A forestry backpack lidar multi-sensor integrated system

By integrating lidar probes, GNSS, true color cameras, binocular cameras and combined navigation modules in the backpack lidar system, high-precision space-time synchronization and time registration of multiple sensors is achieved, and small measurement range and time registration are solved. The problem of insufficient accuracy improves the acquisition accuracy of forest parameters and the stability of the system.

CN110646808BInactive Publication Date: 2024-11-26NORTHEAST FORESTRY UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201911026626.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2019-10-26
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing backpack lidar system has problems in the measurement of forest wood parameters with small measurement range, insufficient time registration accuracy, and poor satellite navigation signal stability, which affects the acquisition accuracy of forest wood parameters.

Method used

A forestry backpack lidar multi-sensor integrated system is designed, combining two lidar probes, GNSS, two true color cameras, a binocular camera and a combined navigation module to realize the multi-sensor between them through the microprocessor module. High-precision spatiotemporal synchronization, using GNSS time system and PPS signals for time synchronization, combined navigation modules provide high-precision navigation information, and enhance the system's anti-interference ability.

Benefits of technology

It improves the accuracy of forest wood parameters, ensures the stability and security of the system, expands the scope of space-time monitoring, enhances the credibility of data acquisition, and improves the system's comprehensive anti-interference ability and navigation continuity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN110646808B_ABST
    Figure CN110646808B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of forest parameter measurement, and in particular to a forestry backpack laser radar multi-sensor integrated system. The present invention is composed of various sensor components and a backpack mechanical structure, including two true color cameras, a binocular camera, a GNSS (Global Navigation Satellite System), two laser radar probes, an industrial computer, a combined navigation module, a battery, and a backpack mechanical mechanism. The present invention closely combines mobile measurement technology and multi-sensor integration technology, and has the characteristics of convenient mobility, strong adaptability, good stability, and portability; the present invention improves the time alignment accuracy of multiple sensors, can quickly and in real time obtain three-dimensional spatial geographic information, and reduce positioning errors; the present invention expands the system's time and space coverage, thereby improving the system's time and space monitoring range; the present invention improves the system's comprehensive anti-interference ability, ensuring the continuity of navigation and the reliability of the system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of forest parameter measurement, and in particular to a forestry backpack laser radar multi-sensor integrated system. Background Art

[0002] Forests are the largest, most widely distributed, most complex ecosystems on land, and richest in material resources. Traditional forest parameter determination requires field measurements, which can only obtain data at some points, which is not conducive to studying regional scale or large-scale forest parameters. With the increasing depth and maturity of the application of remote sensing technology, remote sensing technology has been successfully used in large-area forest mapping, disaster monitoring, etc. In particular, as the fastest-growing LiDAR (Light Detection And Ranging), laser radar, it can overcome the limitations of optical remote sensing and microwave remote sensing signal saturation and improve the accuracy of forest three-dimensional parameter estimation. LiDAR is an active remote sensing technology that has developed very rapidly internationally in recent years, and has been successfully applied in the quantitative measurement and inversion of forest parameters. LiDAR has an imaging mechanism different from passive optical remote sensing, and has a strong ability to detect vegetation spatial structure and terrain, especially the ability to detect forest height, which has advantages that other remote sensing data cannot match.

[0003] There are three-dimensional laser scanning systems with different platforms, such as vehicle-mounted laser radar systems, airborne laser radar systems, and ground-based laser radar systems. Among them, the ground-based laser radar system is widely used in the field of forest parameter measurement. As a rapidly developing measurement technology, it is suitable for the acquisition of three-dimensional point cloud data in forest scenes. It can not only obtain detailed data of individual trees, but also scan spatial data within a sample plot. It is a high-precision non-contact three-dimensional measurement method that can quickly obtain the structural parameters of forest samples and describe the three-dimensional structure of the forest in detail. In practical applications, due to the bulkiness of general ground-based laser radar equipment, the limitation of scanning angles, and the existence of mutual occlusion of trees, the scanning efficiency of the laser radar probe is often affected, which in turn affects the overall progress of the measurement work. The backpack laser radar system in the ground-based laser radar system has the characteristics of easy installation, easy operation, and portability, and can adapt to environments with dense forests and complex terrain. However, in practical applications, the existing backpack laser radar system still has the shortcomings of small scanning coverage, large positioning error, and poor synchronization of data transmission, which in turn affects the accuracy of forest parameter collection. Summary of the invention

[0004] The present invention overcomes the shortcomings of the prior art and realizes a forestry backpack laser radar multi-sensor integrated system. The present invention improves the existing backpack laser radar system's small measurement range, insufficient time registration accuracy, and poor satellite navigation signal stability, thereby improving the system's collection accuracy of tree parameters such as tree height, breast diameter, and number of individual trees.

