Self-following ground penetrating radar detection system and control method thereof

Abstract

This invention provides a self-following ground-penetrating radar (GPR) detection system and its control method. A handheld terminal provides real-time positioning of the operator, issues detection tasks marked with target areas, paths, or follow instructions, and receives detection results. The self-following GPR device adopts a modular architecture. The mobile platform flexibly connects the chassis and drive mechanism via suspension arms and shock-absorbing damping components, enabling all-terrain capability. The self-following control module integrates sensors such as positioning, lidar, vision, and IMU, fusing multimodal data to complete path planning, follow control, and obstacle avoidance decisions, and actively adjusts the suspension based on attitude recognition to maintain vehicle stability. The GPR is installed at the bottom of the platform to collect echo data. The GPR control unit jointly processes the radar data and positioning information to generate radar profile maps, 3D imaging maps, and detection reports, which are then transmitted back to the handheld terminal. This invention balances the efficiency of automated surveys with the flexibility of manual intervention, enabling reliable positioning and following even in environments with satellite signal rejection, and suppressing vibration interference through active attitude adjustment to ensure data quality.

Description

Technical Field

[0001] This invention relates to the field of radar equipment technology, and in particular to a self-following ground-penetrating radar detection system and its control method. Background Technology

[0002] Ground penetrating radar (GPR) is a non-destructive geophysical exploration technology that uses high-frequency electromagnetic waves to detect the distribution of underground media. Its working principle involves transmitting high-frequency electromagnetic waves into the ground via a transmitting antenna. When these waves encounter interfaces between underground media with different dielectric constants, they are reflected. A receiving antenna collects the reflected echo signals. By analyzing the time-domain characteristics and amplitude variations of the echo signals, the distribution structure and properties of the underground media can be deduced. GPR technology has significant advantages such as high resolution, fast detection speed, ease of operation, and non-destructive testing, and has been widely applied in many fields, including road inspection, geological exploration, pipeline detection, tunnel inspection, archaeological excavation, and environmental monitoring.

[0003] Traditional ground-penetrating radar (GPR) operations primarily rely on manual hand-held or vehicle-mounted transport. Manual hand-held methods require operators to bend over or squat for extended periods, resulting in high labor intensity and continuous operation time typically not exceeding two hours, making it difficult to meet the demands of large-area rapid detection. Furthermore, maintaining a constant manual propulsion speed leads to poor uniformity and consistency in data acquisition. While vehicle-mounted transport can improve efficiency, it is limited by road conditions, making it unsuitable for effective operation in complex terrains (such as rugged mountains, soft sand, and muddy roads), confined spaces (such as underground utility tunnels or tunnel interiors), or unstructured environments (such as construction sites or disaster sites). In addition, traditional methods generally suffer from difficulties in controlling the detection trajectory, low accuracy in synchronizing data and location information, and poor safety in harsh environments.

[0004] To overcome the aforementioned shortcomings, automated detection solutions combining autonomous mobile platforms with ground-penetrating radar (GPR) have emerged in recent years. However, existing solutions still suffer from the following deficiencies: First, while GPR robots employing fully autonomous navigation can operate unmanned in ideal environments, in complex work sites (such as urban roads and archaeological sites), the complete separation of the machine from human intervention leads to a loss of flexibility. When encountering sudden obstacles or needing to re-examine abnormal areas, operators cannot easily adjust the detection path and must instead replan the path through the terminal, resulting in decreased operational efficiency. Second, autonomous navigation solutions based on Global Navigation Satellite Systems (GNSS) or Real-Time Kinematics (RTK) suffer from severe positioning drift in environments with weak or absent satellite signals, such as tunnels, tree-lined roads, and under overpasses. Solutions based on simultaneous positioning and mapping using 3D lidar are not only costly but also prone to positioning loss in long, straight corridors or open, featureless areas, leading to spatial misalignment of radar data and affecting detection accuracy. Third, deep learning-based intelligent scanning solutions only focus on underground targets and automatically plan paths based on those targets. However, they rely entirely on pre-stored maps or real-time mapping and cannot perceive the operator's real-time position and movement intentions. In scenarios such as temporary changes to the detection area or on-site obstacle avoidance, natural interaction between the operator and the robot is impossible. Fourth, both multi-robot collaborative solutions and autonomous solutions equipped with high-precision integrated navigation (inertial measurement unit, LiDAR, RTK) involve complex multi-sensor fusion algorithms and expensive hardware configurations. The systems are redundant, complex, and costly, increasing equipment weight and power consumption, raising the barriers to procurement and maintenance, and hindering large-scale engineering applications.

[0005] Therefore, there is an urgent need for a ground-penetrating radar detection system that can achieve automated and efficient detection, allow for flexible on-site intervention by operators, and is highly adaptable to the environment and cost-controllable. Summary of the Invention

[0006] In view of this, embodiments of the present invention provide a self-following ground-penetrating radar detection system and its control method, which solves the problem that existing ground-penetrating radar equipment cannot simultaneously achieve automated and efficient detection and flexible on-site intervention by operators, and has weak adaptability to complex scenarios.

[0007] One aspect of the present invention provides a self-following ground-penetrating radar detection system, the system comprising a handheld terminal and a self-following ground-penetrating radar device; The handheld terminal is used to locate the operator, send the detection task and detection task parameters marked with the target area, target path, target point or follow command to the self-following ground-penetrating device, and receive ground-penetrating radar data; The self-following ground-penetrating radar device includes: The self-following mobile platform includes a walking chassis and a drive mechanism, wherein the drive mechanism is connected and fixed to the walking chassis via a suspension swing arm and a shock-absorbing damping component. The equipment installation platform is located on top of the self-following mobile platform; A self-following control module is mounted on the equipment mounting platform and connected to the drive mechanism. The self-following control module is equipped with a first positioning module, a lidar, a vision camera, an IMU (Inertial Measurement Unit), and a wheel speed encoder to collect multimodal sensing data. Combining this multimodal sensing data, it performs path planning, following control, or obstacle avoidance decisions for the detection task, generating motion control commands to control the drive mechanism for displacement movement. Based on the IMU, it performs attitude recognition and generates attitude control commands to control the suspension arm for attitude adjustment. Ground-penetrating radar is installed at the bottom of the equipment mounting platform to collect ground-penetrating radar data; The ground-penetrating control unit, mounted on the equipment mounting platform, is used to collect and process the ground-penetrating radar data, and send it to the ground station in conjunction with the positioning data collected by the first positioning module to generate radar profile, three-dimensional imaging map and detection report and send them back to the handheld terminal; A power supply unit is used to supply power to the self-following ground-penetrating radar device.

