Six-rotor unmanned aerial vehicle accurate flight system with positioning and navigation functions
By employing a multi-source positioning and sensing module, dynamic weight fusion, and adaptive trajectory control, the positioning accuracy and flight stability of the hexarotor UAV in complex environments have been improved, solving the positioning and trajectory control problems of traditional UAVs in complex environments and enabling high-precision operations.
Patent Information
- Application Number
- CN202511508205.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-22
- Publication Date
- 2026-01-09
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional hexacopter drones suffer from insufficient positioning accuracy in complex environments, unstable trajectory control, and multi-sensor data fusion that is not adapted to scene changes, making it difficult to meet the requirements of high-precision operations.
Employing a multi-source positioning and sensing module, a dynamic weight fusion module, and an adaptive trajectory control module, it integrates a dual-antenna RTK-GNSS, a triple-redundant IMU, a lidar SLAM, a visual anchor point recognition, and a pipeline electromagnetic detection unit. By combining signal preprocessing and scene prediction, it dynamically adjusts PID parameters to achieve high-precision positioning and stable flight.
Achieving positioning accuracy within ±5cm and displacement fluctuation within ±0.5m in complex environments ensures stable flight of drones in multiple scenarios and meets the requirements of high-precision operations.
Smart Images

Figure CN121300402A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of flight systems, in particular to a six-rotor unmanned aerial vehicle precise flight system with positioning and navigation. BACKGROUND
[0002] In the current rapid development of unmanned aerial vehicle technology, its application in the fields of electric power inspection, underground pipe gallery detection, outdoor surveying and mapping, etc. is becoming more and more widespread, but the positioning and navigation system of the traditional six-rotor unmanned aerial vehicle is difficult to adapt to complex operation scenarios, becoming the core bottleneck restricting its precise operation. The current mainstream unmanned aerial vehicle relies on a single GNSS positioning module, which can meet the basic positioning needs in an open scene, but when entering a GNSS signal blocked environment such as a tunnel or an underground pipe gallery, the positioning accuracy will decrease sharply, and even positioning failure may occur, resulting in the unmanned aerial vehicle unable to fly along the preset trajectory, increasing the risk of collision; at the same time, when operating near high-voltage transmission lines, strong electromagnetic interference will further interfere with the positioning signal, making the positioning deviation expand to tens of centimeters, which is difficult to meet the high-precision operation requirements of metal pipeline inspection, etc.
[0003] The trajectory control system of the traditional unmanned aerial vehicle also has obvious limitations, most of which use fixed parameter PID control architecture, which cannot dynamically adjust the control strategy according to environmental changes. When encountering strong wind interference of ≥10m / s, the fixed PID parameters are difficult to quickly respond to changes in wind speed, resulting in unmanned aerial vehicle displacement fluctuations exceeding ±1m, which not only affects the operation efficiency, but also may cause safety accidents due to trajectory deviation; in addition, the existing system lacks the ability of trajectory pre-planning and real-time correction for complex scenarios, cannot import a three-dimensional model of the operation area to generate an adaptive trajectory, and when the actual position deviates from the preset trajectory, the correction response time is usually more than 0.5 seconds, which is significantly lagging, and it is difficult to adapt to scenes such as underground pipe galleries and narrow tunnels that require strict trajectory accuracy.
[0004] In terms of multi-sensor data fusion, the traditional unmanned aerial vehicle mostly uses a static weight distribution method, which always fuses data in a fixed proportion regardless of sensor accuracy and scene changes. For example, when the GNSS signal is weakened, it is still used as the main positioning data source, resulting in a significant reduction in the fusion positioning accuracy; at the same time, the system lacks scene prediction ability and cannot sense the operation scenario switch in advance, and the sensor function activation has a delay, further exacerbating the risk of positioning interruption. These problems together result in insufficient precise flight and operation capability of the traditional six-rotor unmanned aerial vehicle in complex environments, and an integrated system with multi-source positioning, dynamic fusion and adaptive control is urgently needed to break through the scene adaptation and precision bottleneck. SUMMARY
[0005] Technical problems solved In view of the deficiencies of the prior art, the present application provides a six-rotor unmanned aerial vehicle precise flight system with positioning and navigation, which solves the problems raised in the background art.
