Multi-source data fusion and dynamic adjustment algorithm and system for complex flight environment

Abstract

The invention belongs to the technical field of aviation flight, and particularly relates to a multi-source data fusion and dynamic adjustment algorithm and system suitable for a complex flight environment. Fusing multi-source data; path planning and dynamic adjustment; self-adaptive control and execution are carried out; and performing exception processing and redundancy compensation. According to the invention, on the basis of a multi-source data fusion and dynamic adjustment algorithm, various sensor data are fused and a flight strategy is dynamically adjusted in real time, so that the navigation capability and safety of an aircraft in a complex environment are remarkably improved. Compared with a traditional navigation system that a single sensor depends on. Through multi-source data fusion and a dynamic adjustment algorithm, the system significantly improves the autonomous navigation and control capability of an aircraft in a complex flight environment, and effectively solves the problem of uncertainty caused by data loss and environment change in the prior art.

Description

Technical Field

[0001] The present invention belongs to but is not limited to the field of aviation flight technology, and in particular relates to a multi-source data fusion and dynamic adjustment algorithm and system suitable for complex flight environments, especially for dynamic optimization processing of pilot operation data and flight status data. The system ensures high-precision, stable flight control and safe operation in complex flight environments by fusing multi-source data (such as pilot operation input, aircraft sensor data, environmental data, etc.) and adjusting flight paths and operating parameters in real time. Background Art

[0002] In modern aircraft systems, especially in complex flight missions, such as long-distance transportation, military training or emergency rescue missions, pilots need to face a variety of flight states and operation scenarios, and the state information during the flight changes rapidly and complexly. These mission scenarios often involve high altitude, high speed, long-term flight, and are accompanied by complex meteorological conditions, such as strong winds, turbulence, extreme temperature changes, etc. In addition, during the flight, the pilot's physiological state and operational behavior are crucial to the success and safety of the flight mission. However, traditional flight data monitoring systems mainly rely on single sensor data, such as flight control systems, attitude sensors or atmospheric condition monitors, which cannot fully reflect the pilot's operational dynamics and overall flight status. Existing technologies are generally unable to process and optimize multi-source data in real time, and it is difficult to cope with the complex interactions between pilot operation data and flight status data. With the continuous complexity of aircraft systems, the technical problem that the existing technology urgently needs to solve is how to dynamically optimize the processing of pilot operation data and flight status data during flight, and adjust flight strategies and operations in real time to improve the safety, stability and efficiency of flight missions.

[0003] Therefore, how to effectively integrate multi-source data while ensuring computing resource constraints, and dynamically adjust the pilot's operating behavior and aircraft status to ensure the smooth completion of the flight mission has become a key issue that urgently needs to be solved in the current technical field. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention provides a multi-source data fusion and dynamic adjustment algorithm and system suitable for complex flight environments.

[0005] The present invention is implemented as follows: a multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments, the algorithm comprising:

[0006] S1: Data collection and preprocessing: The system first collects multi-source data in the aircraft flight environment in real time through the sensor module; for each type of sensor data, the system will perform basic preprocessing, such as filtering, denoising and other operations to ensure the accuracy and consistency of the data;

[0007] S2: Multi-source data fusion: The pre-processed data enters the multi-source data fusion module; first, the extended Kalman filter (EKF) performs preliminary fusion and state estimation on the sensor data; the system predicts the current position, attitude and speed of the aircraft by establishing a state transition model, combining the current observation data with the historical state; then, the system dynamically adjusts the weight of each sensor in the fusion process according to the confidence of different sensors; for example, when the GPS signal is interfered, the system will increase the weight of IMU and LiDAR to ensure the accuracy of the fusion result;

[0008] S3: Path planning and dynamic adjustment: The system performs local path planning based on the fused data. During this process, the system will detect obstacles in the flight path in real time, calculate the safe path, and dynamically adjust the flight trajectory of the aircraft; at the same time, the system will optimize the flight path according to the global mission requirements to ensure that the aircraft completes the flight mission with the optimal path; in complex environments, such as buildings, forests, etc., the system will give priority to avoiding obstacles to ensure flight safety;

[0009] S4: Adaptive control and execution: After the path planning is completed, the system executes flight control through the adaptive control module; the PID controller adjusts the aircraft's attitude and speed in real time according to the sensor feedback to ensure that the aircraft flies smoothly according to the planned path; at the same time, the system will dynamically adjust the control strategy according to environmental changes; for example, when the wind speed changes or the number of obstacles increases, the system will reduce the flight speed and enhance the obstacle avoidance capability;

[0010] S5: Exception handling and redundancy compensation: The system detects the integrity of sensor data in real time. When some sensor data is lost or abnormal, the system uses the data of other sensors or historical data to compensate through the data redundancy compensation mechanism. For example, when LiDAR data is unavailable due to weather reasons, the system will combine IMU and visual sensors for environmental perception.

[0011] Another object of the present invention is to provide a multi-source data fusion and dynamic adjustment system suitable for complex flight environments based on the multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments, the system specifically comprising:

[0012] The sensor module provides global position information, but the data will be incomplete or distorted when the signal is blocked or interfered. The accelerometer and gyroscope provide the aircraft's attitude and motion information, but the IMU will have drift errors during long-term flight. The laser scans the surrounding environment to generate a three-dimensional point cloud, providing high-precision environmental perception, which is suitable for obstacle detection. The camera obtains environmental images for identifying specific landmarks or environmental features.

[0013] The multi-source data fusion module is connected to the sensor module to combine the sensor data of GPS, IMU and LiDAR to generate a unified flight state estimate; EKF can use historical data and models to perform state prediction and error correction when the sensor data is imperfect; the weight of the sensor is dynamically adjusted according to the confidence of different sensors; for example, when the GPS signal is weak, the system will reduce its reliance on GPS data and increase the weight of IMU and LiDAR data;

[0014] The path planning and dynamic adjustment module is connected to the multi-source data fusion module and combines sensor data to generate a safe flight path based on the current environment in real time. The module detects and avoids obstacles in the flight path based on sensor feedback. Based on the global flight mission requirements, the module dynamically optimizes the flight path in combination with multi-source data to ensure that the aircraft can complete the flight mission efficiently and safely in a complex environment.

