Control method and system for precise landing of unmanned aerial vehicle

Abstract

A control method and system for precise landing of an unmanned aerial vehicle. The method comprises the following steps: on the basis of a multi-sensor fusion technique, using a lidar and a stereo vision algorithm to collect environmental data and perform primary processing, and generating comprehensive environmental perception data (step S1); on the basis of the comprehensive environmental perception data, using a digital elevation model algorithm to re-construct a three-dimensional terrain of a landing area, and generating a three-dimensional terrain model (step S2); on the basis of the three-dimensional terrain model, using an A* search path-planning algorithm to perform risk assessment and plan a safe landing path, and generating an optimized landing path (step S3); on the basis of the optimized landing path, using an adaptive control algorithm to adjust a flight attitude in real time by means of a fuzzy logic controller, and generating flight parameter adjustment (step S4); on the basis of the flight parameter adjustment, using machine vision and a decision tree algorithm to execute autonomous obstacle avoidance and emergency response, and generating a safe landing execution scheme (step S5); and on the basis of the safe landing execution scheme, using an ultrasonic sensor and a ground feedback system to confirm and fine-adjust a landing point, and completing final landing confirmation (step S6).

Description

A UAV precise landing control method and system Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) control systems, and in particular to a UAV precision landing control method and system. Background Art

[0002] The field of unmanned aerial vehicle (UAV) control systems is a key branch of aerospace engineering, involving the automated and semi-automated control of unmanned aerial vehicles (UAVs). This field integrates multiple disciplines, including aerodynamics, automatic control theory, signal processing, and machine vision, aiming to achieve efficient, safe, and intelligent operation of UAVs. UAV control systems encompass not only flight control (such as speed, altitude, and directional control), but also mission management (such as path planning and target recognition) and environmental interaction (such as obstacle avoidance and landing control).

[0003] Precision landing control for drones (UAVs) refers to a technical approach that enables drones to land safely and accurately at a predetermined or specific location. This approach typically requires a comprehensive consideration of the drone's flight dynamics, environmental factors (such as wind speed and terrain), sensor data (such as GPS and vision sensors), and control algorithms. Precision landing is crucial for drone safety and mission efficiency, especially in urban environments, mountainous areas, and other complex terrains. The primary goal is to ensure that the drone safely and accurately returns to the predetermined landing point or landing location after completing its mission. This is crucial for ensuring the safety of the drone and its payload, improving mission efficiency, and optimizing energy utilization. By applying precision landing control methods, drones can achieve highly accurate landings in a variety of environments and conditions. This not only reduces risk and potential losses, but also increases mission reliability and success rates. Furthermore, precision landing technology enables drones to perform missions in remote areas or complex terrain, expanding the scope of drone applications.

[0004] Traditional drone landing control methods have several shortcomings. They lack efficient three-dimensional terrain reconstruction capabilities, resulting in a lack of sufficient terrain information during landing, increasing landing risks. Furthermore, traditional path planning algorithms are inefficient in complex environments and cannot quickly adapt to environmental changes, resulting in suboptimal landing path selection. Traditional methods lack sufficient flexibility and adaptability in flight parameter adjustment, making them difficult to cope with rapidly changing environments. Traditional methods also have limited capabilities in autonomous obstacle avoidance and emergency response, making them ineffective in dealing with unexpected obstacles and increasing safety risks. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a method and system for precise landing control of an unmanned aerial vehicle.

[0006] In order to achieve the above-mentioned object, the present invention adopts the following technical solution: a method for controlling the precise landing of an unmanned aerial vehicle, comprising the following steps:

[0007] S1: Based on multi-sensor fusion technology, it uses lidar and stereo vision algorithms to collect environmental data and perform preliminary processing to generate comprehensive environmental perception data;

[0008] S2: Based on the comprehensive environmental perception data, a digital elevation model algorithm is used to reconstruct the three-dimensional terrain of the landing area to generate a three-dimensional terrain model;

[0009] S3: Based on the three-dimensional terrain model, using the A* search path planning algorithm, performing risk assessment and planning a safe landing path, and generating an optimized landing path;

[0010] S4: Based on the optimized landing path, an adaptive control algorithm is used to adjust the flight attitude in real time through a fuzzy logic controller to generate flight parameter adjustments;

[0011] S5: Based on the flight parameter adjustments, using machine vision and decision tree algorithms, perform autonomous obstacle avoidance and emergency response, and generate a safe landing execution plan;

[0012] S6: Based on the safe landing execution plan, ultrasonic sensors and ground feedback systems are used to confirm and fine-tune the landing point to complete the final landing confirmation;

[0013] The comprehensive environmental perception data specifically refers to multidimensional data including terrain features, obstacle information, and meteorological conditions. The three-dimensional terrain model specifically refers to a three-dimensional map including terrain height, slope, and stability. The optimized landing path specifically refers to the optimal landing trajectory optimized based on environmental risks and drone capabilities. The flight parameter adjustment specifically refers to flight parameter adjustment in response to real-time environmental changes, including speed, angle, and altitude. The safe landing execution plan specifically refers to automatic adjustment and emergency landing strategies when facing sudden obstacles.