[0005] To achieve the above object, the technical solution of the present invention is as follows:

[0006] 1. The present invention provides a backpack-type laser radar multi-sensor integrated system for measuring forest parameters, including: two laser radar probes, GNSS (Global Navigation Satellite System), two true color cameras, a binocular camera, a combined navigation module, an industrial computer, a battery, and a backpack mechanical structure. The backpack mechanical structure includes a support rod, a backpack frame, a backpack bottom plate, and a backpack shell. The support rod is installed and fixed by a locking buckle on the backpack bottom plate through a threaded connection, and is used to install the GNSS and a laser radar probe; the backpack frame is connected to the backpack bottom plate through a threaded connection, and is used to install the two true color cameras, the combined navigation module, the industrial computer, a binocular camera, and a laser radar probe. The backpack bottom plate is used to install the backpack frame, support rod, battery, and backpack shell; the backpack shell is nested with the components on the backpack frame and is installed on the backpack bottom plate through threads.

[0007] 2. The two laser radar probes are used for extracting laser point cloud data in forest scenes. The laser radar probe horizontally installed on the top of the support pole emits laser energy and receives return signals with a scanning range of 270° to the rear, and collects and stores them as point cloud data; the other laser radar probe is installed on the laser radar probe mounting plate on the backpack frame, with the cross-section of the laser radar probe used as the reference plane, and the scanning range with an angle of ±15° to the reference plane is used to extract the laser point cloud data on one side of the backpack shell.

[0008] 3. The industrial computer is installed on the right support plate of the backpack frame through a copper column. The microprocessor module is installed on the mainboard of the industrial computer to coordinate and control the data collection and synchronous transmission between the sensors. The industrial computer is connected to each sensor component through a cable.

[0009] 4. The two true color cameras are symmetrical about the binocular camera in the upper and lower parts and are installed on the true color camera mounting plate of the backpack frame. They collect true color forest image information in a forest environment by taking pictures continuously and at regular intervals in symmetrical positions, thereby improving the richness and accuracy of forest parameter collection. The pictures taken by the true color cameras are matched with the point cloud data collected by the lidar probe, thereby supplementing the tree texture and spectral information missing from the point cloud data.

[0010] 5. The binocular camera is installed on the binocular camera mounting plate of the backpack frame, and the binocular stereo matching algorithm is used to calculate and obtain the depth information and three-dimensional model of the forest scene in real time. When the time difference image is obtained by stereo matching, the depth image can be determined, the three-dimensional scene information can be restored and the three-dimensional point cloud information can be generated; the binocular camera is used to obtain the forest scene image from different spatial perspectives, and the relative position of the binocular camera is calibrated, and the stereo matching algorithm is used to match the feature points of the two collected forest images, and then the accurate and fast feature algorithm and binocular passive stereo vision are used to reconstruct the forest scene in three dimensions, thereby realizing the restoration of the three-dimensional forest scene.

[0011] 6. The combined navigation module is installed on the left support plate of the backpack frame through a copper column, and includes a GPS satellite navigation signal receiver and an IMU inertial measurement unit. The combination of the two can obtain high-precision three-dimensional position, speed, and attitude information in real time, which is convenient for inertial attitude measurement in motion or vibration state, and reduces measurement errors through nonlinear compensation, orthogonal compensation, temperature compensation, and drift compensation. The combined navigation module has high short-term accuracy and can provide effective navigation information when the satellite navigation signal is lost, ensuring the continuity of navigation and the reliability of the system. The combined use of the GPS satellite navigation signal receiver and the IMU inertial measurement unit improves the comprehensive anti-interference ability of the system.

