using GIS tools. The theoretical flight path includes the total theoretical course, the planned flight speed for each path segment, and the planned maneuver sequence. For example, the WPT1-WPT2 segment is planned for level flight, and the WPT2-WPT3 segment is planned for a turn followed by level flight. The planned energy consumption reference values ​​for each segment are also marked.

[0018] Furthermore, such as Figure 2 As shown, the steps for constructing the flight association dataset are as follows: S1.1: Obtain flight-related physical data transmitted by the airborne data module, and add a unified timestamp to the flight-related physical data, flight meteorological data, and flight plan data; In this embodiment, it should be specifically noted that spatial correlation needs to be based on the sampling time of the airborne data module to perform time calibration on the flight meteorological data and flight plan data, so as to ensure that all data at the same timestamp can be called at the same time, avoiding energy consumption calculation deviations caused by time differences.

[0019] S1.2: Based on the flight path point sequence and the real-time location of the UAV in the flight meteorological data, spatial correlation is performed on the flight-related data, flight meteorological data and flight plan data; In this embodiment, it should be specifically noted that the real-time position of the UAV is obtained based on short-term integration of IMU data or short-term GNSS signal assistance. For example, when the UAV flies to the area of ​​30°15'N, 120°20'E, it is ensured that the real-time wind speed and wind direction of that airspace are used to solve the data mismatch caused by the spatial heterogeneity of the wind field.

[0020] S1.3: Store the data after unifying the timestamp and spatial association as a flight association dataset, including timestamps, flight meteorological data, flight plan data, and flight association data.

[0021] In this embodiment, it should be specifically noted that the flight-related dataset is stored in a database that can be read and written in real time. Subsequently, the flight action recognition module, energy consumption decomposition calculation module, and trajectory calculation calibration module can quickly obtain the required full data by calling the dataset record with a specified timestamp, supporting real-time calculation and trajectory measurement.

[0022] Flight motion recognition module: Based on the flight association dataset, it identifies the flight status data of the UAV and transmits the flight status data to the energy consumption decomposition calculation module; Furthermore, the flight status data includes turning status data, climb status data, hovering status data, and level flight status data. The turning status data includes the turn start / end time, instantaneous angular velocity sequence, and cumulative turning angle. The climb status data includes the climb start / end time and estimated climb altitude. The hovering status data includes the hovering start / end time and hovering duration. The level flight status data includes the level flight start / end time, average airspeed, and level flight distance.

[0023] In this embodiment, it should be specifically explained that the turning state recognition needs to be based on the yaw axis angular velocity Wz and turning angle θ in the flight association dataset and the expected action judgment in the flight plan data: when Wz is higher than the preset threshold of 5° / s for three consecutive sampling periods and the turning angle continues to increase, and the expected flight action is a turn, it is directly judged as a turning state; when Wz exceeds the threshold and the turning angle continues to increase, but deviates from the flight plan path, it is judged as an abnormal turning state.

[0024] Climb status identification requires comprehensive judgment based on real-time altitude and airspeed in the flight association dataset: when the airspeed is maintained within the preset level flight speed range, such as 8-12 m / s, and the current expected flight action is climb, it is judged as a climb status; when the airspeed fluctuation is small, but deviates from the planned climb path segment, it is necessary to add auxiliary verification by combining the abnormal total motor energy consumption E. Motor energy consumption E=U×I, which is calculated by real-time voltage and current. After confirmation, it is judged as an unplanned climb status.

[0025] Hovering state recognition requires a comprehensive judgment based on the yaw axis angular velocity, turning angle, and airspeed in the flight association dataset: when Wz is below the threshold, the turning angle remains stable, the airspeed is below the threshold (e.g., less than 1 m / s), there is no level flight, and this continues for more than three sampling cycles, it is judged as hovering state; when hovering is in the expected flight maneuver, the airspeed threshold can be relaxed to avoid errors caused by slight airflow.