[0015] The adaptive control module is connected to the multi-source data fusion module to adjust the aircraft's attitude, speed and direction in real time based on sensor feedback to ensure that the aircraft flies smoothly along the planned path. The system will dynamically adjust the flight strategy according to changes in the flight environment, such as wind speed and obstacle distribution. For example, in areas with dense obstacles, the system will reduce the flight speed to increase obstacle avoidance time.

[0016] The data anomaly processing module is connected to the sensor module to detect the integrity and validity of sensor data in real time. When some sensor data is lost or abnormal, the system will issue a warning and adjust the data fusion strategy. When some sensor data is lost, the system can perform redundant compensation through data from other sensors and historical flight information. For example, in the event of GPS signal loss, IMU and visual sensors can be used for position estimation.

[0017] Furthermore, the sensor module is implemented as follows:

[0018] GPS module: The system integrates a GPS receiver to obtain global location information through satellite signals; the data preprocessing module filters the received GPS signals, removes noise and jump data, and calculates the geographic coordinates of the current location; because the GPS signal is blocked or interfered, it needs to be fused with data from other sensors to supplement the positioning accuracy;

[0019] IMU module: The IMU module includes an accelerometer and a gyroscope, which are used to measure the linear acceleration and angular velocity of the aircraft. Through high-sampling rate data collection, it provides real-time information on the aircraft's attitude and motion. The system uses IMU data to track the aircraft's instantaneous state and calculates displacement information through integration. However, there is a drift problem with IMU data, and the system uses EKF to correct its errors.

[0020] LiDAR module: LiDAR generates 3D point cloud data of the surrounding environment by emitting lasers and measuring their return time. This module is suitable for high-precision environmental perception, especially obstacle detection. During the flight of the aircraft, LiDAR will continuously scan the environment and provide the precise distance and direction of obstacles. The system uses this data for local path planning and dynamic obstacle avoidance.

[0021] Visual sensor module: includes one or more cameras for capturing images of the environment. The system uses visual data to identify specific landmarks or environmental features to assist the aircraft in positioning and navigation. Visual sensors play a key role, especially when the GPS signal is weak or lost. The visual inertial navigation system VINS can combine IMU data for environmental perception and positioning.

[0022] Furthermore, the implementation of the multi-source data fusion module is as follows:

[0023] Extended Kalman filter EKF: It is the core data fusion algorithm of the system. Its implementation steps include:

[0024] (1) Prediction step: Based on the real-time motion data provided by the IMU and LiDAR, the system predicts the current position and attitude of the aircraft through the state transition equation. This step uses the historical state data and the real-time motion information of the IMU to estimate the current state;

[0025] (2) Update step: When new observations are provided by GPS, LiDAR, or vision sensors, the EKF uses these observations to correct the predicted state and calculate the error covariance matrix of the state estimate; the updated state includes the precise position, attitude, and velocity of the aircraft;

[0026] Data weighting strategy: The data weighting strategy in EKF adjusts the data weight of each sensor according to the confidence of different sensors; the system monitors the strength and quality of the GPS signal, reduces the weight of GPS data when the signal is weak or lost, and increases the reliance on IMU, LiDAR and vision data; this dynamic weighting method ensures the robustness and accuracy of data fusion.

[0027] Furthermore, the implementation of the path planning and dynamic adjustment module is as follows:

[0028] Local path planning: The local path planning module detects obstacles in the flight path in real time and calculates a safe path based on the three-dimensional point cloud data generated by LiDAR. The implementation steps of this module are as follows:

[0029] (1) Obstacle detection: By processing LiDAR point cloud data, the system identifies obstacles in the current environment and marks their location and size.

[0030] (2) Path calculation: Based on the aircraft's current flight target, the system uses the A* algorithm or the RRT fast random tree search algorithm to calculate the optimal path and avoid detected obstacles.

[0031] (3) Path adjustment: If the aircraft encounters a new obstacle during flight, the system will dynamically adjust the flight path to avoid the obstacle in real time to ensure the safe movement of the aircraft.

[0032] Global Path Optimization: The global path planning module optimizes the flight path based on mission requirements to ensure that the aircraft completes the mission with the optimal path; global path optimization takes into account the time and energy constraints of the flight mission, combines multi-source data to calculate the optimal flight route before the flight mission begins, and dynamically adjusts it according to the actual environment during the flight.

[0033] Further, the implementation of the adaptive control module is as follows:

[0034] PID controller: The PID controller adjusts the aircraft's attitude, speed, and direction to ensure that it flies smoothly along the planned path. The implementation steps are as follows:

[0035] (1) Attitude control: The system adjusts the aircraft's pitch, roll, and yaw angles in real time based on feedback from the IMU and visual sensors.

[0036] (2) Speed ​​control: The PID controller adjusts the aircraft's forward speed based on obstacles and target locations in the flight path to ensure that it can fly quickly and safely.

[0037] (3) Position control: The controller adjusts the aircraft's trajectory based on GPS and fused position information to keep it on the planned path;

[0038] Adaptive strategy adjustment: During flight, the system will adjust flight parameters according to changes in the external environment, such as wind speed changes, obstacle density, etc.; for example, when dense obstacles are detected, the system will reduce the flight speed to increase obstacle avoidance time, or increase the LiDAR scanning frequency to improve environmental perception accuracy.

[0039] Furthermore, the implementation of the data exception processing module is as follows:

[0040] Data integrity detection: The system monitors the integrity and validity of sensor data in real time. This module detects whether the sensor has data loss, outliers or noise interference, and issues a warning or adjusts the data fusion algorithm. For example, when the GPS signal fails, the system will issue an alarm and automatically switch to the IMU and visual sensor for positioning.

[0041] Data redundancy compensation: When some sensor data is lost, the system uses data and historical information from other sensors for redundancy compensation; for example, in the event of a GPS signal loss, the system calculates the aircraft’s current position through displacement integration from the IMU, while using visual sensors to identify landmarks or environmental features to further correct the position estimate.

[0042] Another object of the present invention is to provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments.

[0043] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments.

[0044] Another object of the present invention is to provide an information data processing terminal, which is used to implement the multi-source data fusion and dynamic adjustment system suitable for complex flight environments.