[0008] In some embodiments, the handheld terminal includes a second positioning module, an interactive display module, a processor unit, and a wireless communication module; the second positioning module uses a UWB positioning tag or an RTK satellite navigation module; the processor unit is used to load a control program to provide a preset interactive interface through the interactive display module for importing detection tasks and displaying the radar profile, the three-dimensional imaging map, or the detection report; the wireless communication module uses a Bluetooth module, a WiFi module, a Zigbee module, or a data network communication module.

[0009] In some embodiments, the chassis adopts a frame structure and is provided with multiple fixed points for mounting the shock-absorbing damping components; the shock-absorbing damping components are hydraulic dampers, shock-absorbing springs, polymer elastic dampers, magnetorheological dampers, or electrorheological dampers. The first end of the suspension arm is fixedly connected to the shock absorber by a rotating connector and the angle is adjusted by a hydraulic rod; the second end of the suspension arm is connected to the drive mechanism, which is a wheeled, tracked, or wheel-tracked hybrid type.

[0010] In some embodiments, the drive mechanism is either wheel-footed or leg-footed.

[0011] In some embodiments, the equipment mounting platform adopts a frame structure and is equipped with multiple shock-absorbing connecting seats and equipment mounting rails. The equipment mounting rails are equipped with a battery compartment, a counterweight adjustment slot, the self-following control module, and a ground-penetrating control unit. The equipment mounting platform has a radar host mounting position in the center for mounting the ground-penetrating radar. The equipment installation platform is connected and fixed to the self-following mobile platform through an active shock absorption mechanism.

[0012] In some embodiments, the active damping mechanism adopts a two-stage damping structure, including: a primary damping layer using polymer gaskets to absorb high-frequency vibrations; a first-end mounting base deployed on the primary damping layer; a limiting buffer block fixedly connected to the first-end mounting base; and a hydraulic damper fixedly connected to the limiting buffer block, wherein the guide rod of the hydraulic damper is connected to and fixedly fixed to a second-end mounting base, and a damping spring is provided around the guide rod to provide support.

[0013] In some embodiments, the self-following control module is further configured to plan a path for the target area to be detected using a bow-shaped round-trip path algorithm, a cell decomposition full coverage algorithm, a spiral path algorithm, an artificial potential field method, or an improved A algorithm; to plan a path for the target point using the Dijkstra algorithm; to execute the follow command using a visual locking real-time follow and historical trajectory delayed follow control algorithm; and to execute the obstacle avoidance decision using a dynamic window algorithm.

[0014] In some embodiments, the self-following control module performs attitude recognition based on the IMU (Inertial Measurement Unit) and generates attitude control commands to control the suspension arm to adjust its attitude, including: The motion data of the equipment installation platform is collected in real time by the IMU inertial measurement unit. After error correction by performing Kalman filtering or particle filtering, the left and right tilt angles and front and back tilt angles of the equipment installation platform are calculated. The left and right tilt angles and the front and back tilt angles are compared with the horizontal target attitude to calculate the correction deviation value; Based on the corrected deviation value and the structural parameters of the suspension arms, the target extension stroke of each suspension arm is calculated, and the attitude control command is generated to control the suspension arms to adjust their attitude to ensure that the equipment installation platform is level.

[0015] On the other hand, the present invention also provides a control method for a self-following ground-penetrating radar detection system, the method being executed by a self-following ground-penetrating radar device in the aforementioned self-following ground-penetrating radar detection system, the method comprising: Receive detection tasks and detection task parameters sent by the handheld terminal. The detection tasks are used to mark the target area, target path, target location, or record follow instructions. For the detection task that marks the target area, target path, or target point, path planning is performed. The current position is collected by the first positioning module, and the vehicle moves autonomously according to the planned path. For the detection task that records follow instructions, the operator's positioning data sent by the handheld terminal is acquired in real time, and the operator's movement path is recorded. The current position is collected by the first positioning module, and the vehicle moves autonomously according to the operator's movement path. Obstacle detection and dynamic obstacle avoidance are performed based on point cloud data collected by lidar and environmental images collected by visual cameras. Based on the detection mission parameters, the ground-penetrating radar is controlled to emit electromagnetic waves and receive echoes during autonomous driving, while simultaneously collecting radar data and its own current position information. The radar data is filtered, gain adjusted, and background removed in real time using edge computing. Then, it is combined with the user's current location information and sent to the ground station to generate a radar profile, a 3D imaging map, and a detection report, which are then transmitted back to the handheld terminal.

[0016] In some embodiments, the method further includes: The system receives terrain feedback data sent by the handheld terminal. For flat roads, it executes a rapid detection mode within a first speed range, adjusting the suspension arm to control the chassis height to the lowest possible level for high-speed scanning. For rugged mountains, soft sand, or muddy roads, it executes a fine detection mode within a second speed range, performing attitude recognition based on the IMU (Inertial Measurement Unit) and generating attitude control commands to control the suspension arm for attitude adjustment, ensuring the ground-penetrating radar is level and improving its transmission gain to compensate for signal attenuation. For underground utility tunnels or underground passages, it employs either the rapid detection mode or the fine detection mode. The first speed range is 20-30 km / h, and the second speed range is 5-10 km / h. In addition, in response to remote commands from the handheld terminal, detection is performed at a specified speed, a specified ground-penetrating radar angle, a specified chassis height, and a specified transmission gain.