[0026] The level flight status identification uses an elimination method, combining flight plan data and airspeed data in the flight association dataset for determination: when Wz is below the threshold, the turning angle does not change significantly, the airspeed is stable within the preset level flight speed range, and the current expected flight action is level flight, it is determined to be in level flight status. Level flight status is the default basic status, ensuring that the flight status covers the entire flight cycle.

[0027] Energy consumption decomposition calculation module: includes a component energy consumption calculation model, calculates component energy consumption and total energy consumption based on flight association dataset and flight status data, and transmits it to the trajectory calculation and calibration module; Furthermore, the energy consumption components include drag energy consumption Ew, turning energy consumption Et, climb energy consumption Ep, hovering energy consumption Eh, and level flight energy consumption Ec. Calculating these components requires importing flight-related datasets and flight status data into the component energy consumption calculation model, obtaining the real-time airspeed Va from the flight-related dataset, and substituting it into the formula. The wind resistance energy consumption Ew is calculated, where Kw is the wind resistance energy consumption coefficient and t is the time variable; the turning angles and instantaneous angular velocities of n turns in the flight correlation dataset are obtained and substituted into the formula. The turning energy consumption Et, θ is calculated. i Let W be the turning angle of the i-th turn. i Let Kc be the instantaneous turning angle of the i-th turn, and Kc be the turning energy consumption coefficient; obtain the climb altitude of m climbs in the flight association dataset, and substitute them into the formula. The climbing energy consumption Ep and ΔH were calculated. j Let Kd be the climb altitude for the j-th climb, and Kd be the climb energy consumption coefficient; obtain the hovering duration T and the drone's hovering power P from the flight association dataset, and substitute them into the formula. The hovering energy consumption Eh was calculated; the level flight distance L from the flight association dataset was obtained and substituted into the formula. The power consumption Ec during level flight was calculated. The calculation of total energy consumption requires obtaining the real-time voltage U(t) and real-time current I(t) from the flight-related dataset and substituting them into the formula. The total energy consumption E is calculated.

[0028] In this embodiment, it should be specifically noted that wind resistance energy consumption refers to the extra energy consumed by the UAV to overcome air resistance when flying in windy conditions, measured in J. The wind resistance energy consumption coefficient Kw is related to the aerodynamic shape and air density of the UAV, and is pre-calibrated and stored in the system through wind tunnel testing, measured in J·s / m. 2 The time variable t refers to the calculation time interval for wind resistance energy consumption; Turning energy consumption refers to the sum of energy consumption of all turning actions during the entire flight of a UAV, with the unit being J. The turning energy consumption coefficient Kc reflects the energy consumption characteristics of the UAV's turning actions and is determined by the UAV model, aerodynamic shape, and motor performance. It is calibrated through wind tunnel tests or measured data, with the unit being joules / (degree·degree / second). Climb energy consumption refers to the sum of energy consumption of all climb actions during the entire flight process, with the unit being J. The climb height of a single climb action refers to the vertical change in altitude of the UAV during that climb segment, calculated from the real-time altitude at the start and end of the climb action. The climb energy consumption coefficient Kd reflects the energy consumption per unit climb height of the UAV, determined by the UAV's weight and motor lift efficiency, calibrated through measured data, with the unit being J / m.

[0029] Hovering energy consumption refers to the energy consumed by a drone when it is hovering in a fixed position, measured in J. Hovering power P refers to the continuous power required for the drone to maintain a hovering state, which is determined by the drone's weight and propeller efficiency. For example, the hovering power of a small quadcopter drone is 50-150W, calibrated through actual measurement data. Hovering duration T refers to the total time the drone is hovering during the entire flight, calculated based on the hovering start / end time.

[0030] Level flight energy consumption refers to the energy consumed by a drone in a windless, uniform straight flight state, solely for maintaining flight level. The unit is J. Level flight distance L refers to the actual horizontal distance the drone flies in level flight state, calculated based on the level flight start / end time and level flight distance in the level flight state data. Level flight energy consumption coefficient K reflects the level flight energy consumption of the drone per unit flight distance. It is calibrated by wind tunnel tests or historical flight data, and the unit is J / m. For example, if a quadcopter drone consumes 1000J of power to fly 100 meters in windless conditions, then K=10J / m.