[0014] As a further solution of the present invention, based on multi-sensor fusion technology, laser radar and stereo vision algorithms are used to collect environmental data and perform preliminary processing to generate comprehensive environmental perception data. The specific steps are as follows:

[0015] S101: Based on multi-sensor fusion technology, it uses LiDAR algorithms to collect environmental distance information and generate preliminary distance perception data;

[0016] S102: Based on the preliminary distance perception data, using a binocular stereo matching algorithm, correcting the image data to generate corrected visual perception data;

[0017] S103: Based on the corrected visual perception data and the preliminary distance perception data, Kalman filtering technology is used to integrate the data to generate optimized environmental perception data;

[0018] S104: Based on the optimized environmental perception data, a data analysis algorithm is used to perform a comprehensive analysis to generate comprehensive environmental perception data.

[0019] As a further solution of the present invention, based on the comprehensive environmental perception data, a digital elevation model algorithm is used to reconstruct the three-dimensional terrain of the landing area. The steps of generating the three-dimensional terrain model are specifically as follows:

[0020] S201: Based on the comprehensive environmental perception data, a terrain analysis algorithm is used to determine terrain features and generate preliminary terrain feature data;

[0021] S202: Based on the preliminary terrain feature data, using a digital elevation model algorithm, measuring the terrain height and shape to generate refined elevation data;

[0022] S203: Reconstructing the three-dimensional terrain based on the refined elevation data using three-dimensional modeling technology to generate a preliminary three-dimensional terrain model;

[0023] S204: Based on the preliminary three-dimensional terrain model, a model optimization algorithm is used to refine and optimize the model to generate a final three-dimensional terrain model.

[0024] As a further solution of the present invention, based on the three-dimensional terrain model, an A* search path planning algorithm is used to perform risk assessment and plan a safe landing path. The steps of generating an optimized landing path are specifically as follows:

[0025] S301: Based on the three-dimensional terrain model, a risk analysis algorithm is used to perform risk assessment and generate a risk assessment report;

[0026] S302: Based on the risk assessment report, using the A* search algorithm to plan a preliminary path and generate a preliminary landing path;

[0027] S303: Based on the preliminary landing path, using path optimization technology, taking into account efficiency and safety factors, to generate an optimized landing path;

[0028] S304: Based on the optimized landing path, final confirmation and adjustment technology is used to ensure the path optimization and generate a final optimized landing path.

[0029] As a further solution of the present invention, based on the optimized landing path, an adaptive control algorithm is used to adjust the flight attitude in real time through a fuzzy logic controller, and the steps of generating flight parameter adjustments are specifically as follows:

[0030] S401: Based on the optimized landing path, a fuzzy logic control algorithm is used to analyze the flight attitude and generate a preliminary flight parameter adjustment plan;

[0031] S402: Based on the preliminary flight parameter adjustment plan, employing adaptive feedback technology to refine the adjustment and generate an advanced flight parameter adjustment plan;

[0032] S403: Based on the advanced flight parameter adjustment solution, using real-time data processing technology to optimize the flight attitude and generate a real-time flight parameter adjustment solution;

[0033] S404: Based on the real-time flight parameter adjustment plan, a dynamic adjustment algorithm is used to complete the final adjustment and generate a final flight parameter adjustment plan.

[0034] As a further embodiment of the present invention, based on the flight parameter adjustment, machine vision and decision tree algorithms are used to perform autonomous obstacle avoidance and emergency response, and the steps of generating a safe landing execution plan are as follows:

[0035] S501: Based on the flight parameter adjustment plan, using machine vision recognition technology, detect obstacles and generate an obstacle detection report;

[0036] S502: Based on the obstacle detection report, a decision tree algorithm is used to replan the path and generate a replanned flight path;

[0037] S503: Based on the re-planned flight path, using obstacle avoidance strategy optimization technology, executing obstacle avoidance and generating an obstacle avoidance execution plan;

[0038] S504: Based on the obstacle avoidance execution plan, an emergency response mechanism is adopted to adjust the flight strategy and generate a safe landing execution plan.

[0039] As a further solution of the present invention, based on the safe landing execution scheme, ultrasonic sensors and a ground feedback system are used to confirm and fine-tune the landing point. The steps for completing the final landing confirmation are as follows:

[0040] S601: Based on the safe landing execution plan, use ultrasonic detection technology to preliminarily detect the landing point and generate a preliminary landing point detection report;

[0041] S602: Based on the preliminary landing point detection report, using ground data analysis technology, analyze the landing point and generate an advanced landing point analysis report;

[0042] S603: Based on the advanced landing point analysis report, fine-tune the landing point position using precise positioning technology to generate a precise landing point adjustment plan;

[0043] S604: Based on the precise landing point adjustment plan, ground adaptability analysis technology and ultrasonic fine adjustment technology are used to confirm the adaptability of the landing point and generate a final landing confirmation report.

[0044] A UAV precision landing control system is used to execute the above-mentioned UAV precision landing control method. The system includes an environment perception module, a terrain reconstruction module, a path planning module, a flight adjustment module, an obstacle avoidance execution module, a landing point fine-tuning module, and a final confirmation module.