[0045] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0046] 1. High robustness of multi-source data fusion: The system uses an extended Kalman filter (EKF) and a dynamic data weighting strategy, which can still provide accurate state estimates when sensor data is incomplete or noisy. This multi-source data fusion greatly improves the robustness of the system in complex environments and ensures the stability and navigation accuracy of the aircraft in different flight environments. Through EKF and dynamic weighting strategies, the system effectively integrates multiple sensor data, and can still provide high-precision state estimates even if some sensor data is lost or abnormal. Compared with navigation systems that rely on a single sensor, multi-source data fusion makes the system more reliable in complex environments.

[0047] 2. Dynamic path planning and real-time adjustment: The system can dynamically adjust the flight path according to real-time changes in the flight environment. Whether it is obstacle detection or environmental changes, the system can respond in real time and automatically adjust the flight trajectory. This allows the aircraft to respond flexibly in complex environments without being restricted by static planning methods.

[0048] 3. Adaptive control enhances flight safety: Through adaptive control strategies, the system can adjust flight parameters in real time according to changes in the flight environment. For example, when encountering sudden wind speeds or obstacles, the system can automatically reduce the flight speed and enhance obstacle avoidance capabilities. This significantly improves the safety of the aircraft in complex flight environments.

[0049] 4. Data anomaly processing and redundancy compensation mechanism: The system has a strong data anomaly processing capability. When some sensor data is lost or abnormal, the system can compensate through other sensors or historical data to ensure the continuity and reliability of navigation. This redundancy compensation mechanism enables the system to maintain good performance even when the data is incomplete.

[0050] 5. Flexibility of dynamic path planning: Adaptive control improves flight safety: The adaptive control module uses PID controller and adaptive strategy adjustment to ensure that the aircraft can fly stably under complex environmental conditions, and adjust flight parameters in time according to changes in the external environment to reduce flight risks.

[0051] This algorithm based on multi-source data fusion and dynamic adjustment significantly improves the aircraft's navigation capability and safety in complex environments by fusing multiple sensor data and dynamically adjusting flight strategies in real time. Compared with traditional navigation systems that rely on a single sensor, this system significantly improves the aircraft's autonomous navigation and control capabilities in complex flight environments through multi-source data fusion and dynamic adjustment algorithms, effectively solving the uncertainty problems caused by data loss and environmental changes in existing technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a flow chart of a multi-source data fusion and dynamic adjustment algorithm applicable to a complex flight environment provided by an embodiment of the present invention;

[0053] Figure 2 It is a structural diagram of a multi-source data fusion and dynamic adjustment system applicable to a complex flight environment provided by an embodiment of the present invention;

[0054] Figure 3 is a structural diagram of a sensor module provided by an embodiment of the present invention;

[0055] In the figure: 1. Sensor module; 2. Multi-source data fusion module; 3. Path planning and dynamic adjustment module; 4. Adaptive control module; 5. Data anomaly processing module; 6. GPS module; 7. IMU module; 8. LiDAR module; 9. Visual sensor module; 10. GPS receiver; 11. Accelerometer; 12. Gyroscope. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0057] like Figure 1 As shown, an embodiment of the present invention provides a multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments, the algorithm comprising:

[0058] S1: Data collection and preprocessing: The system first collects multi-source data in the aircraft flight environment in real time through the sensor module; for each type of sensor data, the system will perform basic preprocessing, such as filtering, denoising and other operations to ensure the accuracy and consistency of the data;

[0059] S2: Multi-source data fusion: The pre-processed data enters the multi-source data fusion module; first, the extended Kalman filter (EKF) performs preliminary fusion and state estimation on the sensor data; the system predicts the current position, attitude and speed of the aircraft by establishing a state transition model, combining the current observation data with the historical state; then, the system dynamically adjusts the weight of each sensor in the fusion process according to the confidence of different sensors; for example, when the GPS signal is interfered, the system will increase the weight of IMU and LiDAR to ensure the accuracy of the fusion result;

[0060] S3: Path planning and dynamic adjustment: The system performs local path planning based on the fused data. During this process, the system will detect obstacles in the flight path in real time, calculate the safe path, and dynamically adjust the flight trajectory of the aircraft; at the same time, the system will optimize the flight path according to the global mission requirements to ensure that the aircraft completes the flight mission with the optimal path; in complex environments, such as buildings, forests, etc., the system will give priority to avoiding obstacles to ensure flight safety;

[0061] S4: Adaptive control and execution: After the path planning is completed, the system executes flight control through the adaptive control module; the PID controller adjusts the aircraft's attitude and speed in real time according to the sensor feedback to ensure that the aircraft flies smoothly according to the planned path; at the same time, the system will dynamically adjust the control strategy according to environmental changes; for example, when the wind speed changes or the number of obstacles increases, the system will reduce the flight speed and enhance the obstacle avoidance capability;

[0062] S5: Exception handling and redundancy compensation: The system detects the integrity of sensor data in real time. When some sensor data is lost or abnormal, the system uses the data of other sensors or historical data to compensate through the data redundancy compensation mechanism. For example, when LiDAR data is unavailable due to weather reasons, the system will combine IMU and visual sensors for environmental perception.

[0063] like Figure 2 As shown, an embodiment of the present invention provides a multi-source data fusion and dynamic adjustment system suitable for a complex flight environment based on the multi-source data fusion and dynamic adjustment algorithm suitable for a complex flight environment, and the system specifically includes:

[0064] Sensor module 1 provides global position information, but the data will be incomplete or distorted when the signal is blocked or interfered. It provides the attitude and motion information of the aircraft through the accelerometer and gyroscope, but the IMU will have drift errors during long-term flight. It generates a three-dimensional point cloud by laser scanning the surrounding environment, providing high-precision environmental perception, which is suitable for obstacle detection. It obtains environmental images through cameras to identify specific landmarks or environmental features.

[0065] The multi-source data fusion module 2 is connected to the sensor module 1 and is used to combine the sensor data of GPS, IMU and LiDAR to generate a unified flight state estimation. The EKF can perform state prediction and error correction through historical data and models when the sensor data is imperfect. The weight of the sensor is dynamically adjusted according to the confidence of different sensors. For example, when the GPS signal is weak, the system will reduce the dependence on GPS data and increase the weight of IMU and LiDAR data.

[0066] The path planning and dynamic adjustment module 3 is connected to the multi-source data fusion module 2, and combines the sensor data to generate a safe flight path based on the current environment in real time. The module detects and avoids obstacles in the flight path based on sensor feedback. Based on the global flight mission requirements, the module dynamically optimizes the flight path in combination with multi-source data to ensure that the aircraft completes the flight mission efficiently and safely in a complex environment.