[0017] The self-following ground-penetrating radar detection system and its control method described in this invention are composed of two parts: a handheld terminal and a self-following ground-penetrating radar device. The handheld terminal is responsible for real-time positioning of the operator and issuing detection tasks and parameters containing target areas, paths, points, or follow instructions to the device, while also receiving the processed detection results. The self-following ground-penetrating radar device is the core of the entire system and adopts a modular integrated architecture: the self-following mobile platform flexibly connects the drive mechanism to the chassis through a suspension swing arm and shock-absorbing damping components, giving the device all-terrain capability; the equipment mounting platform carries various control and detection units; the self-following control module integrates a positioning module, lidar, vision camera, IMU inertial measurement unit, and wheel speed encoder, fusing multimodal sensor data to uniformly complete path planning, follow control, and obstacle avoidance decisions, and actively adjusts the suspension attitude based on attitude recognition results to ensure driving stability in complex terrain; the ground-penetrating radar is installed at the bottom of the platform and is responsible for collecting underground echo data; the ground-penetrating control unit processes the radar data and high-precision positioning data together, transmits them to the ground station to generate radar profile maps, 3D imaging maps, and detection reports, and then sends them back to the handheld terminal. This invention integrates operator positioning, multimodal environmental perception, adaptive suspension control, and ground-penetrating data acquisition into a closed-loop system. It enables operators to perform natural following operations without having to push or pull the equipment. This system retains the flexibility of manual on-site intervention and key re-measurement, while maintaining reliable positioning and following capabilities in environments where satellite signals are denied, such as tunnels and utility tunnels. At the same time, it effectively suppresses vibration interference through active attitude adjustment, ensuring data acquisition quality and reducing the overall hardware cost of the system. It balances the efficiency of automated surveys with the flexibility of manual detailed surveys.

[0018] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the description, or may be learned by practice of the invention. The objects and other advantages of the invention can be realized and obtained by means of the structures specifically pointed out in the description and drawings.

[0019] Those skilled in the art will understand that the objectives and advantages achievable with the present invention are not limited to those specifically described above, and that the above and other objectives achievable with the present invention will become clearer from the following detailed description. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, are not intended to limit the scope of the invention. In the drawings: Figure 1 This is a schematic diagram of the structure of the self-following ground-penetrating radar device in a self-following ground-penetrating radar detection system according to an embodiment of the present invention.

[0021] Figure 2This is a schematic diagram of the equipment installation platform in the self-following ground-penetrating radar detection system according to another embodiment of the present invention.

[0022] Figure 3 This is a schematic diagram of the drive mechanism in a self-following ground-penetrating radar detection system according to another embodiment of the present invention.

[0023] Figure 4 This is a schematic diagram of the shock-absorbing connecting seat in the self-following ground-penetrating radar detection system according to another embodiment of the present invention.

[0024] Figure 5 This is a logical schematic diagram of the control method for a self-following bottom-penetrating radar detection system according to an embodiment of the present invention.

[0025] Figure 6 This is a schematic diagram of the hardware structure of the control method for the self-following bottom-penetrating radar detection system according to an embodiment of the present invention.

[0026] Figure 7 (a) is the operational status diagram for a horizontal road surface, (b) is the operational status diagram for a rugged road surface, (c) is the operational status diagram for a sloping road surface, and (d) is the operational status diagram for an obstacle crossing operation. Attached image description: 1: Drive mechanism; 2: Equipment mounting platform; 3: Ground penetrating radar; 4: Ground-penetrating control unit; 5: Vibration damping components; 6: Self-following control module; 21: Platform frame; 22: Equipment mounting rails; 23: Vibration damping connector; 24: Electrical interface; 25: Radar main unit mounting position; 26: Battery compartment; 27: Counterweight adjustment slot; 31: Traveling wheel; 32: Wheel hub motor; 33: Suspension control arm; 41: Primary damping layer; 42: Damping spring; 43: Hydraulic damper; 44: Guide rod; 45: Limiting buffer block; 46: Install the base at the second end. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and descriptions of this invention are used to explain the invention, but are not intended to limit the invention.

[0029] It should also be noted that, in order to avoid obscuring the invention with unnecessary details, only the structures and / or processing steps closely related to the solution according to the invention are shown in the accompanying drawings, while other details that are not closely related to the invention are omitted.

[0030] It should be emphasized that the term "including / comprises" as used herein refers to the presence of a feature, element, step, or component, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0031] It should also be noted that, unless otherwise specified, the term "connection" in this article can refer not only to a direct connection, but also to an indirect connection involving an intermediary.

[0032] In the following description, embodiments of the invention will be illustrated with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0033] Currently, there are two main types of automated detection solutions that combine mobile platforms with ground-penetrating radar (GPR): fully autonomous navigation solutions and deep learning-based intelligent scanning solutions. However, both types of solutions have insurmountable drawbacks in practical engineering applications. Fully autonomous navigation solutions rely on global navigation satellite systems or real-time dynamic positioning technology for positioning. In environments with satellite signal rejection or weak signals, such as tunnels, under viaducts, and underground utility tunnels, positioning drift is severe or even completely fails. Solutions using 3D lidar for simultaneous positioning and mapping are not only expensive in terms of hardware, but also prone to positioning loss in areas lacking geometric features, such as long straight corridors and open fields, leading to serious errors in the spatial repositioning of GPR data. More importantly, fully autonomous solutions completely exclude operators from the control loop. When encountering sudden obstacles or needing to re-survey suspected abnormal areas, operators cannot easily intervene directly and must replan the route using tablets or other terminals, which actually reduces operational efficiency. Deep learning-based intelligent scanning solutions go to the other extreme. They only focus on underground targets and automatically plan detection paths based on the targets. They rely entirely on pre-stored maps or real-time mapping. They cannot perceive the real-time position and movement intentions of operators, nor can they respond to natural interaction needs such as temporary changes in the detection area on site. They cannot achieve flexible collaboration between humans and machines.

[0034] In view of this, the present invention provides a self-following ground-penetrating radar detection system, the system comprising a handheld terminal and a self-following ground-penetrating radar device.

[0035] The handheld terminal is used to locate the operator, send detection tasks and detection task parameters marked with target area, target path, target point or follow command to the self-following ground-penetrating device, and receive ground-penetrating radar data.