[0045] As a further solution of the present invention, the environment perception module is based on multi-sensor fusion technology, adopts laser radar and stereo vision algorithm to collect environmental data and generate comprehensive environment perception data;

[0046] The terrain reconstruction module reconstructs the landing area terrain based on the comprehensive environmental perception data and uses a digital elevation model algorithm to generate a three-dimensional terrain model;

[0047] The path planning module uses the A* search path planning algorithm based on the three-dimensional terrain model to perform risk assessment and path planning to generate an optimized landing path;

[0048] The flight adjustment module uses an adaptive control algorithm to adjust the flight attitude based on the optimized landing path and generates a flight parameter adjustment plan;

[0049] The obstacle avoidance execution module uses machine vision and decision tree algorithms based on the flight parameter adjustment plan to perform obstacle avoidance and emergency response and generate a safe landing execution plan;

[0050] The landing point fine-tuning module uses ultrasonic sensors and a ground feedback system based on the safe landing execution plan to fine-tune the landing point and generate a precise landing point adjustment plan;

[0051] The final confirmation module is based on a precise landing point adjustment plan, adopts ground adaptability analysis technology and ultrasonic fine adjustment technology to complete the final confirmation and generate a final landing confirmation report.

[0052] As a further solution of the present invention, the environment perception module includes a distance perception submodule, a visual correction submodule, a data fusion submodule, and an analysis and synthesis submodule;

[0053] The terrain reconstruction module includes a terrain feature analysis submodule, an elevation data refinement submodule, a three-dimensional modeling submodule, and a model optimization submodule;

[0054] The path planning module includes a risk assessment submodule, a preliminary path planning submodule, a path optimization submodule, and a path confirmation submodule;

[0055] The flight adjustment module includes an attitude analysis submodule, a feedback adjustment submodule, a real-time data processing submodule, and a dynamic adjustment submodule;

[0056] The obstacle avoidance execution module includes an obstacle detection submodule, a path replanning submodule, an obstacle avoidance strategy optimization submodule, and an emergency response submodule;

[0057] The landing point fine-tuning module includes a preliminary detection submodule, a ground analysis submodule, a precise positioning submodule, and a fine-tuning confirmation submodule;

[0058] The final confirmation module includes an adaptability analysis submodule, an ultrasonic adjustment submodule, a ground feedback submodule, and a final confirmation submodule.

[0059] Compared with the prior art, the advantages and positive effects of the present invention are:

[0060] In the present invention, the three-dimensional terrain of the landing area can be reconstructed through the digital elevation model algorithm, providing the UAV with more detailed terrain information, thereby enhancing the accuracy of landing decisions. Path planning algorithms, such as A* search, effectively perform risk assessments and plan a safe landing path, reducing the risks during the landing process. The adaptive control algorithm, through a fuzzy logic controller, can quickly adjust flight parameters according to real-time environmental changes, improving the flexibility and adaptability of the flight. The combination of machine vision and decision tree algorithms enables the UAV to perform autonomous obstacle avoidance and emergency response, enhancing its ability to deal with emergencies. The application of ultrasonic sensors and ground feedback systems makes the confirmation and fine-tuning of the landing point more accurate, ensuring the safety and accuracy of the landing. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] FIG1 is a schematic diagram of the workflow of the present invention;

[0062] FIG2 is a flow chart of the refinement of S1 of the present invention;

[0063] FIG3 is a flow chart of the refinement of S2 of the present invention;

[0064] FIG4 is a flow chart of the refinement of S3 of the present invention;

[0065] FIG5 is a flow chart of the refinement of S4 of the present invention;

[0066] FIG6 is a flow chart of the refinement of S5 of the present invention;

[0067] FIG7 is a flow chart of the refinement of S6 of the present invention;

[0068] FIG8 is a system flow chart of the present invention;

[0069] FIG9 is a schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0070] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0071] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0072] Example

[0073] Referring to FIG1 , the present invention provides a technical solution: a method for controlling precise landing of a UAV, comprising the following steps:

[0074] S1: Based on multi-sensor fusion technology, it uses lidar and stereo vision algorithms to collect environmental data and perform preliminary processing to generate comprehensive environmental perception data;

[0075] S2: Based on the comprehensive environmental perception data, the digital elevation model algorithm is used to reconstruct the three-dimensional terrain of the landing area and generate a three-dimensional terrain model;

[0076] S3: Based on the 3D terrain model, the A* search path planning algorithm is used to perform risk assessment and plan a safe landing path to generate an optimized landing path;

[0077] S4: Based on the optimized landing path, an adaptive control algorithm is used to adjust the flight attitude in real time through a fuzzy logic controller to generate flight parameter adjustments;

[0078] S5: Based on flight parameter adjustments, it uses machine vision and decision tree algorithms to perform autonomous obstacle avoidance and emergency response, generating a safe landing execution plan;

[0079] S6: Based on the safe landing execution plan, ultrasonic sensors and ground feedback systems are used to confirm and fine-tune the landing point and complete the final landing confirmation;

[0080] The comprehensive environmental perception data specifically refers to multi-dimensional data including terrain features, obstacle information, and meteorological conditions. The three-dimensional terrain model specifically refers to a three-dimensional map including terrain height, slope, and stability. The optimized landing path specifically refers to the optimal landing trajectory optimized according to environmental risks and UAV capabilities. The flight parameter adjustment specifically refers to the adjustment of flight parameters in response to real-time environmental changes, including speed, angle, and altitude. The safe landing execution plan specifically refers to automatic adjustment and emergency landing strategies when facing sudden obstacles.