[0012] 7. The present invention combines the spatiotemporal characteristics of GNSS, two laser radar probes, two true color cameras, a binocular camera and a combined navigation module, and designs and implements the control function of the microprocessor module. The microprocessor module introduces the GNSS time system during the operation process, and can maintain the time system during the GNSS signal interruption, thereby realizing high-precision spatiotemporal synchronization between multiple sensors. The microprocessor module unifies the asynchronous data information of each sensor observing the same target to the same time reference, so that the observation data of each sensor on the same target at the same time corresponds. The microprocessor module obtains the time reference from the industrial computer system clock and the GNSS time system clock, controls the image data acquisition of the true color camera and the binocular camera and the point cloud data acquisition of the laser radar probe, realizes the unification of the time reference of heterogeneous data, and can convert the laser point cloud and optical image from the relative coordinate system to the absolute measurement coordinate system, thereby realizing data synchronization control.

[0013] Due to the existence of unstable factors such as various interference signals, the present invention uses the PPS signal provided by the GPS satellite navigation signal receiver in the combined navigation module and the corresponding GNSS time to achieve strict synchronization of the time of each sensor, eliminating the influence of the interference signal on the time. The synchronization accuracy can reach 0.1ms by timing the PPS signal through GNSS. The microprocessor module receives the time and space information transmitted by GNSS, the relative space information and setting information of the laser radar probe, establishes a time reference, and integrates the position information, distance information and time information. The maximum output frequency of the two laser radar probes that can adjust the scanning data frequency is 100Hz, that is, scanning information can be obtained once every 10ms, and the output frequency and positioning output frequency of the GNSS raw data are up to 50Hz, that is, positioning data information can be obtained once every 20ms, and it also has the function of recording time. In order to ensure that the point cloud data collected by the laser radar probe is synchronized with the GNSS data, the collection of GNSS data is recorded in time while the laser radar probe is collecting point cloud data, thereby ensuring that the sensor data can be solved in the later processing. The time recording information of GNSS can be used to perform time alignment between multiple sensors.

[0014] Compared with the prior art, the present invention has the following beneficial effects:

[0015] To solve the problem of low time registration accuracy of existing laser radar systems, the system combines the time delay characteristics of GNSS, two laser radar probes, two true color cameras, a binocular camera and a combined navigation module. By recording and analyzing the time stamps under the working state of multiple sensors, a time benchmark is established to achieve collaborative work and time synchronization between sensors, and improve the accuracy of time registration. Ensure the stability of the working environment of the backpack laser radar multi-sensor integrated system and the safety during the travel process, enhance the survivability of the system, and improve the credibility of data collection. Expand the time and space coverage, and multiple sensors overlap and cover each other, thereby increasing the time and space monitoring range of the system. The combined navigation module improves the comprehensive anti-interference ability of the system, ensuring the continuity of navigation and the reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a front view of the internal structure of the present invention;

[0017] Figure 2 This is the appearance diagram of the present invention;

[0018] Figure 3 This is a structural diagram of the backpack frame of the present invention;

[0019] Figure 4 This is a schematic diagram of the support rod installation method of the present invention;

[0020] Figure 5It is the left isometric view of the present invention;

[0021] Figure 6 It is the right isometric view of the present invention;

[0022] Figure 7 Schematic diagram of multi-sensor time synchronization control.

[0023] Wherein: laser radar probe (1), GNSS (2), true color camera (3), binocular camera (4), integrated navigation module (5), industrial computer (6), microprocessor module (601), FPGA (602), battery (7), support rod (801), backpack frame (802), laser radar probe mounting plate (8021), true color camera mounting plate (8022), binocular camera mounting plate (8023), left support plate (8024), right support plate (8025), backpack bottom plate (803), locking buckle (8031), backpack shell (804). DETAILED DESCRIPTION

[0024] The specific implementation of the present invention is further described below in conjunction with the drawings and examples.

[0025] A forestry backpack laser radar multi-sensor integrated system consists of various sensor components and a backpack mechanical structure.

[0026] like Figure 1 As shown, the present invention provides a forestry backpack laser radar multi-sensor integrated system, which includes: two laser radar probes (1), a GNSS (Global Navigation Satellite System) (2), two true color cameras (3), a binocular camera (4), a combined navigation module (5), an industrial computer (6), a battery (7), and a backpack mechanical structure.