[0036] Specifically, in some embodiments, the handheld terminal includes a second positioning module, an interactive display module, a processor unit, and a wireless communication module; the second positioning module uses a UWB positioning tag or an RTK satellite navigation module; the processor unit is used to load control programs to provide a preset interactive interface through the interactive display module for importing detection tasks and displaying radar profile maps, three-dimensional imaging maps, or detection reports; the wireless communication module uses a Bluetooth module, a WiFi module, a Zigbee module, or a data network communication module.

[0037] The handheld terminal is configured as the human-machine interface and operator positioning source for the entire detection system. The second positioning module is used for real-time positioning of the operator, specifically employing either a UWB positioning tag or an RTK satellite navigation module: in enclosed environments such as tunnels and underground utility tunnels where satellite signals are blocked, UWB positioning tags are prioritized, utilizing a pre-deployed UWB base station network to achieve high-precision indoor positioning; in open outdoor environments, an RTK satellite navigation module is used to obtain centimeter-level absolute position coordinates. The interactive display module performs both input and output functions. Operators can import detection tasks through a preset interactive interface and view the transmitted radar profile, 3D imaging, and detection reports in real time. The processor unit, as the terminal's computing core, carries a control program responsible for coordinating the work of each module, processing positioning data and task parameters, and driving the interactive display module to present a graphical user interface. The wireless communication module uses any one of Bluetooth, WiFi, Zigbee, or data network communication modules to establish a two-way data link with the self-following ground-penetrating radar device and ground station, enabling task distribution, positioning data upload, and detection result feedback.

[0038] Before the operation begins, the operator uses the interactive display module to define the detection area, set task parameters such as the survey line spacing and detection depth on a preset interactive interface. The processor unit encapsulates these parameters into detection tasks marked with target areas, target paths, target points, or follow instructions. During the operation, the second positioning module continuously acquires the operator's real-time position coordinates. The processor unit binds the positioning data to the detection task and sends it to the self-following ground-penetrating radar device via the wireless communication module, guiding it to autonomously follow the operator's trajectory. When the operator needs to temporarily change the detection area or conduct focused re-measurements on-site, they can directly mark new target points or partial scanning paths on the interactive interface. Task instructions are then sent to the self-following ground-penetrating radar device in real time via the wireless communication module. After data acquisition is complete, the radar profile, 3D imaging map, and detection report generated by the ground station are transmitted back to the handheld terminal via the wireless communication module. The processor unit drives the interactive display module to present these in a graphical format, allowing the operator to instantly confirm the detection quality on-site without returning to the control room to complete preliminary data interpretation and re-measurement decisions.

[0039] like Figure 1 As shown, the self-following ground-penetrating radar device includes: a self-following mobile platform, an equipment installation platform, a self-following control module, a ground-penetrating radar, a ground-penetrating control unit, and a power supply unit.

[0040] A self-following mobile platform includes a chassis and a drive mechanism, such as Figure 3 As shown, the drive mechanism is connected and fixed to the chassis via a suspension arm and a shock-absorbing damping component. In some embodiments, the chassis adopts a frame structure with multiple fixed points for mounting the shock-absorbing damping component; the shock-absorbing damping component is a hydraulic damper, a shock-absorbing spring, a polymer elastic damping component, a magnetorheological damper, or an electrorheological damper. The first end of the suspension arm is fixedly connected to the shock-absorbing damping component via a rotating connector and its angle is adjusted via a hydraulic rod; the second end of the suspension arm is connected to the drive mechanism, which is a wheeled, tracked, or wheel-tracked hybrid type. In some embodiments, the drive mechanism is a wheel-footed or leg-footed type.

[0041] The self-following mobile platform serves as the foundation for the entire device's movement, forming a hierarchical connection architecture encompassing the chassis, shock absorbers, swing arms, and drive system. The chassis is an integral frame structure with multiple standardized mounting points for installing and supporting the shock absorber damping components. These damping components connect the chassis and the suspension swing arms, absorbing and attenuating the impact energy transmitted from ground undulations and driving vibrations. Depending on cost and application scenarios, these damping components can be flexibly selected from passive forms such as hydraulic dampers, shock absorber springs, and polymer elastic dampers. Alternatively, faster-responding and more controllable magnetorheological dampers or electrorheological dampers can be selected. The first end of the suspension swing arm is fixedly connected to the shock absorber damping component via a rotating connector and is equipped with a hydraulic rod for active adjustment of the swing arm angle; the second end connects to the drive mechanism. The drive mechanism is the actuator that directly contacts the ground. It can adopt a traditional wheeled structure to obtain high mobility, or a tracked structure to enhance the ability to pass through soft terrain, or a wheel-tracked hybrid structure to achieve mode switching. In addition, in extremely rugged or obstacle-dense rescue scenarios, the drive mechanism can also be configured as wheel-legged or legged to obtain a more three-dimensional obstacle-crossing ability.

[0042] Once the self-following control module generates motion control commands, the drive mechanism operates at the specified speed and direction. During operation, impacts and vibrations from the ground are first transmitted to the suspension arms by the drive mechanism. The arms, constrained by the rotating connector, swing around their first end to buffer the impact. Simultaneously, the hydraulic rods can adjust their angles as needed to adapt to terrain undulations. Vibration energy is then transmitted to the damping components, which absorb the vibrations through passive dissipation or active adjustment, ensuring that residual vibrations transmitted to the chassis are effectively suppressed. This provides a stable operating environment for the ground-penetrating radar and control unit mounted on the equipment platform. When the self-following control module detects changes in attitude such as roll or pitch based on the IMU (Inertial Measurement Unit), it generates attitude control commands to drive the hydraulic rods to adjust the angle of the suspension arms. By changing the attitude of each arm, it compensates for vehicle tilt caused by the terrain, keeping the equipment platform level and ensuring a constant optimal detection attitude between the ground-penetrating radar antenna and the ground. The use of a multi-fixed-point chassis frame and replaceable damping components and drive mechanism modules significantly improves the overall terrain adaptability and maintainability of the machine.