[0081] Through multi-sensor fusion technology, combined with lidar and stereo vision algorithms, drones can accurately collect and process environmental data in real time, greatly improving the accuracy and reliability of environmental perception. The three-dimensional terrain model reconstructed by the digital elevation model algorithm allows drones to more accurately identify landing areas and reduce landing risks. The use of the A* search path planning algorithm enhances flight safety, and the optimized landing path takes into account environmental risks and drone capabilities to ensure a safe landing. Adaptive control algorithms and fuzzy logic controllers enable drones to adjust flight parameters in real time according to environmental changes, improving adaptability and flexibility during flight. The application of machine vision and decision tree algorithms improves the drone's autonomous obstacle avoidance and emergency response capabilities, reducing the risk of flight failure due to collisions or other emergencies. The combined use of ultrasonic sensors and ground feedback systems ensures the accuracy and safety of the landing point before landing.

[0082] Refer to Figure 2. Based on multi-sensor fusion technology, using LiDAR and stereo vision algorithms, environmental data is collected and initially processed to generate comprehensive environmental perception data. The specific steps are as follows:

[0083] S101: Based on multi-sensor fusion technology, it uses LiDAR algorithms to collect environmental distance information and generate preliminary distance perception data;

[0084] S102: Based on the preliminary distance perception data, a binocular stereo matching algorithm is used to correct the image data to generate corrected visual perception data;

[0085] S103: Based on the corrected visual perception data and preliminary distance perception data, Kalman filtering technology is used to integrate the data and generate optimized environmental perception data;

[0086] S104: Based on the optimized environmental perception data, a data analysis algorithm is used to perform a comprehensive analysis to generate comprehensive environmental perception data.

[0087] In the S101, LiDAR technology is used in the initial phase to collect distance information from the environment. LiDAR can accurately measure the distance to surrounding objects and generate preliminary distance perception data. This data provides basic information about the position and shape of objects in the environment, laying the foundation for subsequent data processing.

[0088] In S102, the image data acquired from the cameras is processed using a binocular stereo matching algorithm. This algorithm calibrates and synthesizes 3D visual perception data by comparing images from two cameras viewed from slightly different angles. This step enhances visual understanding of the environment, making the data more accurate and detailed.

[0089] In S103, the corrected visual perception data and preliminary range perception data are combined using Kalman filtering technology for data fusion. Kalman filtering is a highly efficient algorithm used to fuse data from multiple sources, optimizing and improving data accuracy. This step is the core of multi-sensor fusion, integrating information from different sensors to generate more accurate and reliable environmental perception data.

[0090] In S104, the optimized environmental perception data is comprehensively analyzed using data analysis algorithms to generate comprehensive environmental perception data. This step involves further parsing and understanding the fused data to form a comprehensive understanding of the environment. This includes identifying key features in the environment, dynamic changes, and their potential impact on the current scene.

[0091] Refer to Figure 3. Based on the comprehensive environmental perception data, the digital elevation model algorithm is used to reconstruct the three-dimensional terrain of the landing area. The specific steps for generating the three-dimensional terrain model are as follows:

[0092] S201: Based on the comprehensive environmental perception data, a terrain analysis algorithm is used to determine terrain features and generate preliminary terrain feature data;

[0093] S202: Based on the preliminary terrain feature data, a digital elevation model algorithm is used to measure the terrain height and shape to generate refined elevation data;

[0094] S203: Based on the refined elevation data, the 3D terrain is reconstructed using 3D modeling technology to generate a preliminary 3D terrain model;

[0095] S204: Based on the preliminary three-dimensional terrain model, a model optimization algorithm is used to refine and optimize the model to generate a final three-dimensional terrain model.

[0096] In S201, terrain analysis algorithms are used to process the integrated environmental perception data. This step aims to identify and determine key terrain features, such as slope, terrain type, and key landmarks. By analyzing these features, preliminary terrain feature data can be generated. This stage provides the necessary foundational information for the creation of a digital elevation model (DEM).

[0097] In S202, based on the preliminary terrain feature data, a digital elevation model (DEM) algorithm is used to measure the terrain's height and shape. A DEM is a mathematical model that represents the height of the ground surface and accurately describes the vertical dimensions of the terrain. This step generates refined elevation data, providing accurate height and contour information for 3D modeling.

[0098] In S203, 3D modeling technology is used to reconstruct the 3D terrain based on the refined elevation data. This step involves converting the elevation data into a visual 3D model. This preliminary 3D terrain model can intuitively display the shape, height, and structure of the terrain, and is an important tool for understanding complex terrain.

[0099] In S204, the preliminary 3D terrain model is refined and optimized using a model optimization algorithm. This includes adjusting the model's accuracy, improving visual effects, and enhancing its practicality. The optimized model provides a more accurate and practical representation of the 3D terrain, providing an important basis for landing area selection and planning.

[0100] Refer to Figure 4. Based on the 3D terrain model, the A* search path planning algorithm is used to perform risk assessment and plan a safe landing path. The steps for generating an optimized landing path are as follows:

[0101] S301: Based on the three-dimensional terrain model, a risk analysis algorithm is used to perform risk assessment and generate a risk assessment report;

[0102] S302: Based on the risk assessment report, an A* search algorithm is used to plan a preliminary path and generate a preliminary landing path;

[0103] S303: Based on the preliminary landing path, an optimized landing path is generated by using path optimization technology and taking into account efficiency and safety factors.

[0104] S304: Based on the optimized landing path, final confirmation and adjustment technology is used to ensure the path optimization and generate the final optimized landing path.