[0006] (II) Technical Solution The present application specifically adopts the following technical solutions to achieve the above-mentioned purposes: The six-rotor unmanned aerial vehicle precise flight system with positioning and navigation function comprises a multi-source positioning and sensing module, a dynamic weight fusion module and a self-adaptive trajectory control module. The dynamic weight fusion module is internally provided with a signal confidence evaluation model and a scene prediction sub-module. The self-adaptive trajectory control module adopts a cascade PID + fuzzy adaptive control architecture, with an inner loop of angular velocity loop (bandwidth 50-80Hz) and an outer loop of position loop (bandwidth 20-30Hz).
[0007] Further, the pipeline electromagnetic detection unit has a detection distance of 0.1-2m and a positioning accuracy of ±15cm, and is provided with an anti-electromagnetic interference sub-module.
[0008] Further, the self-adaptive trajectory control module is further internally provided with a trajectory pre-planning and real-time correction sub-module.
[0009] Further, the multi-source positioning perception module can also be equipped with a wheel speed odometer unit, which can assist in calculating displacement when the UAV is flying close to the ground on flat ground (such as the bottom of a concrete tunnel), with a resolution of 0.1 mm and a displacement error of ≤0.5% when traveling 100 m. Working with the IMU unit can further reduce the positioning drift in GNSS long-term rejection scenarios.
[0010] Further, the weight distribution logic of the dynamic weight fusion module can be dynamically adjusted according to the work scene. In an open scene (GNSS signal normal), the RTK-GNSS unit is assigned a weight of 0.7, the IMU unit is assigned a weight of 0.2, and the total weight of other units is 0.1. In a GNSS long-term rejection scenario (lock loss > 5 minutes), the laser radar SLAM unit is assigned a weight of 0.6, the visual anchor recognition unit is assigned a weight of 0.2, the pipeline electromagnetic detection unit is assigned a weight of 0.1, and the IMU unit is assigned a weight of 0.1, ensuring that the fusion positioning accuracy in each scene is not less than ±5 cm.
[0011] (Three) beneficial effects Compared with the prior art, the present application provides a six-rotor UAV precise flight system with positioning and navigation, which has the following beneficial effects: The system greatly improves the positioning accuracy in all scenes through multi-module cooperative linkage, effectively solves the positioning limitations of traditional UAVs, and integrates multiple units such as dual-antenna RTK-GNSS and laser radar SLAM in the multi-source positioning perception module, covering air-ground-underground positioning needs, and ensuring data quality with the preprocessing sub-module; the dynamic weight fusion module dynamically assigns weights based on confidence evaluation and scene prediction, such as laser radar SLAM (weight 0.6) in GNSS rejection scenarios, combined with visual anchor recognition, achieving positioning accuracy within ±5 cm. Near high-voltage lines, the anti-interference sub-module of the pipeline electromagnetic detection unit controls the positioning deviation within ±6 cm, meeting the high-precision operation needs of high-voltage inspection and underground pipe gallery detection.