[0043] The equipment mounting platform is positioned on top of the self-following mobile platform. In some embodiments, such as Figure 2 As shown, the equipment installation platform adopts a frame structure, with multiple shock-absorbing connecting seats and equipment installation rails. The equipment installation rails house a battery compartment, a counterweight adjustment slot, a self-following control module, and a ground-penetrating control unit. A radar host mounting position is located in the center of the equipment installation platform for mounting the ground-penetrating radar. The equipment installation platform is connected to a fixed self-following mobile platform via an active shock-absorbing mechanism.

[0044] In some embodiments, the active damping mechanism employs a two-stage damping structure, such as... Figure 4 As shown, it includes: a primary damping layer, which uses polymer gaskets to absorb high-frequency vibrations; a first-end mounting base, which is deployed on the primary damping layer; a limiting buffer block, which is fixedly connected to the first-end mounting base; and a hydraulic damper, which is fixedly connected to the limiting buffer block. The guide rod of the hydraulic damper is connected to and fixed to the second-end mounting base, and a damping spring is provided around the guide rod to provide support.

[0045] In the two-stage vibration damping structure, the first stage is a high-frequency vibration isolation layer, composed of polymer gaskets deployed at the bottom. Utilizing the viscoelastic properties of polymer materials, it first absorbs low-amplitude, high-frequency vibrations caused by road surface debris, high-frequency motor operation, etc., converting rigid transmission into flexible contact. The remaining mid-to-low frequency, high-amplitude vibrations are then transmitted to the second stage. The second stage is centered on a hydraulic damper, with its upper and lower ends fixed to the equipment mounting platform and chassis via second-end mounting bases and first-end mounting bases, respectively. When vibration causes relative displacement between the platform and chassis, the guide rod reciprocates within the constraint range of the limiting buffer block, forcing hydraulic oil to flow within the damper chamber, converting the vibration kinetic energy into heat dissipation. Simultaneously, the damping springs surrounding the guide rod compress or stretch synchronously to provide elastic support and passively adjust their stiffness according to the amplitude of road surface undulations, thereby significantly reducing the amplitude of the equipment mounting platform in the vertical and pitch directions, ensuring the ground-penetrating radar maintains a stable attitude and that data acquisition is not interfered with by driving vibrations during detection.

[0046] The self-following control module is mounted on the equipment mounting platform and connected to the drive mechanism. The module includes a first positioning module, lidar, vision camera, IMU (Inertial Measurement Unit), and wheel speed encoder to collect multimodal sensing data. Combining this data, it performs path planning, following control, or obstacle avoidance decisions for the detection task, generating motion control commands to control the drive mechanism's displacement. Based on the IMU, it performs attitude recognition and generates attitude control commands to adjust the attitude of the suspension arm. In some embodiments, the self-following control module is further used to plan paths for the target area to be detected using a bow-shaped reciprocating path algorithm, a cell decomposition full-coverage algorithm, a spiral path algorithm, an artificial potential field method, or a modified A algorithm; and to plan paths for target points using the Dijkstra algorithm. For following commands, it employs visual locking real-time following and historical trajectory delayed following control algorithms; and for obstacle avoidance decisions, it uses a dynamic window algorithm.

[0047] Specifically, the first positioning module acquires NMEA data frames containing timestamps, longitude, latitude, elevation, and positioning status identifiers, or a sequence of calculated UTM coordinates. The lidar outputs 3D point cloud data, specifically in the form of each frame containing a timestamp and several scan points, with each scan point consisting of an array of distance, angle, and reflection intensity values. The vision camera outputs a sequence of 2D image frames, specifically in the form of RGB or grayscale image matrices acquired at a fixed frame rate, with each frame accompanied by a timestamp and camera intrinsic parameter information. The IMU (Inertial Measurement Unit) outputs triaxial acceleration and triaxial angular velocity data from the self-following ground-penetrating radar device. The wheel speed encoder outputs the angular displacement increment of each drive wheel, specifically in the form of pulse counts of the left and right wheels per unit time or converted linear velocity values.

[0048] Furthermore, the self-following control module calls different algorithm models to fuse the aforementioned multimodal sensing data according to different types of detection tasks issued by the handheld terminal, and generates corresponding motion control commands.

[0049] For path planning tasks, when a detection task marks a target area or target point, the control module executes the path planning function, referring to... Figure 5 and Figure 6 The specific steps include: Step S101: Environmental mapping and localization initialization. The module mainly uses the 3D point cloud of LiDAR and supplements it with the image of the visual camera. It integrates the data of IMU and wheel speed encoder and constructs a 2D grid map or 3D point cloud map of the working area through the LiDAR synchronous localization and mapping algorithm. At the same time, the first localization module provides the initial global pose value and aligns the local map with the global coordinate system.

[0050] Step S102: Map preprocessing, binarizing the constructed map to distinguish between passable areas and obstacle areas.

[0051] Step S103: Path generation, selecting the appropriate algorithm based on task type. For full coverage detection of the target area, the target area boundary coordinates marked on the handheld terminal are read in. Within the passable area, a set of parallel detection lines is generated using a bow-shaped round-trip path algorithm, a cell decomposition full coverage algorithm, or a spiral path algorithm, ensuring that the spacing between the detection lines meets the detection density requirements.

[0052] For point-to-point navigation tasks targeting a specific location, a shortest reachable path is planned within the passable area using Dijkstra's algorithm or an improved A* algorithm, with the current location as the starting point and the target location coordinates as the ending point.

[0053] Step S104: Path publishing and execution. The planned path is discretized into a series of waypoint sequences and published point by point to the drive mechanism to execute displacement motion. During the motion, the wheel speed encoder provides mileage feedback, the IMU provides heading changes, and the first positioning module monitors the global position error to form a closed-loop correction.

[0054] For follow-control tasks, when the detection task is marked as a follow command, the control module executes the follow control function to maintain a spatial following relationship between the self-following ground-penetrating radar device and the operator. This function employs two modes, which can automatically switch according to environmental conditions.