[0105] In S301, a comprehensive risk assessment is performed using a risk analysis algorithm based on the 3D terrain model. This step includes analyzing terrain characteristics, potential obstacles, and other factors that may affect landing safety. The assessment considers slope, surface material, known obstacles (such as boulders and cliffs), and other environmental factors. After these analyses are completed, a risk assessment report is generated, clearly indicating the potential risk level for each area.

[0106] In S302, based on the risk assessment report, an initial landing path is planned using the A* search algorithm. The A* search algorithm is an efficient path-finding algorithm that considers various factors (such as distance, time, and risk) to find the optimal path. During this phase, the algorithm searches the 3D terrain model, attempting to find a safe and efficient initial landing path.

[0107] In S303, the initially planned landing path is refined and improved using path optimization technology. Factors such as efficiency (e.g., shortest path) and safety (avoiding high-risk areas) are considered to further optimize the path. This involves adjusting the path to avoid more obstacles or select more stable terrain.

[0108] In S304, based on the optimized landing path, final verification and adjustment techniques are implemented to ensure the optimal path. During this phase, the optimized path undergoes final inspection and necessary fine-tuning to ensure its practicality and safety. This includes adapting to environmental changes or adjusting the path based on the latest intelligence.

[0109] Referring to Figure 5, based on the optimized landing path, the adaptive control algorithm is used to adjust the flight attitude in real time through the fuzzy logic controller. The specific steps for generating flight parameter adjustments are as follows:

[0110] S401: Based on the optimized landing path, the fuzzy logic control algorithm is used to analyze the flight attitude and generate a preliminary flight parameter adjustment plan;

[0111] S402: Based on the preliminary flight parameter adjustment plan, the adaptive feedback technology is used to refine the adjustment and generate an advanced flight parameter adjustment plan;

[0112] S403: Based on the advanced flight parameter adjustment plan, real-time data processing technology is used to optimize the flight attitude and generate a real-time flight parameter adjustment plan;

[0113] S404: Based on the real-time flight parameter adjustment plan, a dynamic adjustment algorithm is used to complete the final adjustment and generate a final flight parameter adjustment plan.

[0114] In S401, based on the optimized landing path, a fuzzy logic control algorithm analyzes the current flight attitude. Fuzzy logic controllers are capable of handling uncertainty and ambiguity, effectively adapting to complex or changing flight environments. During this phase, the algorithm evaluates the aircraft's current attitude and environmental conditions, such as speed, altitude, and wind direction, and generates a preliminary flight parameter adjustment plan. This plan is intended to initially adjust the aircraft to the planned landing path.

[0115] In S402, based on the preliminary flight parameter adjustment plan, adaptive feedback technology is used to refine adjustments. Adaptive feedback technology dynamically adjusts control parameters based on the aircraft's response and environmental changes, ensuring the aircraft follows the optimal path. During this step, the system continuously monitors flight status and external conditions, adjusting flight parameters in a timely manner to maintain stable and efficient flight.

[0116] In S403, the system further optimizes flight attitude using real-time data processing technology based on an advanced flight parameter adjustment scheme. During this phase, the system processes data collected from the aircraft and the environment in real time to precisely adjust flight parameters, ensuring the aircraft's stability and adaptability in complex environments.

[0117] In S404, the final parameter adjustments are made using a dynamic adjustment algorithm based on the real-time flight parameter adjustment plan. This stage is the final optimization of the entire flight control process, ensuring that the aircraft can accurately and safely adjust to the optimal attitude and position when approaching landing.

[0118] Refer to Figure 6. The steps for generating a safe landing plan based on flight parameter adjustments, machine vision, and a decision tree algorithm to perform autonomous obstacle avoidance and emergency response are as follows:

[0119] S501: Based on the flight parameter adjustment plan, use machine vision recognition technology to detect obstacles and generate an obstacle detection report;

[0120] S502: Based on the obstacle detection report, a decision tree algorithm is used to replan the path and generate a replanned flight path;

[0121] S503: Based on the re-planned flight path, obstacle avoidance strategy optimization technology is used to execute obstacle avoidance and generate an obstacle avoidance execution plan;

[0122] S504: Based on the obstacle avoidance execution plan, an emergency response mechanism is adopted to adjust the flight strategy and generate a safe landing execution plan.

[0123] In S501, based on the flight parameter adjustment plan, machine vision technology is used to detect obstacles in the flight path. Using high-precision cameras and image processing algorithms, the machine vision system analyzes the environment in real time, identifying potential obstacles such as other aircraft, ground obstacles, and weather conditions. Detected obstacle information is compiled into an obstacle detection report.

[0124] In S502, based on the obstacle detection report, the flight path is replanned using a decision tree algorithm. The decision tree algorithm selects the optimal flight path through a series of decision nodes according to different obstacle types and locations.

[0125] In S503, based on the re-planned flight path, obstacle avoidance strategy optimization technology is used to perform obstacle avoidance, which includes changing the altitude, speed or direction of the aircraft to avoid collision with obstacles.

[0126] In S504, based on the obstacle avoidance execution plan, an emergency response mechanism is activated to adjust the flight strategy and generate a safe landing execution plan.