[0012] At the same time, the system significantly enhances flight trajectory stability and scene adaptability, ensuring safe operation in complex environments. The adaptive trajectory control module uses a cascade PID + fuzzy adaptive architecture, automatically adjusts parameters when wind speed > 10 m / s, and displacement fluctuation ≤±0.5 m; the trajectory pre-planning and real-time correction sub-module imports a CAD three-dimensional model to generate an adaptive trajectory, and triggers correction within 0.1 seconds when the deviation > 5 cm. The dynamic weight fusion module predicts the scene 0.5-1 seconds in advance and pre-activates the corresponding positioning unit, such as starting laser radar SLAM before entering a tunnel, avoiding sensor switching delay, so that the UAV can stably operate in multiple scenes such as open ground and tunnels, breaking through the single scene adaptation problem of traditional UAVs. BRIEF DESCRIPTION OF DRAWINGS
[0013] Fig. 1 This is a schematic diagram of the precision flight system of the six-rotor unmanned aerial vehicle of the present invention; Fig. 2 This is a schematic diagram of the multi-source positioning and sensing module system of the present invention; Fig. 3 This is a schematic diagram of the adaptive trajectory control module system of the present invention. Detailed Implementation
[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0015] Example like Figs. 1-3 As shown, an embodiment of the present invention proposes a precision flight system for a six-rotor unmanned aerial vehicle with positioning and navigation, including a multi-source positioning and sensing module, a dynamic weight fusion module, and an adaptive trajectory control module. The multi-source positioning and sensing module integrates a dual-antenna RTK-GNSS unit, a triple-redundant IMU unit, a lidar SLAM unit, a visual anchor point recognition unit, and a pipeline electromagnetic detection unit. It can simultaneously collect multi-dimensional positioning signals from air, ground, and underground. Each unit is equipped with a signal preprocessing submodule, which can reduce noise and remove outliers from the original signal. The pipeline electromagnetic detection unit has a detection range of 0.1-2m and a positioning accuracy of ±15cm. It is also equipped with an anti-electromagnetic interference submodule. When working near high-voltage transmission lines (±500kV), the positioning deviation caused by electromagnetic interference can be controlled within ±6cm, which meets the positioning requirements of underground utility tunnels and tunnels for metal pipeline inspection. The multi-source positioning and sensing module can also be equipped with a wheel speed odometer unit. When the UAV flies close to the ground on flat ground (such as the bottom of a concrete tunnel), the wheel speed odometer unit can assist in calculating displacement. Its resolution is 0.1mm, and the displacement error when traveling 100m is ≤0.5%. Working in conjunction with the IMU unit can further reduce positioning drift in long-term GNSS denial scenarios. The dynamic weight fusion module incorporates a signal confidence assessment model and a scene prediction submodule. The signal confidence assessment model can calculate the confidence value of each positioning unit in real time, with a value range of 0-100%. When the GNSS signal strength is ≥-120dBm, the confidence value is assigned to 90%, and when the point cloud matching rate of the lidar SLAM unit is <60%, the confidence value is assigned to 40%. The scene prediction submodule analyzes the UAV flight trajectory and the trend of GNSS signal strength changes, and can predict the upcoming operation scene 0.5-1 seconds in advance, and trigger the pre-activation command of the corresponding positioning unit. The weight distribution logic of the dynamic weight fusion module can be dynamically adjusted according to the job scene. In an open scene (GNSS signal is normal), the RTK-GNSS unit weight is assigned 0.7, the IMU unit weight is assigned 0.2, and the weights of other units are assigned 0.1 in total. In a GNSS long-term rejection scene (loss of lock > 5 minutes), the laser radar SLAM unit weight is assigned 0.6, the visual anchor point recognition unit weight is assigned 0.2, the pipeline electromagnetic detection unit weight is assigned 0.1, and the IMU unit weight is assigned 0.1, so as to ensure that the fusion positioning accuracy in each scene is not less than ±5 cm. The adaptive trajectory control module adopts a cascade PID + fuzzy adaptive control architecture. The inner loop is an angular velocity loop (bandwidth 50-80Hz), and the outer loop is a position loop (bandwidth 20-30Hz). The fuzzy adaptive control submodule can dynamically adjust the PID parameters according to the environmental wind speed (0-20m / s). When the wind speed is >10m / s, the damping coefficient is automatically increased, so that the displacement fluctuation of the unmanned aerial vehicle under this wind speed is controlled within ±0.5m. The adaptive trajectory control module also has a built-in trajectory pre-planning and real-time correction submodule. It can import an external CAD format three-dimensional model of the job scene (such as a tunnel or pipe gallery model), automatically generate a pre-planned trajectory along the axis of the scene, and based on the real-time data of the multi-source positioning and perception module, when the actual position of the unmanned aerial vehicle deviates from the pre-planned trajectory by >5cm, trigger an automatic correction instruction, and the correction response time is ≤0.1s. The three modules form a multi-dimensional and full-process linkage relationship around the technical logic of "high-precision positioning data acquisition-high-quality data fusion-high stability trajectory control", which is specifically reflected in three aspects of data flow, parameter coordination and scene adaptation. From the perspective of data flow, the multi-source positioning and sensing module serves as the data collection source, integrating a dual-antenna RTK-GNSS unit, a three-redundancy IMU unit, a laser radar SLAM unit, a visual anchor point recognition unit, a pipeline electromagnetic detection unit, and a wheel speed odometer unit according to demand. It can synchronously collect multi-dimensional positioning signals in air-ground-underground, and each unit is configured with a signal preprocessing submodule to complete noise reduction and outlier elimination on the original signal. After preprocessing, the "preprocessed positioning data + real-time working state information of each unit" is transmitted to the dynamic weight fusion module. After receiving the data, the dynamic weight fusion module calculates the confidence value of each positioning unit (value range 0-100%, such as GNSS signal strength ≥-120dBm, confidence value 90%, laser radar SLAM unit point cloud matching rate <60%, confidence value 40%) through the built-in signal confidence evaluation model, and dynamically allocates the weight of each positioning unit according to the judgment result of the scene prediction submodule (by analyzing the flight trajectory and GNSS signal strength trend, predicting the work scene 0.5-1 seconds in advance), generating "fusion positioning data + scene type instruction" and transferring to the adaptive trajectory control module. Based on the fusion positioning data and scene instruction, the adaptive trajectory control module generates trajectory control signals to drive the UAV flight, and at the same time feeds back the "deviation data of actual flight position and pre-planned trajectory" to the dynamic weight fusion module to assist it in further optimizing the weight allocation logic, forming a data closed-loop circulation.
[0016] From the perspective of parameter coordination, the multi-source positioning perception module transmits parameters to the dynamic weight fusion module, which not only includes positioning coordinate data, but also includes key working parameters of each unit, such as GNSS signal strength, laser radar point cloud matching rate, IMU redundant data consistency, etc. These parameters are directly used as input basis for signal confidence evaluation model; the dynamic weight fusion module outputs dynamic weight coefficients according to these parameters, for example, in an open scene (GNSS signal is normal), the RTK-GNSS unit weight is assigned 0.7, the IMU unit weight is assigned 0.2, and the weight of other units is assigned 0.1. In the GNSS long-term rejection scene (lock loss > 5 minutes), the laser radar SLAM unit weight is assigned 0.6, the visual anchor point recognition unit weight is assigned 0.2, the pipeline electromagnetic detection unit weight is assigned 0.1, and the IMU unit weight is assigned 0.1. At the same time, the positioning accuracy level information corresponding to the weight coefficient is synchronized to the adaptive trajectory control module; after receiving this information, the adaptive trajectory control module adjusts the control parameters in combination with the environmental parameters (such as wind speed), such as using a cascade PID + fuzzy adaptive control architecture, the inner loop angular velocity loop (bandwidth 50-80Hz), the outer loop position loop (bandwidth 20-30Hz), when the wind speed > 10m / s, the damping coefficient is automatically increased, and the trajectory correction response time (≤0.1s) and other control effect parameters are fed back to the multi-source positioning perception module to assist it in adjusting the signal sampling frequency, ensuring parameter coordination adaptation.
