Unmanned aerial vehicle carrying system for complex terrain surveying and mapping

Abstract

The invention provides an unmanned aerial vehicle carrying system for complex terrain surveying and mapping, and relates to the technical field of unmanned aerial vehicle surveying and mapping. The unmanned aerial vehicle carrying system for complex terrain surveying and mapping comprises a sensor unit, the sensor unit comprises an integrated laser radar, a multispectral camera, a double-antenna RTK. PPK positioning module and an intelligent control unit, and the intelligent control unit comprises an environment sensing processor, a dynamic path planning module, a multi-source data fusion engine and a power system. The power system comprises an oil-electricity hybrid module, the oil-electricity hybrid module is composed of a lithium battery and a magnetic suction type fuel oil range extender, the magnetic suction type fuel oil range extender is mounted on the belly through a magnetic suction connector, and the connector comprises an axial positioning pin and a rare earth magnetic ring. Oil-electricity seamless switching is achieved through the magnetic attraction quick-release range extender, the endurance is greatly improved, the multi-degree-of-freedom posture adjusting platform is matched with the millimeter-level sensor for hard synchronization, and vibration distortion is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) surveying and mapping, and in particular to an UAV-mounted system for surveying and mapping complex terrain. Background Art

[0002] Topographic surveying is crucial for obtaining information about the Earth's surface terrain and is indispensable in engineering construction, land planning and other fields. Traditional topographic surveying methods are inefficient, costly, and easily restricted by complex terrain. In recent years, drone technology has developed rapidly. Drone mapping has become a key means of topographic surveying due to its high efficiency, flexibility, and high precision. Existing drone mapping technology mainly solves the mapping problems of steep slopes, dense forests, canyons and other scenes through innovations such as vertical take-off and landing capabilities, high-precision sensor integration, intelligent obstacle avoidance and collaborative operations.

[0003] At present, when drones in existing technologies are used for surveying and mapping in complex terrains such as mountains, forests, urban areas and other areas with relatively complex terrain, drone surveying and mapping faces interference from many problems such as navigation positioning, environmental interference, data processing and safety supervision, which in turn leads to low surveying and mapping efficiency and untimely transmission of surveying and mapping data. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides a UAV carrying system for complex terrain mapping, which solves the problems of low efficiency of complex terrain mapping by UAVs and untimely transmission of mapping information.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a UAV-mounted system for complex terrain mapping, comprising a sensor unit, the sensor unit including an integrated laser radar, a multispectral camera, and a dual-antenna RTK·PPK positioning module, an intelligent control unit, the intelligent control unit including an environmental perception processor, a dynamic path planning module, and a multi-source data fusion engine, a power system, the power system including a hybrid module, the hybrid module consisting of a lithium battery and a magnetic fuel range extender, a multi-degree-of-freedom attitude adjustment platform, the multi-degree-of-freedom attitude adjustment platform being connected to the body via a ball joint mechanism, a cluster communication relay, the cluster communication relay supporting MESH networking, and a ground station, the ground station being used to provide satellite data and mission parameters.

[0006] Preferably, the sensor unit outputs point cloud, spectrum and positioning data to the intelligent control unit, and the laser radar of the sensor unit and the multispectral camera are hard synchronized through the 1PPS pulse of RTK, with a time error of ≤1 microsecond.

[0007] Preferably, the intelligent control unit sends a course correction instruction to the power system and receives flight status feedback. The environment perception processor of the intelligent control unit deploys the LOGA·Net algorithm, and the terrain analysis delay is less than 50 milliseconds.

[0008] Preferably, the intelligent control unit sends an attitude calibration instruction to the attitude adjustment platform and receives pan-tilt angle feedback. The ball joint mechanism of the attitude adjustment platform contains an angle encoder to achieve attitude closed-loop control with a compensation accuracy of ≤0.1 degrees.

[0009] Preferably, the cluster communication relay communicates with the intelligent control unit via a MESH network, and the MESH network of the cluster communication relay supports a network of ≥3 units, with a communication radius of ≥10 kilometers.

[0010] Preferably, the multi-source data fusion engine outputs a soil data model and a geological hazard risk assessment map.