[0067] The adaptive control module 4 is connected to the multi-source data fusion module 2, and adjusts the attitude, speed and direction of the aircraft in real time based on sensor feedback to ensure that the aircraft flies smoothly according to the planned path; the system will dynamically adjust the flight strategy according to changes in the flight environment, such as wind speed, obstacle distribution, etc. For example, in an area with dense obstacles, the system will reduce the flight speed to increase the obstacle avoidance time;

[0068] The data anomaly processing module 5 is connected to the sensor module 1 to detect the integrity and validity of the sensor data in real time. When some sensor data is lost or abnormal, the system will issue a warning and adjust the data fusion strategy. When some sensor data is lost, the system can perform redundant compensation through data from other sensors and historical flight information. For example, in the case of GPS signal loss, IMU and visual sensors can be used for position estimation.

[0069] like Figure 3 As shown, the implementation of the sensor module is as follows:

[0070] GPS module 6: The system integrates a GPS receiver 10 to obtain global location information through satellite signals; the data preprocessing module filters the received GPS signals, removes noise and jump data, and calculates the geographic coordinates of the current location; because the GPS signal is blocked or interfered, it needs to be fused with data from other sensors to supplement the positioning accuracy;

[0071] IMU module 7: IMU module 7 includes an accelerometer 11 and a gyroscope 12, which are used to measure the linear acceleration and angular velocity of the aircraft; through high sampling rate data collection, the aircraft's attitude and motion information are provided in real time; the system uses IMU data to track the instantaneous state of the aircraft and calculates displacement information by integration; however, there is a drift problem in the IMU data, and the system corrects its error through EKF;

[0072] LiDAR module 8: LiDAR generates 3D point cloud data of the surrounding environment by emitting laser light and measuring its return time. This module is suitable for high-precision environmental perception, especially obstacle detection. During the flight of the aircraft, LiDAR continuously scans the environment and provides the precise distance and position of obstacles. The system uses this data for local path planning and dynamic obstacle avoidance.

[0073] Visual sensor module 9: includes one or more cameras 13 for capturing environmental images; the system uses visual data to identify specific landmarks or environmental features to assist the aircraft in positioning and navigation; visual sensors play a key role especially when the GPS signal is weak or lost, and the visual inertial navigation system VINS can combine IMU data for environmental perception and positioning.

[0074] The implementation of the multi-source data fusion module is as follows:

[0075] Extended Kalman filter EKF: It is the core data fusion algorithm of the system. Its implementation steps include:

[0076] (1) Prediction step: Based on the real-time motion data provided by the IMU and LiDAR, the system predicts the current position and attitude of the aircraft through the state transition equation. This step uses the historical state data and the real-time motion information of the IMU to estimate the current state;

[0077] (2) Update step: When new observations are provided by GPS, LiDAR, or vision sensors, the EKF uses these observations to correct the predicted state and calculate the error covariance matrix of the state estimate; the updated state includes the precise position, attitude, and velocity of the aircraft;

[0078] Data weighting strategy: The data weighting strategy in EKF adjusts the data weight of each sensor according to the confidence of different sensors; the system monitors the strength and quality of the GPS signal, reduces the weight of GPS data when the signal is weak or lost, and increases the reliance on IMU, LiDAR and vision data; this dynamic weighting method ensures the robustness and accuracy of data fusion.

[0079] The implementation of the path planning and dynamic adjustment module is as follows:

[0080] Local path planning: The local path planning module detects obstacles in the flight path in real time and calculates a safe path based on the three-dimensional point cloud data generated by LiDAR. The implementation steps of this module are as follows:

[0081] (1) Obstacle detection: By processing LiDAR point cloud data, the system identifies obstacles in the current environment and marks their location and size.

[0082] (2) Path calculation: Based on the aircraft's current flight target, the system uses the A* algorithm or the RRT fast random tree search algorithm to calculate the optimal path and avoid detected obstacles.

[0083] (3) Path adjustment: If the aircraft encounters a new obstacle during flight, the system will dynamically adjust the flight path to avoid the obstacle in real time to ensure the safe movement of the aircraft.

[0084] Global Path Optimization: The global path planning module optimizes the flight path based on mission requirements to ensure that the aircraft completes the mission with the optimal path; global path optimization takes into account the time and energy constraints of the flight mission, combines multi-source data to calculate the optimal flight route before the flight mission begins, and dynamically adjusts it according to the actual environment during the flight.

[0085] The implementation of the adaptive control module is as follows:

[0086] PID controller: The PID controller adjusts the aircraft's attitude, speed, and direction to ensure that it flies smoothly along the planned path. The implementation steps are as follows:

[0087] (1) Attitude control: The system adjusts the aircraft's pitch, roll, and yaw angles in real time based on feedback from the IMU and visual sensors.

[0088] (2) Speed ​​control: The PID controller adjusts the aircraft's forward speed based on obstacles and target locations in the flight path to ensure that it can fly quickly and safely.

[0089] (3) Position control: The controller adjusts the aircraft's trajectory based on GPS and fused position information to keep it on the planned path;

[0090] Adaptive strategy adjustment: During flight, the system will adjust flight parameters according to changes in the external environment, such as wind speed changes, obstacle density, etc.; for example, when dense obstacles are detected, the system will reduce the flight speed to increase obstacle avoidance time, or increase the LiDAR scanning frequency to improve environmental perception accuracy.

[0091] The implementation of the data exception processing module is as follows:

[0092] Data integrity detection: The system monitors the integrity and validity of sensor data in real time. This module detects whether the sensor has data loss, outliers or noise interference, and issues a warning or adjusts the data fusion algorithm. For example, when the GPS signal fails, the system will issue an alarm and automatically switch to the IMU and visual sensor for positioning.

[0093] Data redundancy compensation: When some sensor data is lost, the system uses data and historical information from other sensors for redundancy compensation; for example, in the event of a GPS signal loss, the system calculates the aircraft’s current position through displacement integration from the IMU, while using visual sensors to identify landmarks or environmental features to further correct the position estimate.

[0094] An embodiment of the present invention provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments.

[0095] An embodiment of the present invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments.