[0027] like Figure 1 , Figure 2 , Figure 3 As shown, the backpack mechanical structure includes a support rod (801), a backpack frame (802), a backpack bottom plate (803) and a backpack shell (804).

[0028] like Figure 4 As shown, the support rod (801) is fixed by a locking buckle (8031) threadedly connected to the backpack bottom plate (803), and is used to install the GNSS (2) and a laser radar probe (1);

[0029] like Figure 1 , Figure 3As shown, the backpack frame (802) is fixed on the backpack bottom plate (803) and is used to install the two true color cameras (3), the combined navigation module (5), the industrial computer (6), a binocular camera (4) and a laser radar probe (1);

[0030] like Figure 1 As shown, the backpack bottom plate (803) is used to install the backpack frame (802), the support rod (801), the battery (7), and the backpack shell (804);

[0031] like Figure 2 As shown, the backpack shell (804) is nested with the components on the backpack frame (802) and is installed on the backpack bottom plate (803) through threads.

[0032] Further, such as Figure 1 As shown, the two laser radar probes (1) are used for extracting laser point cloud data in a forest scene, wherein the laser radar probe (1) horizontally installed on the top of the support rod (801) emits laser energy and receives return signals in a scanning range with an angle of 270° to the rear, and collects and stores them as point cloud data; the other laser radar probe (1) is installed on the laser radar probe mounting plate (8021) on the backpack frame (802), and the cross section of the laser radar probe (1) is used as a reference plane, and the scanning range with an angle of ±15° with the reference plane is used to extract laser point cloud data on one side of the backpack shell (804).

[0033] Further, such as Figure 6 As shown, the industrial computer (6) is installed on the right support plate (8025) of the backpack frame (802) through a copper column, and the microprocessor module (601) is installed on the mainboard of the industrial computer (6). The microprocessor module (601) coordinates and controls the data collection and synchronous transmission between the sensors, and the industrial computer (6) is connected to the sensor components through cables.

[0034] Further, such as Figure 1 As shown, the two true color cameras (3) are symmetrical about the binocular camera (4) in the upper and lower parts, and are installed on the true color camera mounting plate (8022) on the backpack frame (802). By continuously taking pictures at regular intervals in symmetrical positions, full color image information in a forest environment is collected, thereby improving the richness and accuracy of the collection of forest parameters; the pictures taken by the true color camera (3) are matched with the point cloud data collected by the laser radar probe (1), thereby supplementing the tree texture and spectral information missing from the point cloud data.

[0035] Further, such as Figure 1As shown, the binocular camera (4) is mounted on the binocular camera mounting plate (8023) of the backpack frame (802). The binocular camera (4) uses a binocular stereo matching algorithm to calculate and obtain the depth information and three-dimensional model of the forest scene in real time. After the time difference image is obtained by stereo matching, the depth image can be determined, the three-dimensional scene information can be restored and the three-dimensional point cloud information can be generated; the binocular camera (4) is used to obtain the forest scene image from different spatial perspectives, and the relative position of the binocular camera (4) is calibrated, and the feature points of the two collected forest images are matched using a stereo matching algorithm, and then the forest scene is reconstructed in three dimensions using an accurate and fast feature algorithm and binocular passive stereo vision, thereby realizing the restoration of the three-dimensional forest scene.

[0036] Further, such as Figure 5 As shown, the combined navigation module (5) is mounted on the left support plate (8024) of the backpack frame (802) through a copper column, and includes a GPS satellite navigation signal receiver and an IMU inertial measurement unit. The combination of the two can obtain high-precision three-dimensional position, speed, and attitude information in real time, facilitate inertial attitude measurement in a motion or vibration state, and reduce measurement errors through nonlinear compensation, orthogonal compensation, temperature compensation, and drift compensation. The combined navigation module has high short-term accuracy and can provide effective navigation information when the satellite navigation signal is lost, ensuring the continuity of navigation and the reliability of the system. The combined use of the GPS satellite navigation signal receiver and the IMU inertial measurement unit improves the comprehensive anti-interference ability of the system.