[0011] Preferably, the ground station and the intelligent control unit bidirectionally transmit satellite data, mission parameters and fusion modeling results.

[0012] Preferably, the magnetic interface of the magnetic fuel range extender includes an axial positioning pin and a rare earth magnetic ring, the axial positioning pin limits radial displacement, and the rare earth magnetic ring provides a vertical adsorption force of ≥800N.

[0013] Preferably, the multi-source data fusion engine fuses the slope curvature characteristics of the drone point cloud, the vegetation cover index of the satellite image and the ground sensor data through the XGBOOST algorithm to generate a soil moisture thermal map and a landslide risk probability model.

[0014] A method for surveying and mapping complex terrain using an unmanned aerial vehicle (UAV) system includes the following steps:

[0015] Step 1: The sensor unit collects terrain point cloud and surface spectrum;

[0016] Step 2: The intelligent control unit generates attitude adjustment instructions based on the point cloud distortion rate and generates an obstacle avoidance route based on the terrain curvature;

[0017] Step 3: The power system switches the power supply mode according to the voltage threshold;

[0018] Step 4: The ground station and the intelligent control unit collaborate to output a three-dimensional real-scene model.

[0019] The present invention provides a UAV-mounted system for complex terrain mapping. It has the following beneficial effects:

[0020] 1. This invention achieves seamless switching between oil and electricity through a magnetic quick-release range extender, significantly improving endurance. The multi-degree-of-freedom attitude adjustment platform is hard-synchronized with millimeter-level sensors to improve vibration distortion. Dual-antenna RTK / PPK and MESH multi-machine networking ensures full terrain coverage in blind spots such as canyons and dense forests, greatly improving surveying and mapping efficiency.

[0021] 2. This invention uses the LOGA·Net algorithm to generate obstacle avoidance routes in real time, improving the pass rate in complex terrain. The multi-source fusion engine integrates satellite, drone, and ground data to output soil thermal maps and risk assessment models, achieving high disaster identification accuracy. The closed loop from data collection to decision-making provides rapid response capabilities for disaster prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Schematic diagram of the control module of the present invention;

[0023] Figure 2 Schematic diagram of the decision-making process of the present invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0025] Example:

[0026] like Figure 1 and Figure 2 As shown, an embodiment of the present invention provides an unmanned aerial vehicle (UAV) carrying system for complex terrain mapping, including a sensor unit, which includes an integrated laser radar, a multispectral camera, and a dual-antenna RTK·PPK positioning module. The intelligent control unit includes an environmental perception processor, a dynamic path planning module, and a multi-source data fusion engine. The power system includes a hybrid module, which consists of a lithium battery and a magnetic fuel range extender. The magnetic fuel range extender is mounted on the belly of the aircraft through a magnetic interface, which contains an axial positioning pin and a rare earth magnetic ring. The magnetic fuel range extender has a built-in micro-turbine engine with a power density of 3.5kW / kg. Status data is transmitted to the power system via a CAN bus. The six-degree-of-freedom attitude adjustment platform is connected to the aircraft body through a ball joint mechanism. The cluster communication relay supports MESH networking. The ground station is used to provide satellite data and mission parameters.

[0027] The sensor unit outputs point cloud, spectral, and positioning data to the intelligent control unit. The multispectral and thermal imaging sensors integrate wide-angle, zoom, thermal imaging, and laser ranging modules to achieve all-weather monitoring. Thermal imaging can also be used to identify forest fire sources or monitor farmland encroachment at night. The sensor unit's lidar and multispectral camera are hard-synchronized via RTK's 1PPS pulse, with a time error of less than 1 microsecond. The intelligent control unit sends course correction commands to the power system and receives flight status feedback. The intelligent control unit's environmental perception processor deploys the LOGA Net algorithm, with a terrain resolution delay rate of less than 50 microseconds. The intelligent control unit sends attitude calibration commands to the attitude adjustment platform and receives gimbal angle feedback. The attitude adjustment platform's ball joint mechanism contains an angle encoder to achieve attitude closed-loop control. The compensation accuracy is controlled at 0.1 degrees. The cluster communication relay communicates with the intelligent control unit through the MESH network. The MESH network of the cluster communication relay supports a 3-unit network with a communication radius of 12 kilometers. The multi-source data fusion engine outputs the soil data model and the geological disaster risk assessment map. The ground station and the intelligent control unit transmit satellite data, mission parameters and fusion modeling results in both directions. The magnetic interface of the magnetic fuel range extender includes an axial positioning pin and a rare earth magnetic ring. The axial positioning pin limits radial displacement, and the rare earth magnetic ring provides a vertical adsorption force of 800N. The multi-source data fusion engine uses the XGBOOST algorithm to fuse the slope curvature characteristics of the drone point cloud, the vegetation cover index of the satellite image and the ground sensor data to generate a soil moisture heat map and a landslide risk probability model.