[0096] An embodiment of the present invention provides an information data processing terminal, which is used to implement the multi-source data fusion and dynamic adjustment system suitable for complex flight environments.

[0097] Specific implementation plan: Multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments

[0098] Through the fusion processing of multi-source sensor data, the aircraft can achieve accurate positioning, path planning and flight control in complex flight environments, and has the ability to process abnormal data and dynamically adjust, thereby ensuring the aircraft's flight safety and the stability of mission execution.

[0099] Implementation steps:

[0100] S1: Data collection and preprocessing

[0101] 1) Multi-source data collection:

[0102] Various sensors such as IMU (inertial measurement unit), LiDAR (laser radar), GPS, visual sensors, etc. are installed on the aircraft to collect data in the flight environment in real time.

[0103] IMU provides the aircraft's acceleration and angular velocity data; LiDAR provides environmental distance and obstacle information; GPS is used to obtain the aircraft's global position coordinates; and the visual sensor captures images of the flight environment through the camera to assist in perceiving the surrounding environment.

[0104] 2) Data preprocessing:

[0105] Filtering: Perform basic filtering on the collected sensor data, such as using a Kalman filter (KF) or a low-pass filter to remove noise and sudden abnormal data to ensure the smoothness and continuity of the sensor data.

[0106] Data synchronization: Since different sensors have different sampling frequencies, multi-source sensor data needs to be synchronized. By aligning timestamps, we can ensure that the data from each sensor can be fused at the same time to avoid errors caused by data delays.

[0107] S2: Multi-source data fusion

[0108] 1) Preliminary fusion and state estimation:

[0109] Extended Kalman Filter (EKF): The pre-processed sensor data is input into the EKF, and its nonlinear system state estimation capability is used to estimate the real-time position, attitude, and speed of the aircraft. The EKF uses the state transition model and observation model, combines the historical state and current observation data, performs iterative updates, and finally outputs the precise state of the aircraft.

[0110] 2) Dynamic weight adjustment:

[0111] The system dynamically adjusts the weight of each sensor data in the fusion process according to the confidence of the sensor (such as GPS signal strength, IMU stability, etc.). For example, when the GPS signal is weak or interfered, the system will increase the weight of IMU and LiDAR to ensure the accuracy of position estimation; conversely, when LiDAR is unavailable due to weather conditions, the system will rely more on GPS and visual sensor data.

[0112] S3: Path planning and dynamic adjustment

[0113] 1) Local path planning:

[0114] The system performs local path planning based on fused multi-source data during real-time flight. Obstacles in the forward flight path are detected by LiDAR and visual sensors, and the obstacle avoidance path is calculated in real time in combination with state estimation data (aircraft position, attitude, speed).

[0115] Use the A* algorithm or RRT (Rapid Random Tree) algorithm to generate a safe flight path to ensure that the aircraft can avoid obstacles in complex environments.

[0116] 2) Dynamic path adjustment:

[0117] In complex environments, such as when flying over buildings or forests, the system will prioritize avoiding obstacles and, if necessary, reduce the flight speed to enhance obstacle avoidance capabilities. The system combines the flight mission requirements and, based on global path planning, adjusts the flight trajectory in real time to ensure that the aircraft completes the mission and ensures safety.

[0118] S4: Adaptive Control and Execution

[0119] 1) Adaptive control system:

[0120] Using a PID controller, the aircraft's flight attitude and speed are dynamically adjusted based on real-time sensor data feedback. The PID controller ensures that the aircraft flies smoothly along the planned path by constantly adjusting the control input (such as motor speed).

[0121] 2) Dynamic adjustment strategy:

[0122] The system dynamically adjusts the control strategy according to changes in the flight environment (such as increased wind speed, increased number of obstacles, etc.). For example, when encountering strong winds, the system will reduce the flight speed and improve attitude stability; when encountering a complex obstacle environment, the system will enhance the obstacle avoidance strategy to prioritize flight safety.

[0123] S5: Exception handling and redundancy compensation

[0124] 1) Sensor data integrity detection:

[0125] The system detects the integrity and accuracy of sensor data in real time. If it finds that the data of a sensor is missing or fluctuates abnormally (such as GPS signal loss, LiDAR data cannot be used due to weather conditions, etc.), the system will immediately activate the redundancy compensation mechanism.

[0126] 2) Data redundancy compensation mechanism:

[0127] GPS signal loss: When the GPS signal is lost due to occlusion or interference, the system combines the data from the IMU and visual sensors to estimate the position. Through the visual SLAM (simultaneous localization and mapping) algorithm, the system uses the environmental features captured by the camera and the acceleration and angular velocity data of the IMU to perform positioning and track calculation.

[0128] LiDAR data is unavailable: When LiDAR data is damaged due to bad weather (such as fog, rain or snow), the system uses IMU and visual sensors to compensate for the missing LiDAR data, ensuring that the aircraft continues to perceive the surrounding environment and maintain safe flight.

[0129] Historical data compensation: If multiple sensors malfunction at the same time, the system will use historical flight data (such as the prediction results of the state transition model) to perform short-term data compensation to ensure that the flight mission is not interrupted.

[0130] Effect:

[0131] 1) High-precision data fusion: Through the multi-source data fusion method of the extended Kalman filter (EKF), the aircraft can obtain high-precision position, attitude and velocity estimates in real time to ensure the accuracy of the flight path.

[0132] 2) Enhanced dynamic path planning and obstacle avoidance capabilities: The system combines real-time data from multi-source sensors and has the ability to dynamically adjust the flight path, especially in complex environments, to effectively avoid obstacles and ensure flight safety.

[0133] 3) Stability of adaptive control: Through the dynamic adjustment strategy of the PID controller, the aircraft can respond promptly to environmental changes (such as changes in wind speed, increase in obstacles, etc.) to maintain flight stability and accuracy of mission execution.

[0134] 4) Redundancy compensation ensures reliability: The system has the ability to handle sensor data loss and anomalies. Through the data redundancy compensation mechanism, it ensures that the flight mission can still proceed smoothly in complex and harsh environments, enhancing the system's robustness and fault tolerance.

[0135] This implementation effectively improves the aircraft's data processing capabilities, path planning accuracy and flight safety in complex flight environments, and has broad application prospects.