[0127] Refer to Figure 7. Based on the safe landing execution plan, ultrasonic sensors and ground feedback systems are used to confirm and fine-tune the landing point. The specific steps to complete the final landing confirmation are as follows:

[0128] S601: Based on the safe landing execution plan, use ultrasonic detection technology to perform preliminary detection of the landing point and generate a preliminary landing point detection report;

[0129] S602: Based on the preliminary landing point detection report, use ground data analysis technology to analyze the landing point and generate an advanced landing point analysis report;

[0130] S603: Based on the advanced landing point analysis report, use precise positioning technology to fine-tune the landing point position and generate a precise landing point adjustment plan;

[0131] S604: Based on the precise landing point adjustment plan, ground adaptability analysis technology and ultrasonic fine-tuning technology are used to confirm the adaptability of the landing point and generate a final landing confirmation report.

[0132] In S601, ultrasonic detection technology is used to preliminarily detect the intended landing point. The ultrasonic sensor transmits sound waves and receives their echoes, and determines the distance and characteristics of the ground by calculating the propagation time of the sound waves.

[0133] In S602, based on the preliminary inspection report, ground data analysis technology is used to conduct a more in-depth analysis of the landing site. This includes analysis of ground firmness, slope, potential obstacles, and other environmental factors.

[0134] In S603, based on the advanced analysis report, precise positioning technology is used to fine-tune the landing point location. This step involves adjusting the aircraft's final landing trajectory and direction to ensure the optimal landing point.

[0135] In S604, based on the precise landing point adjustment plan, ground adaptability analysis technology and ultrasonic fine-tuning technology are used to confirm the adaptability of the landing point. This step ensures that the landing point is feasible under various environmental conditions and generates a final landing confirmation report.

[0136] Please refer to Figure 8, which shows a UAV precision landing control system. The UAV precision landing control system is used to execute the above-mentioned UAV precision landing control method. The system includes an environment perception module, a terrain reconstruction module, a path planning module, a flight adjustment module, an obstacle avoidance execution module, a landing point fine-tuning module, and a final confirmation module.

[0137] The environmental perception module is based on multi-sensor fusion technology, using lidar and stereo vision algorithms to collect environmental data and generate comprehensive environmental perception data;

[0138] The terrain reconstruction module reconstructs the landing area terrain based on comprehensive environmental perception data and uses a digital elevation model algorithm to generate a three-dimensional terrain model;

[0139] The path planning module uses the A* search path planning algorithm based on the 3D terrain model to perform risk assessment and path planning to generate an optimized landing path;

[0140] The flight adjustment module uses an adaptive control algorithm to adjust the flight attitude based on the optimized landing path and generate a flight parameter adjustment plan;

[0141] The obstacle avoidance execution module uses machine vision and decision tree algorithms based on the flight parameter adjustment plan to perform obstacle avoidance and emergency response and generate a safe landing execution plan;

[0142] The landing point fine-tuning module uses ultrasonic sensors and a ground feedback system based on the safe landing execution plan to fine-tune the landing point and generate a precise landing point adjustment plan.

[0143] The final confirmation module is based on a precise landing point adjustment plan, using ground adaptability analysis technology and ultrasonic fine-tuning technology to complete the final confirmation and generate a final landing confirmation report.

[0144] The Environmental Perception Module's multi-sensor fusion technology enables the system to accurately perceive its surroundings in real time, significantly reducing risks caused by environmental uncertainties. The three-dimensional terrain model generated by the Terrain Reconstruction Module enables the drone to land safely on rugged or irregular terrain, significantly improving its applicability in special environments. The A* search algorithm employed by the Path Planning Module not only improves the efficiency and accuracy of path planning but also ensures flight safety through risk assessment. The combined use of the Flight Adjustment Module and the Obstacle Avoidance Execution Module provides the drone with high flexibility and adaptability during flight, effectively reducing the risk of accidents. The Landing Point Fine-Tuning Module uses high-precision sensors to ensure the accuracy and safety of the landing process. The Final Confirmation Module ensures that all preparations and adjustments can be effectively verified and confirmed under any ground conditions, further enhancing the reliability and safety of the entire landing process.

[0145] Please refer to Figure 9, the environment perception module includes a distance perception submodule, a visual correction submodule, a data fusion submodule, and an analysis and synthesis submodule;

[0146] The terrain reconstruction module includes terrain feature analysis submodule, elevation data refinement submodule, 3D modeling submodule, and model optimization submodule;

[0147] The path planning module includes a risk assessment submodule, a preliminary path planning submodule, a path optimization submodule, and a path confirmation submodule;

[0148] The flight adjustment module includes attitude analysis submodule, feedback adjustment submodule, real-time data processing submodule, and dynamic adjustment submodule;

[0149] The obstacle avoidance execution module includes an obstacle detection submodule, a path replanning submodule, an obstacle avoidance strategy optimization submodule, and an emergency response submodule;

[0150] The landing point fine-tuning module includes a preliminary detection submodule, a ground analysis submodule, a precise positioning submodule, and a fine-tuning confirmation submodule;

[0151] The final confirmation module includes an adaptability analysis submodule, an ultrasonic adjustment submodule, a ground feedback submodule, and a final confirmation submodule.

[0152] The environmental perception module utilizes multi-sensor fusion technology. The distance perception submodule uses sensors such as lidar to detect distance and obstacles, while the visual correction submodule applies stereo vision algorithms to correct images. The data fusion submodule integrates data from various sensors to obtain comprehensive environmental information. The analysis and synthesis submodule then conducts in-depth analysis of this data to generate a comprehensive environmental perception report.

[0153] In the Terrain Reconstruction module, the Terrain Feature Analysis submodule analyzes terrain characteristics based on data provided by the Environmental Perception module, and the Elevation Data Refinement submodule obtains detailed elevation data. The 3D Modeling submodule uses this data to create a 3D model of the landing area, and the Model Optimization submodule further refines this model to ensure its accuracy and practicality.