[0028] Step 1: Align the FRE-20 fuel range extender with the drone's belly and magnetically attach it. Insert the axial positioning pin into the base, and the rare earth magnetic ring automatically locks. Secure the lidar and multispectral camera to the attitude adjustment platform via the shock-absorbing base. The ball joint at the bottom of the platform connects to the carbon fiber body. The sensor unit collects terrain point clouds and surface spectra.

[0029] Step 2: LiDAR scans the terrain to calculate curvature in real time, which is then transmitted to LOGA Net for real-time decision-making. This in turn controls the environmental perception processor to quickly generate a zigzag obstacle avoidance route. The power system then performs a climb. Dynamic attitude adjustment is performed in the order of point cloud distortion rate, which is fed back to the attitude adjustment platform via the ball joint encoder to compensate for the yaw angle. The intelligent control unit generates attitude adjustment instructions based on the point cloud distortion rate. The attitude adjustment platform uses a six-degree-of-freedom parallel Stewart architecture. When the LiDAR detects that the point cloud distortion rate is greater than 5% due to body vibration, the intelligent control unit calculates the PID parameters and drives the electric cylinder to complete the adjustment within 200ms. Attitude compensation reduces the yaw angle deviation of the sensor platform to less than 0.05°. The six-degree-of-freedom attitude adjustment platform achieves precise attitude adjustment through a parallel electric cylinder-driven ball joint mechanism. Multiple servo motors work together to extend and retract the electric cylinder to push the upper platform carrying the sensor to perform pitch, yaw, roll, and multi-axis translation in space. The precision angle encoder embedded in the center of the ball joint provides real-time feedback on attitude deviation. The intelligent control unit calculates the point cloud distortion rate and generates PID instructions to drive the electric cylinder to complete vibration compensation in a short time, ensuring that the angle between the lidar beam and the terrain normal changes to an angle that ensures passing, suppressing point cloud distortion caused by vibration, and generating obstacle avoidance routes based on the terrain curvature.

[0030] Step 3: During use, the built-in voltage sensor will monitor the voltage. If the voltage of the lithium battery drops below 22V, the power system will switch the power supply mode according to the voltage threshold. The voltage sensor will transmit the command to the ignition controller of the magnetic fuel range extender through the controller to start the drive mode and improve the overall endurance of the drone.

[0031] Step 4. The ground station and the intelligent control unit work together to output a three-dimensional real-scene model. The multi-source data fusion engine of the drone synchronously receives various ground input data. The drone point cloud obtains high-precision terrain slope and curvature data through lidar scanning. All point clouds are synchronized at the microsecond level through RTK PPS pulse timestamp binding. Satellite images generate NDVI vegetation index through Landsat-9 near-infrared band images provided by the ground station, which are converted and aligned to drone coordinates through geographic coordinate conversion. Multiple sensors deployed in the ground monitoring area transmit soil moisture and other conditions in real time at a frequency of 1Hz. The fusion engine calls the XGBOOST algorithm for joint modeling, extracts the slope and curvature characteristics of the drone point cloud, overlays the vegetation cover index of the satellite image, and uses ground sensor data as training labels to calibrate the model.