[0031] Trajectory calculation and calibration module: includes a trajectory length calculation unit and a model parameter calibration unit, calculates the trajectory length based on the energy consumption of each component, and calibrates the energy consumption calculation model parameters. Furthermore, the trajectory length calculation unit is used to calculate the trajectory length. The calculation of the trajectory length requires obtaining the energy consumption of each component and the total energy consumption. Based on wind resistance energy consumption Ew, turning energy consumption Et, climbing energy consumption Ep, hovering energy consumption Eh, and total energy consumption E, the calculation is performed using the formula... The equivalent energy consumption for level flight is calculated, and the ratio of the equivalent energy consumption for level flight to the energy consumption coefficient for level flight is used to obtain the trajectory length Lc=Ece / K, where K is the energy consumption coefficient for level flight.

[0032] In this embodiment, it should be specifically explained that the trajectory calculation calibration module obtains the theoretical flight path and the actual trajectory length of the area with good GNSS signal from the flight association dataset in the ground cloud data management module. The theoretical flight path is used to initially verify the rationality of the calculated trajectory, and the actual trajectory length is used to calibrate the parameters of the sub-item energy consumption calculation model.

[0033] Furthermore, such as Figure 3 As shown, the calibration steps for the parameters of the sub-item energy consumption calculation model are as follows: S2.1: When the UAV flies to an area with good GNSS signal, the ground cloud data management module acquires GNSS positioning data, calculates the actual trajectory length of the area with good GNSS signal through continuous positioning coordinates, and triggers parameter calibration; S2.2: Calculate the virtual and real trajectory length error based on the real trajectory length in the area with good GNSS signal and the trajectory length calculation unit, and identify the associated energy consumption coefficient in the sub-item energy consumption calculation model based on the virtual and real trajectory length error; S2.3: The least squares method is used to establish a correlation model between the correlation energy consumption coefficient and the error of the virtual and real trajectory length. The corrected correlation energy consumption coefficient is calculated and multi-parameter collaborative calibration is performed. The corrected correlation energy consumption coefficient is synchronized to the energy consumption decomposition calculation module in real time.

[0034] In this embodiment, it should be specifically explained that the associated energy consumption coefficients include wind resistance energy consumption coefficient, turning energy consumption coefficient, climb energy consumption coefficient, hovering energy consumption coefficient, and level flight energy consumption coefficient. The identification of associated energy consumption coefficients in the sub-item energy consumption calculation model needs to be based on the actual flight state of the UAV. For example, if the trajectory length obtained by the trajectory length calculation unit is too large, it means that the K value is too small, and the level flight energy consumption coefficient should be calibrated first. Multi-parameter collaborative calibration means that if the error is caused by multiple factors, multiple associated energy consumption coefficients need to be calibrated simultaneously. For example, by comparing the turning angle, angular velocity, and turning energy consumption in the real trajectory, the Kc value can be corrected to ensure that the subsequent energy consumption calculation is more accurate.

[0035] The trajectory correction control module includes a trajectory deviation analysis unit and a control command generation unit. It continuously receives the trajectory length and performs real-time trajectory control.

[0036] Furthermore, the trajectory correction control module includes a trajectory deviation analysis unit and a control command generation unit, which are used to perform real-time analysis and comparison between the calculated trajectory length and the theoretical flight path in the flight association dataset, and generate flight control commands based on the deviation, which are then sent to the UAV through the ground cloud data management module to achieve closed-loop trajectory correction and control.

[0037] In this embodiment, it should be specifically explained that continuous trajectory control refers to the system not only continuously calculating the trajectory length in areas without GNSS signals, but also applying it to flight control. The trajectory deviation analysis unit compares the calculated trajectory with the theoretical route in the flight association dataset in real time to analyze lateral deviation, longitudinal progress deviation, and altitude deviation.