[0037] Furthermore, the combined navigation module (5) is used to calculate in real time the motion trajectory information in the forest scene, including the position, speed information, attitude information and acceleration information; the GPS satellite navigation receiver and the IMU inertial measurement unit are driven in the forest scene, and the GPS satellite navigation receiver and the IMU inertial measurement unit are driven to operate and information is integrated according to the scene data information calculated in real time by the system, so as to realize the combined navigation operation of the system in the forest scene; the real-time satellite navigation information generated by the GPS satellite navigation receiver and the real-time attitude and acceleration information output by the IMU inertial measurement unit are transmitted to the microprocessor module (601) through a cable, so as to realize the real-time coordinate solution and obtain the positioning result in real time;

[0038] Furthermore, the microprocessor module (601) obtains a time reference from the system clock of the industrial computer (6) and the time system clock of the GNSS (2), controls the image data acquisition of the true color camera (3) and the binocular camera (4) and the point cloud data acquisition of the laser radar probe (1), and realizes the unification of the time reference of heterogeneous data, thereby being able to convert the laser point cloud and the optical image from a relative coordinate system to an absolute measurement coordinate system, and realize data synchronization control.

[0039] Furthermore, the present invention utilizes the PPS signal provided by the GPS satellite navigation signal receiver in the combined navigation module (5) and the corresponding GNSS (2) time to achieve strict synchronization of the time of each sensor, thereby eliminating the influence of interference signals on the time. Figure 7 As shown, the PPS signal is timed by GNSS (2), and the synchronization accuracy can reach 0.1ms. The microprocessor module (601) receives the time and space information transmitted by GNSS (2), the relative space information and setting information of the laser radar probe (1), establishes a time reference, and integrates the position information, distance information and time information.

[0040] Furthermore, in view of the characteristics of the present invention, the industrial computer system clock uses a 100MHz high-stability crystal oscillator and a team-style parallel working FPGA (602). When debugging or changing a certain submodule, it will not affect the implementation results of other modules. FPGA (602) completes the key part of the time calibration, and generates us, ms and s pulse generation units through programming control. When the alignment pulse generation circuit detects a valid 1PPS pulse, an alignment pulse is generated. At this time, the ms and s pulse generation circuits are reset to complete a time calibration, that is, when the 1PPS pulse edge reaches the FPGA (602) circuit, the circuit clears the decimal part of the time "second".

[0041] Furthermore, the scanning data frequency of the two laser radar probes (1) can be adjusted to a maximum output frequency of 100 Hz, that is, scanning information can be obtained every 10 ms, and the output frequency of the original data of the GNSS (2) and the positioning output frequency are up to 50 Hz, that is, positioning data information can be obtained every 20 ms, and it also has the function of recording time. In order to ensure that the point cloud data collected by the laser radar probe (1) and the GNSS (2) data can maintain time synchronization, while the laser radar probe (1) is collecting data, the collection of the GNSS (2) data is recorded in time, thereby ensuring that the sensor data can be solved in the later processing. Time recording information of GNSS (2) is used to perform time alignment between multiple sensors.

[0042] Furthermore, since the sampling frequencies of the sensors are different, it is necessary to perform internal difference and extrapolation processing on the target data of different frequencies collected by the sensors, and at the same time select the observation time series of the sensor with the highest output frequency as the time reference, and then extrapolate the observation data collected by the low-frequency sensor to this time reference to improve the accuracy of multi-sensor time registration. The data collection and recording time of each sensor can be synchronized and converted into a unified time coordinate system. Time registration is to accurately fuse the image information collected by the two true color cameras (3) and the binocular camera (4), the three-dimensional space point cloud data obtained by the two laser radar probes (1), and the positioning and attitude data information obtained by the GPS satellite navigation receiver and the IMU inertial measurement unit in the combined navigation module (5) at each sampling moment, so as to keep the time series of data collection synchronized.