[0136] Specific implementation plan: Multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments

[0137] Through multi-source data fusion and dynamic adjustment algorithms, accurate positioning, path planning, adaptive control and exception handling of the aircraft in complex flight environments can be achieved, ensuring the safe flight and mission execution of the aircraft.

[0138] S1: Data collection and preprocessing

[0139] 1. Multi-source data collection:

[0140] -The system uses multiple sensors to collect aircraft flight data in real time. The sensors include IMU (Inertial Measurement Unit), LiDAR (Laser Radar), GPS, visual sensors, etc.

[0141] -IMU provides acceleration and angular velocity information; LiDAR measures the distance to obstacles in the flight environment; GPS obtains global position; and vision sensors capture images of the flight environment.

[0142] 2. Data preprocessing:

[0143] - Filter the acquired data (using Kalman filtering or low-pass filtering) to remove noise and sudden data anomalies.

[0144] -Time synchronization: synchronize the sampling frequencies of different sensors to ensure that the data can be matched and used at the same timestamp.

[0145] Python implementation example:

[0146] ```Python

[0147] import numpy as np

[0148] from pykalman import KalmanFilter

[0149] defpreprocess_data(sensor_data):

[0150] # Use Kalman filter to denoise

[0151] kf=KalmanFilter(initial_state_mean=0,n_dim_obs=1)

[0152] filtered_data=kf.em(sensor_data).smooth(sensor_data)[0]

[0153] return filtered_data

[0154] #Sample data

[0155] imu_data=np.array([1,2,3,4,5])

[0156] lidar_data=np.array([10,12,14,16,18])

[0157] gps_data=np.array([100,102,104,106,108])

[0158] #Preprocess data

[0159] filtered_imu=preprocess_data(imu_data)

[0160] filtered_lidar=preprocess_data(lidar_data)

[0161] filtered_gps=preprocess_data(gps_data)

[0162] ```

[0163] S2: Multi-source data fusion

[0164] 1. Extended Kalman Filter (EKF) initial fusion:

[0165] -Establish a state transfer model, define state variables (such as aircraft position, speed, attitude) and observation model (observation values ​​of each sensor).

[0166] -Use EKF to estimate the aircraft state based on historical state and current sensor data. Python implements the EKF core algorithm:

[0167] ```Python

[0168] import numpy as np defekf_predict(state,control,P,F,Q):

[0169] #Status prediction

[0170] state_pred=F@state+control

[0171] P_pred=F@P@F.T+Q

[0172] return state_pred,P_pred

[0173] defekf_update(state_pred,P_pred,Z,H,R):

[0174] # Kalman gain

[0175] K=P_pred@HT@np.linalg.inv(H@P_pred@H.T+R)

[0176] state_upd=state_pred+K@(ZH@state_pred)

[0177] P_upd = (np.eye(len(K))-K@H)@P_pred

[0178] return state_upd,P_upd

[0179] # Initialize state and covariance matrix

[0180] state = np.array([0,0,0]) #Assume state variables [x,y,θ]

[0181] P = np.eye(3)

[0182] F = np.eye(3) #state transfer matrix

[0183] Q = np.eye(3)*0.01#process noise

[0184] H = np.eye(3) #Observation matrix

[0185] R = np.eye(3)*0.05 # Observation noise

[0186] #Sensor measurement value

[0187] Z = np.array([10,5,0.1]) #Assume [x,y,θ]

[0188] control=np.array([0.1,0.05,0.01])

[0189] #EKF prediction and update

[0190] state_pred,P_pred=ekf_predict(state,control,P,F,Q)

[0191] state_upd,P_upd=ekf_update(state_pred,P_pred,Z,H,R)

[0192] ```

[0193] 2. Dynamic weight adjustment:

[0194] - Dynamically adjust the weight of sensors in EKF fusion according to the confidence of the sensors. For example, when the GPS signal is weak, the weight of IMU and LiDAR is increased.

[0195] Python implements weight adjustment:

[0196] ```Python

[0197] defadjust_weights(sensor_data,sensor_confidence):

[0198] #Sensor confidence weight adjustment

[0199] weighted_data=sensor_data*sensor_confidence

[0200] return np.sum(weighted_data) / np.sum(sensor_confidence)

[0201] #Assume the confidence of different sensors

[0202] imu_confidence=0.8

[0203] gps_confidence=0.6

[0204] lidar_confidence=0.7

[0205] # Dynamically adjusted fusion results

[0206] sensor_data=np.array([filtered_imu,filtered_gps,filtered_lidar])

[0207] confidences=np.array([imu_confidence,gps_confidence,lidar_confidence])

[0208] fused_data=adjust_weights(sensor_data,confidences)

[0209] S3: Path planning and dynamic adjustment

[0210] 1. Local path planning:

[0211] -Use the fused data to perform local path planning using A* or RRT algorithm and avoid obstacles in real time.

[0212] -Python implements A* path planning:

[0213]

[0214]

[0215] 2. Dynamic path adjustment:

[0216] -Adjust the flight path in real time based on sensor feedback and environmental changes to ensure the aircraft can avoid obstacles.

[0217] S4: Adaptive Control and Execution

[0218] 1.PID controller:

[0219] - Use a PID controller to adjust the aircraft's attitude and speed based on sensor feedback.

[0220] Python implements PID controller:

[0221]

[0222]

[0223] #Use PID to control the attitude

[0224] pid_controller=PIDController(1.0,0.1,0.05)

[0225] setpoint=10#target height

[0226] measured_value=8#actual height

[0227] control_signal=pid_controller.compute(setpoint,measured_value)

[0228] S5: Exception handling and redundancy compensation

[0229] 1. Redundancy compensation mechanism:

[0230] -When a sensor data is lost, use data from other sensors or historical data to compensate. For example, when LiDAR data is lost, combine IMU and visual sensor data for environmental perception.

[0231] Compensation mechanism implementation:

[0232]

[0233] Effect:

[0234] 1. High-precision fusion: EKF is used to achieve efficient fusion of multi-source data and dynamically adjust sensor weights to improve data fusion accuracy in complex environments.

[0235] 2. Real-time path planning and dynamic adjustment: Use A* or RRT algorithm to implement real-time path planning, and dynamically adjust the flight path based on sensor feedback to ensure that the aircraft can effectively avoid obstacles.