[0154] In the path planning module, the risk assessment submodule assesses the flight path based on the 3D terrain model. The preliminary path planning submodule then plans a preliminary landing path based on this risk assessment. The path optimization submodule further optimizes this path, and the path confirmation submodule ultimately determines the optimal landing path.

[0155] In the flight adjustment module, the attitude analysis submodule analyzes the drone's current flight attitude based on the optimized landing path, and the feedback adjustment submodule adjusts flight parameters based on these analysis results. The real-time data processing submodule processes the real-time data collected during flight for use by the dynamic adjustment submodule, which dynamically adjusts the flight attitude and parameters based on this data.

[0156] Within the Obstacle Avoidance Execution Module, the Obstacle Detection submodule is responsible for detecting obstacles in the flight path, while the Path Replanning submodule replans the flight path when obstacles are detected. The Obstacle Avoidance Strategy Optimization submodule optimizes these strategies to ensure flight safety, while the Emergency Response submodule implements rapid response to avoid collisions in emergency situations.

[0157] In the landing point fine-tuning module, the preliminary detection submodule performs preliminary inspection of the selected landing point, the ground analysis submodule analyzes the ground conditions at the landing point, the precise positioning submodule accurately locates the landing point, and the fine-tuning confirmation submodule performs final fine-tuning confirmation.

[0158] In the Final Confirmation module, the Adaptability Analysis submodule analyzes the drone's compatibility with the intended landing site, and the Ultrasonic Adjustment submodule uses ultrasonic technology to make subtle adjustments. The Ground Feedback submodule makes adjustments based on feedback from ground conditions, and the Final Confirmation submodule completes the final confirmation, ensuring the drone can land safely and accurately.

[0159] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. An accurate landing control method for an unmanned aerial vehicle, characterized in that, it includes the following steps: Based on multi-sensor fusion technology, using lidar and stereo vision algorithms, collect environmental data and perform preliminary processing to generate comprehensive environmental perception data; Based on the comprehensive environmental perception data, using the digital elevation model algorithm, reconstruct the three-dimensional terrain of the landing area to generate a three-dimensional terrain model; Based on the three-dimensional terrain model, using the A* search path planning algorithm, conduct risk assessment and plan a safe landing path to generate an optimized landing path; Based on the optimized landing path, using an adaptive control algorithm, through a fuzzy logic controller, adjust the flight attitude in real time to generate flight parameter adjustments; Based on the flight parameter adjustments, using machine vision and decision tree algorithms, perform autonomous obstacle avoidance and emergency response to generate a safe landing execution plan; Based on the safe landing execution plan, using ultrasonic sensors and a ground feedback system, confirm and fine-tune the landing point to complete the final landing confirmation; The comprehensive environmental perception data specifically refers to multi-dimensional data including terrain features, obstacle information, and meteorological conditions. The three-dimensional terrain model specifically refers to a three-dimensional map including terrain height, slope, and stability. The optimized landing path specifically refers to the best landing trajectory optimized according to environmental risks and the capabilities of the unmanned aerial vehicle. The flight parameter adjustments specifically refer to flight parameter adjustments for real-time environmental changes, including speed, angle, and altitude. The safe landing execution plan specifically refers to automatic adjustment and emergency landing strategies in the face of sudden obstacles.

2. The accurate landing control method for an unmanned aerial vehicle according to claim 1, characterized in that, the step of collecting environmental data and performing preliminary processing to generate comprehensive environmental perception data based on multi-sensor fusion technology, using lidar and stereo vision algorithms is specifically as follows: Based on multi-sensor fusion technology, using the lidar algorithm, collect environmental distance information to generate preliminary distance perception data; Based on the preliminary distance perception data, using the binocular stereo matching algorithm, correct the image data to generate corrected visual perception data; Based on the corrected visual perception data and the preliminary distance perception data, using the Kalman filtering technology, integrate the data to generate optimized environmental perception data; Based on the optimized environmental perception data, using a data analysis algorithm, conduct comprehensive analysis to generate comprehensive environmental perception data.

3. The accurate landing control method for an unmanned aerial vehicle according to claim 1, characterized in that, the step of reconstructing the three-dimensional terrain of the landing area to generate a three-dimensional terrain model based on the comprehensive environmental perception data, using the digital elevation model algorithm is specifically as follows: Based on the comprehensive environmental perception data, using a terrain analysis algorithm, determine terrain features to generate preliminary terrain feature data; Based on the preliminary terrain feature data, using the digital elevation model algorithm, measure the terrain height and shape to generate refined elevation data; Based on the refined elevation data, using three-dimensional modeling technology, reconstruct the three-dimensional terrain to generate a preliminary three-dimensional terrain model; Based on the preliminary three-dimensional terrain model, using a model optimization algorithm, refine and optimize the model to generate the final three-dimensional terrain model.

4. The method for precise landing control of an unmanned aerial vehicle according to claim 1, characterized in that based on the three-dimensional terrain model, the A* search path planning algorithm is used to perform risk assessment and plan a safe landing path. The steps of generating an optimized landing path are specifically as follows: Based on the three-dimensional terrain model, a risk analysis algorithm is used to perform risk assessment and generate a risk assessment report; Based on the risk assessment report, the A* search algorithm is used to plan a preliminary path and generate a preliminary landing path; Based on the preliminary landing path, path optimization technology is used to generate an optimized landing path with reference to efficiency and safety factors; Based on the optimized landing path, final confirmation and adjustment technology is used to ensure path optimization and generate a final optimized landing path.