[0043] Furthermore, the GPS satellite navigation receiver data output frequency is set to Δf GPS =20Hz, the laser radar sensor (1) data output frequency is Δf LiDAR =50Hz, the sampling frequency of the IMU inertial measurement unit is Δf IMU =20Hz, the frequency of GNSS output is Δf GNSS =50Hz, and then the BP neural network method is used to align the GPS satellite navigation receiver positioning data and the IMU inertial measurement unit observation data to ensure that the output time interval of each sensor in the system is unified to Δt = 1 / Δf GNSS =1 / Δf LiDAR , so that the attitude and position parameters of the laser radar probe (1) data can be obtained, and the conversion equation between the world coordinate system and the space coordinate system of the backpack laser radar multi-sensor integrated system at the nth moment can be obtained as follows:

[0044]

[0045] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A forestry backpack laser radar multi-sensor integrated system, characterized in that: include: Two laser radar probes, GNSS (Global Navigation Satellite System), two true color cameras, a binocular camera, a combined navigation module, an industrial computer, a battery, and a backpack mechanical structure; the backpack mechanical structure includes a support rod, a backpack frame, a backpack bottom plate, and a backpack shell; the support rod is installed and fixed by a locking buckle connected to the backpack bottom plate by a thread, and is used to install the GNSS and a laser radar probe; the backpack frame is connected to the backpack bottom plate by a thread, and is used to install the two true color cameras, the combined navigation module, the industrial computer, a binocular camera, and a laser radar probe; the backpack bottom plate is used to install the backpack frame, the support rod, the battery, and the backpack shell; the backpack shell is nested with the components on the backpack frame, and is installed on the backpack bottom plate by a thread; The two laser radar probes are used for extracting laser point cloud data in forest scenes, wherein the laser radar probe horizontally mounted on the top of the support rod emits laser energy and receives return signals within a scanning range of 270° to the rear, and collects and stores the data as point cloud data; the other laser radar probe is mounted on the laser radar probe mounting plate on the backpack frame, and the cross section of the laser radar probe is used as a reference plane, and the scanning range with an angle of ±15° with the reference plane is used to extract laser point cloud data on one side of the backpack shell; The industrial computer is installed on the right support plate of the backpack frame through a copper column; the microprocessor module is installed on the mainboard of the industrial computer to coordinate and control the data collection and synchronous transmission between the sensors; The two true color cameras are symmetrical about the binocular camera and are mounted on the true color camera mounting plate of the backpack frame. They continuously and regularly take pictures in symmetrical positions to collect true color forest image information in a forest environment, thereby improving the richness and accuracy of the collection of forest parameters. The combined navigation module is installed on the left support plate of the backpack frame through a copper column, and includes a GPS satellite navigation signal receiver and an IMU inertial measurement unit. The combination of the two can obtain high-precision three-dimensional position, speed, and attitude information in real time, which is convenient for inertial attitude measurement in motion or vibration state; the combined navigation module has high short-term accuracy and can provide effective navigation information when the satellite navigation signal is lost, ensuring the continuity of navigation and the reliability of the system; The microprocessor module introduces the GNSS time system during the operation process. The microprocessor module obtains the time reference from the industrial computer system clock and the GNSS time system clock, controls the image data acquisition of the true color camera and the binocular camera and the point cloud data acquisition of the laser radar probe, realizes the unification of the time reference of heterogeneous data, and realizes high-precision spatiotemporal synchronization among multiple sensors. The two true color cameras continuously and regularly take pictures in symmetrical positions to collect full color image information in a forest environment, thereby improving the richness and accuracy of the collection of forest parameters; the pictures taken by the true color cameras are matched with the point cloud data collected by the laser radar probe, thereby supplementing the tree texture and spectral information missing from the point cloud data; The binocular camera uses a binocular stereo matching algorithm to calculate and obtain the depth information and three-dimensional model of the forest scene in real time; when the stereo matching obtains the time difference image, the depth image can be determined, the three-dimensional scene information can be restored and the three-dimensional point cloud information can be generated; the binocular camera is used to obtain the forest scene image from different spatial perspectives, and the relative position of the binocular camera is calibrated, and the stereo matching algorithm is used to match the feature points of the two collected forest images, and then the accurate and fast feature algorithm and binocular passive stereo vision are used to reconstruct the forest scene in three dimensions, thereby realizing the restoration of the three-dimensional forest scene; The integrated navigation module includes a GPS satellite navigation signal receiver and an IMU inertial measurement