[0055] The first method is real-time visual locking and tracking. A visual camera captures real-time image frames in front of the operator. A deep learning human detection model identifies a preset visual target pattern on the operator or handheld terminal, frames the target in the image, and calculates the target's azimuth and distance in the camera coordinate system. Using the operator's current position as the dynamic target point, the visually calculated azimuth and distance are used as control inputs. The control module uses a motion prediction model to calculate the required linear and angular velocities for tracking in real time, generating motion control commands. Simultaneously, it integrates lidar point cloud data to detect the actual distance between the operator and the self-following ground-penetrating radar device, and maintains a preset safe following distance through a PID controller.

[0056] The second method is delayed historical trajectory following. When the line of sight is obstructed, the system automatically switches from real-time visual lock-on following to delayed historical trajectory following. During real-time visual lock-on following, the self-following control module continuously records the real-time position coordinate sequence of the self-following ground-penetrating radar device output by the first positioning module. Simultaneously, it integrates the IMU and wheel speed encoder for inertial recursion, forming a complete historical trajectory line of the operator's movement. This trajectory line is stored as a standard target path, containing the coordinates, speed, and heading information of each trajectory point. When the visual signal is lost or the operator's instructions are changed, the control module locks this historical trajectory line as the target path, and the self-following ground-penetrating radar device autonomously travels along the trajectory point sequence. During travel, the first positioning module compares the lateral deviation between the current position and the trajectory points in real time, and corrects the steering angle through motion control commands to ensure the straightness of the survey line is reproduced. When the self-following ground-penetrating radar device reaches the end of the historical trajectory and the operator has not yet left the new trajectory, the control module controls the self-following ground-penetrating radar device to automatically decelerate and wait in place until a new trajectory update command is received.

[0057] The obstacle avoidance decision-making task runs continuously throughout the autonomous navigation and following control process, executing in parallel with the aforementioned tasks to ensure driving safety. The operation steps are as follows: Real-time reading of the current obstacle's position, size, and contour information perceived by the lidar and visual camera; simultaneously, acquiring the current speed and angular velocity from the wheel speed encoder and IMU; calculating the achievable speed range within the speed space, forming a dynamic window. Within the dynamic window, multiple candidate linear and angular velocity combinations are sampled at time steps, generating several short-term predicted trajectories through forward simulation. Each predicted trajectory is scored according to a preset cost function. The cost function comprehensively considers the following factors: 1) Obstacle distance cost: the closer the trajectory is to the nearest obstacle, the higher the cost. 2) Target orientation cost: the greater the deviation between the trajectory's end orientation and the target direction, the higher the cost. 3) Speed ​​cost: under safe conditions, the closer the speed is to the set cruise value, the lower the cost. The speed combination corresponding to the trajectory with the lowest cost is selected, converted into motion control commands, and sent to the drive mechanism for execution. This process cycles every 100 to 200 milliseconds, achieving real-time avoidance of dynamic obstacles.

[0058] In some embodiments, the self-following control module performs attitude recognition based on the IMU (Inertial Measurement Unit) and generates attitude control commands to control the suspension arm to adjust its attitude, including steps S201-S203: Step S201: The motion data of the equipment installation platform is collected in real time by the IMU inertial measurement unit. After error correction by performing Kalman filtering or particle filtering, the left and right tilt angles and front and back tilt angles of the equipment installation platform are calculated.

[0059] Step S202: Compare the left and right tilt angles and the front and back tilt angles with the horizontal target attitude, and calculate the correction deviation value.

[0060] Step S203: Calculate the target extension stroke of each suspension arm based on the correction deviation value and the structural parameters of the suspension arm, and generate attitude control commands to control the suspension arms to adjust their attitude to ensure that the equipment installation platform is level.

[0061] Ground-penetrating radar (GPR) is installed at the bottom of the equipment platform to collect GPR data. Specifically, the form and structure of the GPR can be flexibly configured according to the detection target and operational scenario. Regarding antenna polarization, a single-polarized dipole antenna, or a dual-polarized or multi-polarized antenna structure, can be used to obtain richer directional information about underground targets by switching the transmission and reception polarization directions. In terms of antenna arrangement, a single-channel structure with one transmitter and one receiver can be used, suitable for shallow detection and rapid surveys; a multi-channel array structure with one transmitter and multiple receivers or multiple transmitters and multiple receivers can also be used, covering a wider underground profile in a single scan, significantly improving operational efficiency and suitable for large-area road defect detection. Regarding antenna frequency configuration, a fixed single-frequency antenna can be used for optimized detection at a specific depth; a dual-frequency or multi-frequency combined antenna can be used, integrating high and low frequency units within the same antenna housing, simultaneously acquiring high-resolution shallow data and deep penetration data in a single pass; a stepped-frequency or frequency-modulated continuous wave system can also be used, dynamically adjusting the operating frequency band through software to adapt to different detection depth requirements. In terms of coupling method, radars are divided into two categories: air-coupled and ground-coupled. Air-coupled radar antennas are suspended at the bottom of the equipment mounting platform, maintaining a fixed distance from the ground, and are suitable for high-speed mobile detection. Ground-coupled radars use an adaptive mounting mechanism with a flexible floating connection, ensuring the antenna's bottom surface remains in close contact with the ground, making them suitable for fine-grained detection in rugged terrain. For structural protection, the antenna shell can be designed to be sealed and waterproof, and fitted with a high-polymer wear-resistant base plate to adapt to harsh conditions such as mud and sand. An attitude sensor can be embedded to monitor the antenna's attitude towards the ground in real time, providing feedback signals to the attitude-maintaining mechanism. In actual deployments, a combination of multi-frequency, multi-channel systems with air-coupled and ground-coupled methods is often used. The ground-penetrating control unit automatically switches operating modes according to mission instructions, enabling a single system to handle multiple detection scenarios.

[0062] The ground-penetrating control unit is mounted on the equipment installation platform and is used to collect and process ground-penetrating radar data. It is combined with the positioning data collected by the first positioning module and sent to the ground station to generate radar profile maps, three-dimensional imaging maps and detection reports, which are then transmitted back to the handheld terminal.