5. The method for precise landing control of an unmanned aerial vehicle according to claim 1, characterized in that based on the optimized landing path, an adaptive control algorithm is used, and through a fuzzy logic controller, the flight attitude is adjusted in real time. The steps of generating flight parameter adjustment are specifically as follows: Based on the optimized landing path, a fuzzy logic control algorithm is used to analyze the flight attitude and generate a preliminary flight parameter adjustment plan; Based on the preliminary flight parameter adjustment plan, adaptive feedback technology is used to refine the adjustment and generate an advanced flight parameter adjustment plan; Based on the advanced flight parameter adjustment plan, real-time data processing technology is used to optimize the flight attitude and generate a real-time flight parameter adjustment plan; Based on the real-time flight parameter adjustment plan, a dynamic adjustment algorithm is used to complete the final adjustment and generate a final flight parameter adjustment plan.

6. The method for precise landing control of an unmanned aerial vehicle according to claim 1, characterized in that based on the flight parameter adjustment, machine vision and decision tree algorithms are used to perform autonomous obstacle avoidance and emergency response. The steps of generating a safe landing execution plan are specifically as follows: Based on the flight parameter adjustment plan, machine vision recognition technology is used to detect obstacles and generate an obstacle detection report; Based on the obstacle detection report, a decision tree algorithm is used to re-plan the path and generate a re-planned flight path; Based on the re-planned flight path, obstacle avoidance strategy optimization technology is used to perform obstacle avoidance and generate an obstacle avoidance execution plan; Based on the obstacle avoidance execution plan, an emergency response mechanism is used to adjust the flight strategy and generate a safe landing execution plan.

7. The method for precise landing control of an unmanned aerial vehicle according to claim 1, characterized in that based on the safe landing execution plan, an ultrasonic sensor and a ground feedback system are used to confirm and fine-tune the landing point. The steps of completing the final landing confirmation are specifically as follows: Based on the safe landing execution plan, ultrasonic detection technology is used to preliminarily detect the landing point and generate a preliminary landing point detection report; Based on the preliminary landing point detection report, ground data analysis technology is used to analyze the landing point and generate an advanced landing point analysis report; Based on the advanced landing point analysis report, precise positioning technology is used to fine-tune the position of the landing point and generate a precise landing point adjustment plan; Based on the precise landing point adjustment plan, ground adaptability analysis technology and ultrasonic fine adjustment technology are used to confirm the adaptability of the landing point and generate a final landing confirmation report.

8. An unmanned aerial vehicle (UAV) precise landing control system, characterized in that According to the UAV precise landing control method described in any one of claims 1-7, the system includes an environmental perception module, a terrain reconstruction module, a path planning module, a flight adjustment module, an obstacle avoidance execution module, a landing point fine-tuning module, and a final confirmation module.

9. According to the UAV precise landing control system described in claim 8, characterized in that The environmental perception module is based on multi-sensor fusion technology and uses lidar and stereo vision algorithms to collect environmental data and generate comprehensive environmental perception data; The terrain reconstruction module is based on the comprehensive environmental perception data and uses a digital elevation model algorithm to reconstruct the terrain of the landing area and generate a three-dimensional terrain model; The path planning module is based on the three-dimensional terrain model and uses the A* search path planning algorithm to perform risk assessment and path planning to generate an optimized landing path; The flight adjustment module is based on the optimized landing path and uses an adaptive control algorithm to adjust the flight attitude and generate a flight parameter adjustment plan; The obstacle avoidance execution module is based on the flight parameter adjustment plan and uses machine vision and decision tree algorithms to perform obstacle avoidance and emergency response and generate a safe landing execution plan; The landing point fine-tuning module is based on the safe landing execution plan and uses ultrasonic sensors and a ground feedback system to fine-tune the landing point and generate a precise landing point adjustment plan; The final confirmation module is based on the precise landing point adjustment plan and uses ground adaptability analysis technology and ultrasonic fine adjustment technology to complete the final confirmation and generate a final landing confirmation report.

10. According to the UAV precise landing control system described in claim 8, characterized in that The environmental perception module includes a distance perception sub-module, a vision correction sub-module, a data fusion sub-module, and an analysis and synthesis sub-module; The terrain reconstruction module includes a terrain feature analysis sub-module, an elevation data refinement sub-module, a three-dimensional modeling sub-module, and a model optimization sub-module; The path planning module includes a risk assessment sub-module, a preliminary path planning sub-module, a path optimization sub-module, and a path confirmation sub-module; The flight adjustment module includes an attitude analysis sub-module, a feedback adjustment sub-module, a real-time data processing sub-module, and a dynamic adjustment sub-module; The obstacle avoidance execution module includes an obstacle detection sub-module, a path re-planning sub-module, an obstacle avoidance strategy optimization sub-module, and an emergency response sub-module; The landing point fine-tuning module includes a preliminary detection sub-module, a ground analysis sub-module, a precise positioning sub-module, and a fine-tuning confirmation sub-module; The final confirmation module includes an adaptability analysis sub-module, an ultrasonic adjustment sub-module, a ground feedback sub-module, and a final confirmation sub-module.

Images