[0038] It should be further explained that the control command generation unit makes intelligent decisions based on the deviation analysis results, for example: When a lateral deviation exceeding the threshold is detected, a command to "adjust the heading XX degrees to the left / right" is generated. When a flight progress delay is detected due to headwind, an instruction to "increase airspeed by XX%" is generated, which should meet the condition that the airspeed is within the safety margin. When the system detects strong winds ahead and abnormally high energy consumption, it generates an instruction to "suggest climbing / descending to XX altitude to avoid wind disturbance".

[0039] Control commands are transmitted in real time to the UAV's flight control system via the data link of the ground cloud data management module. The UAV's flight control system uses the commands as advanced guidance commands to actively correct the flight trajectory. By identifying flight status and energy consumption decomposition in real time, combined with wind resistance compensation and model calibration, this system not only achieves high-precision trajectory measurement in the absence of GNSS signals, but also provides real-time feedback for the UAV's autonomous navigation and path control, improving the control accuracy and navigation reliability of the aircraft in complex environments.

[0040] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion, characterized in that, It includes an airborne data acquisition module, a ground cloud data management module, a flight action recognition module, an energy consumption decomposition and calculation module, a trajectory calculation and calibration module, and a trajectory correction and control module. The airborne data acquisition module includes a motor sensing unit, an inertial measurement unit, and an airspeed meter unit, which collects flight-related physical data in real time and transmits it to the ground cloud data management module. The ground cloud data management module includes a meteorological data access unit and a flight plan management unit, which acquires flight meteorological data and flight plan data in real time and constructs a flight-related dataset; The flight action recognition module identifies the UAV's flight status data based on the flight association dataset and transmits the flight status data to the energy consumption decomposition calculation module. The energy consumption decomposition calculation module includes a component energy consumption calculation model, which calculates component energy consumption and total energy consumption based on flight-related datasets and flight status data, and transmits them to the trajectory calculation and calibration module. The trajectory calculation and calibration module includes a trajectory length calculation unit and a model parameter calibration unit, which calculates the trajectory length based on the energy consumption of each component and calibrates the energy consumption calculation model parameters. The trajectory correction control module includes a trajectory deviation analysis unit and a control command generation unit, continuously receives the trajectory length, and performs real-time trajectory control.

2. The intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion according to claim 1, characterized in that: The flight-related physical data includes basic energy consumption data, inertial correlation data, and airspeed data. The motor sensing unit is used to collect basic energy consumption data, the inertial measurement unit is used to collect inertial correlation data, and the airspeed meter unit is used to collect airspeed data. The basic energy consumption data includes voltage U and current I. The inertial correlation data includes yaw axis angular velocity Wz, real-time altitude H, instantaneous angular velocity W during turning, and turning angle θ. The airspeed data refers to airspeed Va.

3. The intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion according to claim 1, characterized in that: The meteorological data access unit is used to acquire flight meteorological data, and the flight plan management unit is used to acquire flight plan data. The meteorological data unit acquires flight meteorological data based on the real-time location information of the UAV through a preset meteorological data interface, specifically including wind speed Vw and wind direction θw. The flight plan management unit receives the flight plan pre-configured by the user, acquires the flight path point sequence, generates a theoretical route based on the flight path point sequence, and obtains flight plan data, specifically including the flight path point sequence and the theoretical route.

4. The intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion according to claim 1, characterized in that: The steps for constructing the flight association dataset are as follows: S1.1: Obtain flight-related physical data transmitted by the airborne data module, and add a unified timestamp to the flight-related physical data, flight meteorological data, and flight plan data; S1.2: Based on the flight path point sequence and the real-time location of the UAV in the flight meteorological data, spatial correlation is performed on the flight-related data, flight meteorological data and flight plan data; S1.3: Store the data after unifying the timestamp and spatial association as a flight association dataset, including timestamps, flight meteorological data, flight plan data, and flight association data.