[0063] For example, the ground-penetrating control unit is installed in the instrument mounting rack on the equipment installation platform. It is electrically interconnected with the ground-penetrating radar, the first positioning module, the self-following control module and the power supply unit through a standardized interface. Its internal structure adopts a modular board architecture, consisting of six parts: radar signal acquisition board, positioning synchronization board, edge computing board, data storage board, communication interface board and synchronization trigger management module. The radar signal acquisition board connects to the ground-penetrating radar antenna via a high-frequency coaxial cable. It incorporates a high-speed analog-to-digital converter to sample and quantize the radar echo signal, supporting multi-channel parallel acquisition. The positioning synchronization board receives real-time position data from the first positioning module via a serial interface, and a unified hardware synchronization pulse is generated by the synchronization trigger management module. Each radar data point is timestamped with its positioning coordinates to achieve a synchronization accuracy better than 1ms. The edge computing board, equipped with an embedded processor, runs preprocessing algorithms such as real-time filtering, background removal, and gain control, converting the raw radar data into signal-to-noise ratio optimized processed data. Simultaneously, it dynamically adjusts the radar transmission parameters based on the driving speed and terrain features fed back by the self-following control module. The data storage board is equipped with solid-state storage media, providing local dual backups of both raw and processed data. The communication interface board integrates a gigabit Ethernet interface and a wireless communication module, packaging and sending the radar data with bound position information to the ground station via wired or wireless links. It also receives control commands and parameter configurations forwarded from the handheld terminal via the ground station. All boards are interconnected via a high-speed backplane bus. The entire board employs a low-power design and is supplied with a unified regulated power supply by a power supply unit, meeting the reliability requirements of long-term field operations.

[0064] The power supply unit is used to power the self-following ground-penetrating radar device.

[0065] For example, the power supply unit is deployed in the lower equipment enclosure of the equipment installation platform, adopting a three-tier architecture including a main battery pack, secondary power distribution, and intelligent management. The main battery pack consists of several high-energy-density lithium-ion battery modules connected in series and parallel, with a rated output voltage of 48V and a total capacity of not less than 2kWh, supporting continuous operation for more than 4 hours. The battery modules are encapsulated in an aluminum alloy sealed chamber with heat dissipation fins and shock-absorbing pads, meeting the IP67 protection level. The secondary power distribution module converts the 48V DC bus into 48V power for the drive mechanism, 12V / 5V regulated power for the control and sensing modules, and an independent high-voltage pulse power for the ground-penetrating radar transmitter. Each output circuit is independently equipped with overcurrent protection and filtering isolation circuits. The battery management system (BMS) monitors the voltage, temperature, and charging / discharging current of each cell in real time, communicates with the self-following control module via the CAN bus, sends the remaining power, estimated range, and fault alarms to the handheld terminal, and automatically reduces the transmission power or prompts the operator to return when the battery is low. The charging interface uses a magnetic waterproof quick-connect connector, which supports external chargers or photovoltaic supplementary charging, further enhancing the continuous support capability for field operations.

[0066] On the other hand, the present invention also provides a control method for a self-following ground-penetrating radar detection system, the method being executed by a self-following ground-penetrating radar device in the aforementioned self-following ground-penetrating radar detection system, the method comprising steps S301 to S304: Step S301: Receive the detection task and detection task parameters sent by the handheld terminal. The detection task is used to mark the target area, target path, target location or record the follow instruction.

[0067] Step S302: For detection tasks involving marking target areas, target paths, or target points, perform path planning, collect the current position through the first positioning module, and autonomously drive according to the planned path; for detection tasks involving recording follow instructions, acquire the operator's positioning data sent by the handheld terminal in real time, record and form the operator's movement path, collect the current position through the first positioning module, and autonomously drive according to the operator's movement path; perform obstacle detection and dynamic obstacle avoidance based on the point cloud data collected by the lidar and the environmental images collected by the visual camera.

[0068] Step S303: Based on the detection mission parameters, control the ground penetrating radar to emit electromagnetic waves and receive echoes during autonomous driving, and simultaneously collect radar data and its own current position information.

[0069] Step S304: After real-time filtering, gain adjustment and background removal of radar data based on edge computing, the data is sent to the ground station in conjunction with its current location information to generate radar profile map, three-dimensional imaging map and detection report and then sent back to the handheld terminal.

[0070] In some embodiments, such as Figure 7 As shown in (a) to 7(d), the method further includes: The system receives terrain feedback data from a handheld terminal. For flat roads, it executes a rapid detection mode within a first speed range, adjusting the suspension arm to control the chassis height to the lowest possible level for high-speed scanning. For rugged mountains, soft sand, or muddy roads, it executes a fine detection mode within a second speed range, using an IMU (Inertial Measurement Unit) for attitude recognition and generating attitude control commands to adjust the suspension arm's attitude, ensuring the ground-penetrating radar is level and improving its transmission gain to compensate for signal attenuation. For underground utility tunnels or underground passages, it employs either a rapid detection mode or a fine detection mode. The first speed range is 20-30 km / h, and the second speed range is 5-10 km / h.

[0071] In addition, in response to remote commands from a handheld terminal, it can perform detection at a specified speed, a specified ground-penetrating radar angle, a specified chassis height, and a specified transmission gain.

[0072] Specifically, this invention can be adapted to exploration operations under various complex terrain conditions: Operation on flat roads: Using the rapid movement mode, the chassis height is adjusted to the lowest level, and the driving speed is 20~30km / h to achieve high-efficiency large-area scanning.

[0073] Operation in rugged mountainous terrain: Employing a fine detection mode or a hybrid mode, the chassis height is adjusted in real time according to the terrain, the attitude control mechanism ensures the radar is level, and the active damping system suppresses vibration, achieving stable detection in complex terrain.

[0074] For operations on soft sandy / muddy roads: a fine detection mode is adopted to increase the grounding area and reduce the grounding specific voltage to avoid vehicles getting stuck; the detection control module automatically increases the transmission gain to compensate for signal attenuation.