[0236] 3. Adaptive control: The PID controller dynamically adjusts the aircraft's attitude and speed to ensure the aircraft's flight stability and accuracy.

[0237] 4. Redundancy compensation: Through the redundant compensation mechanism, when a sensor fails, the system can quickly compensate to maintain the continuity and safety of the flight mission.

[0238] Example 1: Aircraft data fusion and dynamic adjustment under complex weather conditions

[0239] In complex flight environments, such as storms or strong winds, aircraft need to cope with severe airflow disturbances and uncertain environmental changes. The system collects data from multiple sensors in real time through the S1 step, including barometers, accelerometers, wind speed sensors, GPS and other sensors. Through data preprocessing, noise and sudden outliers are filtered out to ensure data accuracy. After entering the S2 multi-source data fusion stage, the system uses the extended Kalman filter (EKF) to fusion estimate the state of the aircraft, and combines the outputs of multiple sensors to estimate the precise position, speed and attitude of the aircraft. The system will dynamically weight each sensor according to the current wind speed, wind direction and GPS signal strength to cope with the uncertain impact of meteorological conditions on sensor data.

[0240] When the aircraft encounters strong winds or sudden weather changes, the system automatically adjusts the aircraft's heading through the path planning and dynamic adjustment function of S3 to avoid areas with high wind speeds, and optimizes the flight path in accordance with mission requirements. During the execution process (S4), the PID controller makes real-time adjustments based on the aircraft's attitude and speed changes to ensure stable flight of the aircraft. When sensor data is abnormal or the signal is lost (S5), the system uses the historical wind speed data or the aircraft's inertial measurement data to compensate through a redundant compensation mechanism to ensure the aircraft's continued stable flight.

[0241] Example 2: Obstacle avoidance and path adjustment during high-altitude flight

[0242] In complex flight environments, such as high-altitude flights in cities or mountainous areas, aircraft need to avoid obstacles such as buildings and mountains in real time. In this scenario, the system collects data from sensors such as lidar, cameras, GPS, and inertial measurement units (IMUs) through step S1 to perceive the surrounding environment in real time. The preprocessed data is combined with the distance information detected by the lidar and the visual data of the camera through the EKF fusion estimation module in step S2 to accurately estimate the distance, position, and relative motion state of surrounding obstacles.

[0243] During the S3 path planning and dynamic adjustment process, the system calculates the safe distance from obstacles in real time and adjusts the path of the aircraft to avoid collisions. In complex high-altitude flight scenarios, the system prioritizes avoiding obstacles and optimizes the global flight route while planning paths in the local environment. During the S4 adaptive control execution process, the PID controller automatically adjusts the aircraft's attitude based on sensor feedback to ensure that the aircraft passes through areas with dense obstacles in a safe path. If a sensor such as a lidar signal is lost or a camera fails, the system uses the S5 redundant compensation mechanism to compensate using data from the inertial measurement unit or GPS to ensure the normal flight of the aircraft until the sensor returns to normal.

[0244] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. It can be understood by a person of ordinary skill in the art that the above-mentioned devices and methods can be implemented using computer executable instructions and / or contained in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium. The device and its modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0245] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments, characterized in that: The algorithm includes: S1: Data collection and preprocessing: The system first collects multi-source data in the aircraft flight environment in real time through the sensor module; for each type of sensor data, the system will perform basic preprocessing to ensure the accuracy and consistency of the data; S2: Multi-source data fusion: The pre-processed data enters the multi-source data fusion module; first, the extended Kalman filter (EKF) performs preliminary fusion and state estimation on the sensor data; the system predicts the current position, attitude and speed of the aircraft by establishing a state transition model, combining the current observation data with the historical state; then, the system dynamically adjusts the weight of each sensor in the fusion process according to the confidence of different sensors; S3: Path planning and dynamic adjustment: The system performs local path planning based on the fused data. During this process, the system detects obstacles in the flight path in real time, calculates a safe path, and dynamically adjusts the aircraft's flight trajectory. At the same time, the system optimizes the flight path according to the global mission requirements to ensure that the aircraft completes the flight mission with the optimal path. In complex environments, the system will prioritize avoiding obstacles to ensure flight safety. S4: Adaptive control and execution: After the path planning is completed, the system executes flight control through the adaptive control module; the PID controller adjusts the aircraft's attitude and speed in real time according to the sensor feedback to ensure that the aircraft flies smoothly according to the planned path; at the same time, the system will dynamically adjust the control strategy according to environmental changes; S5: Exception handling and redundancy compensation: The system detects the integrity of sensor data in real time; when some sensor data is lost or abnormal, the system uses the data of other sensors or historical data to compensate through the data redundancy compensation mechanism.

2. The multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments as claimed in claim 1, characterized in that: The data collection and preprocessing in step S1 includes using the sensor module to collect multi-source data in the aircraft flight environment in real time, and filtering and denoising the data of each sensor to improve the accuracy and consistency of the data, ensuring that the sensor data can provide reliable input for subsequent processing.

3. The multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments as claimed in claim 1, characterized in that: The multi-source data fusion in step S2 uses an extended Kalman filter (EKF) to perform preliminary fusion of multi-source sensor data, and predicts the real-time state of the aircraft by establishing a state transition model and combining the current sensor observation data with the historical state; this process dynamically adjusts the weights according to the confidence of different sensors to improve the data fusion accuracy in complex environments.

4. The multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments as claimed in claim 1, characterized in that: The path planning and dynamic adjustment in the S3 step performs local path planning based on the fused data, detects obstacles in the flight path in real time, calculates a safe path, and dynamically adjusts the path according to obstacles, flight missions, and the real-time status of the aircraft; the system prioritizes obstacle avoidance in complex environments based on changes in sensor data to ensure aircraft flight safety.

5. The multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments as claimed in claim 1, characterized in that: The exception handling and redundancy compensation mechanism in step S5 monitors the integrity of sensor data in real time. When some sensor data is lost or abnormal, the system uses the data of other sensors to compensate; for example, when the LiDAR signal is lost, the system combines the data of IMU and visual sensors or historical data for environmental perception to ensure the continuity and reliability of data during flight.