unit. The combination of the two can obtain high-precision three-dimensional position, speed, and attitude information in real time, which is convenient for inertial attitude measurement in motion or vibration state. The measurement error is reduced through nonlinear compensation, orthogonal compensation, temperature compensation, and drift compensation. The integrated navigation module has high short-term accuracy and can provide effective navigation information when the satellite navigation signal is lost, thereby ensuring the continuity of navigation and the reliability of the system. The combined use of the GPS satellite navigation signal receiver and the IMU inertial measurement unit improves the comprehensive anti-interference ability of the system. The combined navigation module calculates the motion trajectory information in the forest scene in real time, including position, speed, attitude and acceleration information; drives the GPS satellite navigation receiver and the IMU inertial measurement unit in the forest scene, and drives the GPS satellite navigation receiver and the IMU inertial measurement unit to operate and fuse information according to the scene data information calculated by the system in real time, so as to realize the combined navigation operation of the system in the forest scene; transmits the real-time satellite navigation information generated by the GPS satellite navigation receiver and the real-time attitude and acceleration information output by the IMU inertial measurement unit to the microprocessor module through cables, so as to realize real-time coordinate solution and obtain positioning results in real time; The microprocessor module obtains the time reference from the industrial computer system clock and the GNSS time system clock, controls the image data acquisition of the true color camera and the binocular camera and the point cloud data acquisition of the laser radar probe, realizes the unification of the time reference of heterogeneous data, and thus can convert the laser point cloud and optical image from the relative coordinate system to the absolute measurement coordinate system, and realizes data synchronization control; The PPS signal provided by the GPS satellite navigation signal receiver in the integrated navigation module and the corresponding GNSS time realize strict synchronization of the time of each sensor, eliminating the influence of interference signals on the time; the PPS signal is timed by GNSS, and the synchronization accuracy can reach 0.1ms; The microprocessor module receives the time and space information transmitted by GNSS, the relative space information and setting information of the laser radar probe, establishes a time reference, and integrates the position information, distance information and time information; The system clock of the industrial computer uses a 100MHz high-stable crystal oscillator and a team-based parallel FPGA. When debugging or changing a submodule, the implementation results of other modules will not be affected. The FPGA completes the key part of the time calibration, and generates the us, ms and s pulse generation units through programming control. When the alignment pulse generation circuit detects a valid 1PPS pulse, an alignment pulse is generated. At this time, the ms and s pulse generation circuits are reset to complete a time calibration. That is, when the 1PPS pulse edge reaches the FPGA circuit, the circuit clears the decimal part of the time "second" to zero. The scanning data frequency of the two laser radar probes can be adjusted to a maximum output frequency of 100Hz, that is, scanning information can be obtained every 10ms, and the output frequency of GNSS raw data and positioning output frequency is up to 50Hz, that is, positioning data information can be obtained every 20ms, and it also has the function of recording time; in order to ensure that the point cloud data collected by the laser radar probe and the GNSS data can maintain time synchronization, the collection of GNSS data is recorded at the same time as the laser radar probe collects data, thereby ensuring that the sensor data can be solved in the later processing. Time recording information of GNSS can be used to perform time alignment between multiple sensors; The target data of different frequencies collected by each sensor are processed by internal difference and extrapolation. At the same time, the observation time series of the sensor with the highest output frequency is selected as the time base, and then the observation data collected by the low-frequency sensor is extrapolated to this time base to improve the accuracy of multi-sensor time registration; the data collection and recording time of each sensor can be synchronized and converted to a unified time coordinate system; time registration is to accurately integrate the image information collected by the two true color cameras and binocular cameras, the three-dimensional space point cloud data obtained by the two laser radar probes, and the positioning and attitude data information obtained by the GPS satellite navigation receiver and the IMU inertial measurement unit in the combined navigation module at each sampling moment, so as to keep the time series of data collection synchronized; Set the data output frequency of the GPS satellite navigation receiver to ΔfGPS = 20Hz, the data output frequency of the lidar probe to ΔfLiDAR = 50Hz, the sampling frequency of the IMU inertial measurement unit to ΔfIMU = 20Hz, and the frequency of GNSS output to ΔfGNSS = 50Hz. Then use the BP neural network method to align the GPS satellite navigation receiver positioning data and the IMU inertial measurement unit observation data to ensure that the output time interval of each sensor in the system is unified to Δt = 1 / ΔfGNSS = 1 / ΔfLiDAR, so that the attitude and position parameters of the lidar probe data can be calculated.