[0017] From the perspective of scene adaptation, when the unmanned aerial vehicle is about to enter a special operation scene, the scene prediction sub-module of the dynamic weight fusion module captures the scene change signal first, such as predicting the entry into the GNSS rejection scene when the GNSS signal strength continuously weakens, and then triggering the pre-activation instruction of the corresponding positioning unit, activating the laser radar SLAM unit and the visual anchor point recognition unit in advance to avoid sensor switching delay; at the same time, the pipeline electromagnetic detection unit in the multi-source positioning perception module automatically starts the anti-electromagnetic interference sub-module when approaching the high-voltage transmission line (±500kV) to adapt to the electromagnetic environment near the high-voltage line; the adaptive trajectory control module imports the corresponding external CAD format three-dimensional model (such as tunnel, pipe gallery model) according to the scene instruction, automatically generates a pre-planned trajectory along the axis of the scene, and realizes precise adaptation of the scene and module function.
[0018] (1) Improve the positioning accuracy in all scenes The multi-unit design of the multi-source positioning perception module realizes multi-dimensional coverage of positioning signals. The dual-antenna RTK-GNSS unit ensures high precision in open scenarios. The three-redundant IMU units solve the short-term signal positioning needs. The laser radar SLAM unit is suitable for indoor and tunnel closed scenarios. The pipeline electromagnetic detection unit meets the underground pipeline positioning needs. The selected wheel speed odometer unit (resolution 0.1mm, displacement error ≤0.5% for 100m travel) assists in flat ground flight positioning. After preprocessing, the data of each unit provides a high-quality data basis for the dynamic weight fusion module. The dynamic weight fusion module selects high-reliability data for fusion through confidence evaluation and dynamic weight allocation, avoiding positioning deviation caused by single sensor failure or insufficient precision. For example, in the GNSS long-term denial scenario, relying on the high-weight cooperation of the laser radar SLAM and visual anchor point recognition unit ensures that the fusion positioning accuracy is not less than ±5cm. At the same time, when the pipeline electromagnetic detection unit is operating near high-voltage lines, the positioning deviation can be controlled within ±6cm, thereby improving the positioning accuracy in different scenarios.
[0019] (II) Ensure flight trajectory stability The cascade PID architecture of the adaptive trajectory control module quickly responds to positioning deviation through the cooperation of the inner loop angular velocity ring and the outer loop position ring. The fuzzy adaptive control submodule dynamically adjusts the PID parameters based on the environmental data (such as wind speed 0-20m / s) transmitted by the dynamic weight fusion module. When the wind speed is >10m / s, the displacement fluctuation is controlled within ±0.5m by increasing the damping coefficient, effectively resisting external wind interference. At the same time, the trajectory pre-planning and real-time correction submodule built-in the adaptive trajectory control module is based on the real-time data of the multi-source positioning perception module. When the actual position of the UAV deviates from the pre-planned trajectory by >5cm, it can trigger automatic correction instructions within ≤0.1 seconds, ensuring that the UAV's trajectory always adheres to the pre-planned path when flying in narrow spaces (such as underground pipe galleries), avoiding collision risks and ensuring flight stability and safety.
[0020] (III) Enhance scene adaptation capability The scene prediction submodule of the dynamic weight fusion module predicts scene changes 0.5-1 seconds in advance, allowing each module to reserve time for function switching and parameter adjustment, avoiding system response lag caused by sudden scene changes. Each unit of the multi-source positioning perception module can be flexibly started according to scene needs, such as starting the laser radar SLAM and pipeline electromagnetic detection unit in the underground pipe gallery scenario and starting the anti-electromagnetic interference submodule near high-voltage lines. The adaptive trajectory control module imports scene-specific three-dimensional models and adjusts trajectory correction logic to adapt to the spatial characteristics of different work scenarios, enabling the UAV to operate stably in multiple scenarios such as open spaces, underground pipe galleries, tunnels, and near high-voltage lines, breaking the limitations of traditional single-scenario adaptation.