5. The intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion according to claim 1, characterized in that: The flight status data includes turning status data, climb status data, hovering status data, and level flight status data. The turning status data includes the turn start / end time, instantaneous angular velocity sequence, and cumulative turning angle. The climb status data includes the climb start / end time and estimated climb altitude. The hovering status data includes the hovering start / end time and hovering duration. The level flight status data includes the level flight start / end time, average airspeed, and level flight distance.

6. The intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion according to claim 1, characterized in that: The energy consumption components include drag energy consumption Ew, turning energy consumption Et, climb energy consumption Ep, hovering energy consumption Eh, and level flight energy consumption Ec. Calculating these components requires importing flight-related datasets and flight status data into the component energy consumption calculation model, obtaining the real-time airspeed Va from the flight-related dataset, and substituting it into the formula. The wind resistance energy consumption Ew is calculated, where Kw is the wind resistance energy consumption coefficient and t is the time variable; the turning angles and instantaneous angular velocities of n turns in the flight correlation dataset are obtained and substituted into the formula. The turning energy consumption Et, θ is calculated. i Let W be the turning angle of the i-th turn. i Let Kc be the instantaneous turning angle of the i-th turn, and Kc be the turning energy consumption coefficient; obtain the climb altitude of m climbs in the flight association dataset, and substitute them into the formula. The climbing energy consumption Ep and ΔH were calculated. j Let Kd be the climb altitude for the j-th climb, and Kd be the climb energy consumption coefficient; obtain the hovering duration T and the drone's hovering power P from the flight association dataset, and substitute them into the formula. The hovering energy consumption Eh was calculated. Obtain the level flight distance L from the flight association dataset and substitute it into the formula. The power consumption Ec during level flight was calculated. The calculation of total energy consumption requires obtaining the real-time voltage U(t) and real-time current I(t) from the flight-related dataset and substituting them into the formula. The total energy consumption E is calculated.

7. The intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion according to claim 1, characterized in that: The trajectory length calculation unit is used to calculate the trajectory length. The calculation of the trajectory length requires obtaining the energy consumption of each component and the total energy consumption. Based on wind resistance energy consumption Ew, turning energy consumption Et, climbing energy consumption Ep, hovering energy consumption Eh, and total energy consumption E, the calculation is performed using the formula... The equivalent energy consumption for level flight is calculated, and the ratio of the equivalent energy consumption for level flight to the energy consumption coefficient for level flight is used to obtain the trajectory length Lc=Ece / K.

8. The intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion according to claim 1, characterized in that: The calibration steps for the parameters of the sub-item energy consumption calculation model are as follows: S2.1: When the UAV flies to an area with good GNSS signal, the ground cloud data management module acquires GNSS positioning data, calculates the actual trajectory length of the area with good GNSS signal through continuous positioning coordinates, and triggers parameter calibration; S2.2: Calculate the virtual and real trajectory length error based on the real trajectory length in the area with good GNSS signal and the trajectory length calculation unit, and identify the associated energy consumption coefficient in the sub-item energy consumption calculation model based on the virtual and real trajectory length error; S2.3: The least squares method is used to establish a correlation model between the correlation energy consumption coefficient and the error of the virtual and real trajectory length. The corrected correlation energy consumption coefficient is calculated and multi-parameter collaborative calibration is performed. The corrected correlation energy consumption coefficient is synchronized to the energy consumption decomposition calculation module in real time.

9. The intelligent unmanned aerial vehicle trajectory control and navigation system based on multi-source data fusion according to claim 1, characterized in that: The trajectory correction control module includes a trajectory deviation analysis unit and a control command generation unit. It is used to perform real-time analysis and comparison between the calculated trajectory length and the theoretical flight path in the flight association dataset, and generate flight control commands based on the deviation. These commands are then sent to the UAV through the ground cloud data management module to achieve closed-loop trajectory correction and control.

Images