6. A multi-source data fusion and dynamic adjustment system suitable for complex flight environments based on the multi-source data fusion and dynamic adjustment algorithm suitable for complex flight environments as claimed in claim 1, characterized in that: The system specifically includes: The sensor module provides global position information, but the data will be incomplete or distorted when the signal is blocked or interfered. The accelerometer and gyroscope provide the aircraft's attitude and motion information, but the IMU will have drift errors during long-term flight. The laser scans the surrounding environment to generate a three-dimensional point cloud, providing high-precision environmental perception, which is suitable for obstacle detection. The camera obtains environmental images for identifying specific landmarks or environmental features. The multi-source data fusion module is connected to the sensor module to combine the sensor data of GPS, IMU and LiDAR to generate a unified flight state estimation; EKF can perform state prediction and error correction through historical data and models when the sensor data is imperfect; and dynamically adjust the weight of the sensor according to the confidence of different sensors; The path planning and dynamic adjustment module is connected to the multi-source data fusion module and combines sensor data to generate a safe flight path based on the current environment in real time. The module detects and avoids obstacles in the flight path based on sensor feedback. Based on the global flight mission requirements, the module dynamically optimizes the flight path in combination with multi-source data to ensure that the aircraft can complete the flight mission efficiently and safely in a complex environment. The adaptive control module is connected to the multi-source data fusion module to adjust the aircraft's attitude, speed and direction in real time based on sensor feedback to ensure that the aircraft flies smoothly according to the planned path; the system will adjust the aircraft's attitude, speed and direction in real time based on changes in the flight environment; The data anomaly processing module is connected to the sensor module to detect the integrity and validity of sensor data in real time. When some sensor data is lost or abnormal, the system will issue a warning and adjust the data fusion strategy. When some sensor data is lost, the system can perform redundancy compensation through data from other sensors and historical flight information.

7. The multi-source data fusion and dynamic adjustment system suitable for complex flight environments as claimed in claim 6, characterized in that: The implementation of the sensor module is as follows: GPS module: The system integrates a GPS receiver to obtain global location information through satellite signals; the data preprocessing module filters the received GPS signals, removes noise and jump data, and calculates the geographic coordinates of the current location; because the GPS signal is blocked or interfered, it needs to be fused with data from other sensors to supplement the positioning accuracy; IMU module: The IMU module includes an accelerometer and a gyroscope, which are used to measure the linear acceleration and angular velocity of the aircraft. Through high-sampling rate data collection, it provides real-time information on the aircraft's attitude and motion. The system uses IMU data to track the aircraft's instantaneous state and calculates displacement information through integration. However, there is a drift problem with IMU data, and the system uses EKF to correct its errors. LiDAR module: LiDAR generates 3D point cloud data of the surrounding environment by emitting lasers and measuring their return time. This module is suitable for high-precision environmental perception, especially obstacle detection. During the flight of the aircraft, LiDAR will continuously scan the environment and provide the precise distance and direction of obstacles. The system uses this data for local path planning and dynamic obstacle avoidance. Visual sensor module: includes one or more cameras for capturing images of the environment. The system uses visual data to identify specific landmarks or environmental features to assist the aircraft in positioning and navigation. Visual sensors play a key role, especially when the GPS signal is weak or lost. The visual inertial navigation system VINS can combine IMU data for environmental perception and positioning.

8. The multi-source data fusion and dynamic adjustment system suitable for complex flight environments as claimed in claim 6, characterized in that: The implementation of the multi-source data fusion module is as follows: Extended Kalman filter EKF: It is the core data fusion algorithm of the system. Its implementation steps include: (1) Prediction step: Based on the real-time motion data provided by the IMU and LiDAR, the system predicts the current position and attitude of the aircraft through the state transition equation. This step uses the historical state data and the real-time motion information of the IMU to estimate the current state. (2) Update step: When new observations are provided by GPS, LiDAR, or vision sensors, the EKF uses these observations to correct the predicted state and calculate the error covariance matrix of the state estimate; the updated state includes the precise position, attitude, and velocity of the aircraft; Data weighting strategy: The data weighting strategy in EKF adjusts the data weight of each sensor according to the confidence of different sensors; the system monitors the strength and quality of the GPS signal, reduces the weight of GPS data when the signal is weak or lost, and increases the reliance on IMU, LiDAR and vision data; this dynamic weighting method ensures the robustness and accuracy of data fusion.

9. The multi-source data fusion and dynamic adjustment system suitable for complex flight environments as claimed in claim 6, characterized in that: The implementation of the path planning and dynamic adjustment module is as follows: Local path planning: The local path planning module detects obstacles in the flight path in real time and calculates a safe path based on the three-dimensional point cloud data generated by LiDAR. The implementation steps of this module are as follows: (1) Obstacle detection: By processing LiDAR point cloud data, the system identifies obstacles in the current environment and marks their location and size; (2) Path calculation: Based on the aircraft's current flight target, the system uses the A* algorithm or the RRT fast random tree search algorithm to calculate the optimal path and avoid detected obstacles; (3) Path adjustment: If the aircraft encounters a new obstacle during flight, the system will dynamically adjust the flight path to avoid the obstacle in real time to ensure the aircraft's safe progress; Global Path Optimization: The global path planning module optimizes the flight path based on mission requirements to ensure that the aircraft completes the mission with the optimal path; global path optimization takes into account the time and energy constraints of the flight mission, combines multi-source data to calculate the optimal flight route before the flight mission begins, and dynamically adjusts it according to the actual environment during the flight.

10. The multi-source data fusion and dynamic adjustment system suitable for complex flight environments as claimed in claim 6, characterized in that: The implementation of the adaptive control module is as follows: PID controller: The PID controller adjusts the aircraft's attitude, speed, and direction to ensure that it flies smoothly along the planned path. The implementation steps are as follows: (1) Attitude control: The system adjusts the aircraft's pitch, roll, and yaw angles in real time based on feedback from the IMU and visual sensors. (2) Speed ​​control: The PID controller adjusts the aircraft's forward speed based on obstacles and target locations in the flight path to ensure that it can fly quickly and safely. (3) Position control: The controller adjusts the aircraft's trajectory based on GPS and fused position information to keep it on the planned path; Adaptive strategy adjustment: During flight, the system will adjust flight parameters according to changes in the external environment, such as wind speed changes, obstacle density, etc.; for example, when dense obstacles are detected, the system will reduce the flight speed to increase obstacle avoidance time, or increase the LiDAR scanning frequency to improve environmental perception accuracy.

Images