[0021] It should be pointed out finally that the above only describes the preferred embodiments of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent replacements to some technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims
1. A six-rotor unmanned aerial vehicle precision flight system with positioning and navigation, comprising a multi-source positioning perception module, a dynamic weight fusion module and an adaptive trajectory control module, characterized in that: The multi-source positioning perception module integrates a dual-antenna RTK-GNSS unit, a three-redundancy IMU unit, a laser radar SLAM unit, a visual anchor point recognition unit, and a pipeline electromagnetic detection unit, can simultaneously collect air-ground-underground multi-dimensional positioning signals, and each unit is configured with a signal preprocessing submodule, which can denoise and remove outliers from the original signal. The dynamic weight fusion module has a signal confidence evaluation model and a scene prediction submodule. The signal confidence evaluation model can calculate the confidence value of each positioning unit in real time, with a value range of 0-100%. When the GNSS signal strength is ≥-120 dBm, the confidence value is 90%. When the laser radar SLAM unit point cloud matching rate is <60%, the confidence value is 40%. The scene prediction submodule can predict the upcoming work scene 0.5-1 seconds in advance by analyzing the unmanned aerial vehicle flight trajectory and GNSS signal strength trend, and trigger the pre-activation instruction of the corresponding positioning unit. The adaptive trajectory control module adopts a cascade PID + fuzzy adaptive control architecture, with an inner loop of angular velocity and an outer loop of position. The fuzzy adaptive control submodule can dynamically adjust the PID parameters according to the environmental wind speed. When the wind speed is >10 m / s, the damping coefficient is automatically increased to control the displacement fluctuation of the unmanned aerial vehicle within ±0.5 m under this wind speed.
2. The precision flight system of hexa-copter with positioning and navigation according to claim 1, wherein: The pipeline electromagnetic detection unit has a detection distance of 0.1-2 m and a positioning accuracy of ±15 cm. It is configured with an anti-electromagnetic interference submodule. When working near high-voltage transmission lines, the positioning deviation caused by electromagnetic interference can be controlled within ±6 cm, meeting the positioning requirements of underground pipe gallery and tunnel metal pipeline inspection scenes.
3. The precision flight system of hexa-copter with positioning and navigation according to claim 1, wherein: The adaptive trajectory control module also has a trajectory pre-planning and real-time correction submodule, which can import an external CAD format three-dimensional model of the work scene, automatically generate a pre-planned trajectory along the scene axis, and trigger an automatic correction instruction when the actual position of the unmanned aerial vehicle deviates from the pre-planned trajectory by >5 cm, with a correction response time ≤0.1 s.
4. The precision flight system of hexa-copter with positioning and navigation according to claim 1, wherein: The multi-source positioning perception module can also be equipped with a wheel speed odometer unit, which can assist in calculating displacement when the unmanned aerial vehicle is flying close to the ground on flat ground. Its resolution is 0.1 mm, and the displacement error is ≤0.5% when traveling 100 m. Working with the IMU unit can further reduce the positioning drift in GNSS long-term rejection scenarios.
5. The precision flight system of hexa-copter with positioning and navigation according to claim 1, wherein: The weight distribution logic of the dynamic weight fusion module can be dynamically adjusted according to the work scene. In an open scene, the RTK-GNSS unit weight is assigned 0.7, the IMU unit weight is assigned 0.2, and the weights of other units are assigned a total of 0.
1. In a GNSS long-term rejection scenario, the laser radar SLAM unit weight is assigned 0.6, the visual anchor point recognition unit weight is assigned 0.2, the pipeline electromagnetic detection unit weight is assigned 0.1, and the IMU unit weight is assigned 0.1, ensuring that the fusion positioning accuracy in each scene is not less than ±5 cm.
Citation Information
Cited By
Power grid operation unmanned aerial vehicle navigation system based on RTK and laser radar cooperative positioning
CN121612309A