[0032] Output two core mapping results:

[0033] Soil moisture heat map: Generate meter-level resolution distribution map;

[0034] Landslide risk probability model: Marks multiple high-risk areas, which are basically consistent with the actual landslide locations;

[0035] The final generated three-dimensional real-scene model improves the accuracy of the point cloud and superimposes the soil thermal map to form a decision-making base map. The ground station and intelligent decision-making control system automatically identify high-risk areas and generate the optimal evacuation path to avoid landslide points. When the landslide risk value is greater than 0.65, the ground station automatically marks the high-risk area and pushes the warning coordinates and evacuation path to the emergency center through the MESH network. At the same time, it pushes a warning to the emergency center to evacuate residents in the area in advance.

[0036] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A UAV-mounted system for complex terrain mapping, characterized by: include: A sensor unit comprising an integrated lidar, a multispectral camera, and a dual-antenna RTK / PPK positioning module; an intelligent control unit comprising an environmental perception processor, a dynamic path planning module, and a multi-source data fusion engine; a power system comprising a hybrid module consisting of a lithium battery and a magnetic fuel range extender, the magnetic fuel range extender being mounted on the aircraft belly via a magnetic interface containing an axial locating pin and a rare earth magnetic ring; a multi-degree-of-freedom attitude adjustment platform connected to the aircraft body via a ball joint mechanism; a cluster communication relay supporting MESH networking; and a ground station for providing satellite data and mission parameters.

2. The UAV-mounted system for complex terrain mapping according to claim 1, characterized in that: The sensor unit outputs point cloud, spectrum and positioning data to the intelligent control unit. The laser radar of the sensor unit is hard-synchronized with the multispectral camera through the 1PPS pulse of RTK, and the time error is ≤1 microsecond.

3. The UAV-mounted system for complex terrain mapping according to claim 1, characterized in that: The intelligent control unit sends course correction instructions to the power system and receives flight status feedback. The environment perception processor of the intelligent control unit deploys the LOGA·Net algorithm, and the terrain parsing delay is less than 50 milliseconds.

4. The UAV-mounted system for complex terrain mapping according to claim 1, characterized in that: The intelligent control unit sends attitude calibration instructions to the attitude adjustment platform and receives pan / tilt angle feedback. The ball joint mechanism of the attitude adjustment platform contains an angle encoder to achieve attitude closed-loop control with a compensation accuracy of ≤0.1 degrees.

5. The UAV-mounted system for complex terrain mapping according to claim 1, characterized in that: The cluster communication relay communicates with the intelligent control unit through a MESH network. The MESH network of the cluster communication relay supports a network of ≥3 units and a communication radius of ≥10 kilometers.

6. The UAV-mounted system for complex terrain mapping according to claim 1, characterized in that: The multi-source data fusion engine outputs a soil data model and a geological hazard risk assessment map.

7. The UAV-mounted system for complex terrain mapping according to claim 1, characterized in that: The ground station and the intelligent control unit bidirectionally transmit satellite data, mission parameters and fusion modeling results.

8. The UAV-mounted system for complex terrain mapping according to claim 1, characterized in that: The magnetic interface of the magnetic fuel range extender includes an axial positioning pin and a rare earth magnetic ring. The axial positioning pin limits radial displacement, and the rare earth magnetic ring provides a vertical adsorption force of ≥800N.

9. The UAV-mounted system for complex terrain mapping according to claim 1, characterized in that: The multi-source data fusion engine uses the XGBOOST algorithm to fuse the slope curvature characteristics of the drone point cloud, the vegetation cover index of the satellite image, and the ground sensor data to generate a soil moisture thermal map and a landslide risk probability model.

10. A method for surveying and mapping complex terrain using an unmanned aerial vehicle (UAV) system, based on the UAV system for surveying and mapping complex terrain described in claims 1-7, characterized in that: The following steps are involved: Step 1: The sensor unit collects terrain point cloud and surface spectrum; Step 2: The intelligent control unit generates attitude adjustment instructions based on the point cloud distortion rate and generates an obstacle avoidance route based on the terrain curvature; Step 3: The power system switches the power supply mode according to the voltage threshold; Step 4: The ground station and the intelligent control unit collaborate to output a three-dimensional real-scene model.

Images