[0075] Underground utility tunnel / tunnel operations: Employ a rapid movement mode, low-speed fine detection (5~10km / h), and use lidar to build a 3D map in real time to achieve accurate navigation in the absence of GPS.

[0076] Disaster site operations: Operators can remotely control the system from a safe area, and the system will autonomously complete the detection tasks of dangerous areas such as ruins and landslides, ensuring personnel safety.

[0077] The present invention has the following significant beneficial effects: (1) Significantly improved work efficiency: The self-following mobile platform can work continuously for more than 4 hours with a range of up to 50km. Compared with manual hand-held operation, the efficiency is increased by 5 to 10 times. It is not limited by the physical strength of the operator and can realize rapid detection operations over large areas and long distances.

[0078] (2) Significantly enhanced terrain adaptability: The self-following mobile platform has a 300mm obstacle crossing capability and a ±280mm chassis height adjustment function, which can effectively operate in complex terrain (rugged mountains, soft sand, muddy roads), narrow spaces (underground pipe corridors, tunnel interiors) and unstructured environments (construction sites, disaster sites), greatly expanding the application scenarios of ground penetrating radar.

[0079] (3) Effective guarantee of detection accuracy: The self-following mobile platform operates at a uniform and controllable speed (0~30km / h stepless speed regulation), combined with centimeter-level RTK positioning accuracy and millisecond-level data synchronization, ensuring the accuracy of the detection trajectory and data quality; the active damping system and attitude maintenance mechanism effectively suppress the impact of vibration and attitude changes on detection.

[0080] (4) Achieve fully automated operation: Through preset detection paths and intelligent control algorithms, the system can automatically complete detection tasks, including automatic navigation, obstacle avoidance, path correction, parameter adjustment and other functions, reducing manual intervention, operation difficulty and labor costs.

[0081] (5) Precise synchronization of data and location: The hardware synchronization triggering technology is adopted to achieve precise synchronization of radar data acquisition and RTK location information (synchronization accuracy better than 1ms), eliminating the need for manual matching in the later stage, thus improving data processing efficiency and reliability of results.

[0082] (6) Safe operation in harsh environments: The system has an IP67 protection level and an operating temperature range of -20℃ to +50℃. It can operate safely in harsh environments such as high temperature, high humidity, toxic and harmful substances, and radioactivity, protecting personnel safety and expanding the application boundaries of ground penetrating radar.

[0083] (7) Adaptive detection capability: The system can adaptively adjust the radar transmission parameters (transmission frequency, pulse repetition frequency, gain, etc.) according to driving speed, terrain conditions, and geological characteristics to optimize the detection effect; the attitude maintenance mechanism ensures that the radar antenna is always in the best detection attitude.

[0084] (8) The equipment is effectively protected: the active damping system effectively absorbs driving vibrations, protects the ground-penetrating radar device from impact damage, extends the service life of the equipment, and reduces maintenance costs.

[0085] In summary, the self-following ground-penetrating radar detection system and its control method described in this invention are composed of two parts: a handheld terminal and a self-following ground-penetrating radar device. The handheld terminal is responsible for real-time positioning of the operator and issuing detection tasks and parameters containing target areas, paths, points, or follow instructions to the device, while simultaneously receiving the processed detection results. The self-following ground-penetrating radar device is the core of the entire system, adopting a modular integrated architecture: the self-following mobile platform flexibly connects the drive mechanism to the chassis through a suspension swing arm and shock-absorbing damping components, giving the device all-terrain capability; the equipment mounting platform carries various control and detection units; the self-following control module integrates a positioning module, lidar, visual camera, IMU inertial measurement unit, and wheel speed encoder, fusing multimodal sensor data to uniformly complete path planning, follow control, and obstacle avoidance decisions, and actively adjusts the suspension attitude based on attitude recognition results to ensure driving stability in complex terrain; the ground-penetrating radar is installed at the bottom of the platform and is responsible for collecting underground echo data; the ground-penetrating control unit processes the radar data and high-precision positioning data together, transmits them to the ground station to generate radar profile maps, 3D imaging maps, and detection reports, and then sends them back to the handheld terminal. This invention integrates operator positioning, multimodal environmental perception, adaptive suspension control, and ground-penetrating data acquisition into a closed-loop system. It enables operators to perform natural following operations without having to push or pull the equipment. This system retains the flexibility of manual on-site intervention and key re-measurement, while maintaining reliable positioning and following capabilities in environments where satellite signals are denied, such as tunnels and utility tunnels. At the same time, it effectively suppresses vibration interference through active attitude adjustment, ensuring data acquisition quality and reducing the overall hardware cost of the system. It balances the efficiency of automated surveys with the flexibility of manual detailed surveys.

[0086] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave.

[0087] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0088] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0089] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A self-following ground-penetrating radar detection system, characterized in that, The system includes a handheld terminal and a self-following ground-penetrating radar device; The handheld terminal is used to locate the operator, send the detection task and detection task parameters marked with the target area, target path, target point or follow command to the self-following ground-penetrating device, and receive ground-penetrating radar data; The self-following ground-penetrating radar device includes: The self-following mobile platform includes a walking chassis and a drive mechanism, wherein the drive mechanism is connected and fixed to the walking chassis via a suspension swing arm and a shock-absorbing damping component. The equipment installation platform is located on top of the self-following mobile platform; A self-following control module is mounted on the equipment mounting platform and connected to the drive mechanism. The self-following control module is equipped with a first positioning module, a lidar, a vision camera, an IMU (Inertial Measurement Unit), and a wheel speed encoder to collect multimodal sensing data. Combining this multimodal sensing data, it performs path planning, following control, or obstacle avoidance decisions for the detection task, generating motion control commands to control the drive mechanism for displacement movement. Based on the IMU, it performs attitude recognition and generates attitude control commands to control the suspension arm for attitude adjustment. Ground-penetrating radar is installed at the bottom of the equipment mounting platform to collect ground-penetrating radar data; The ground-penetrating control unit, mounted on the equipment mounting platform, is used to collect and process the ground-penetrating radar data, and send it to the ground station in conjunction with the positioning data collected by the first positioning module to generate radar profile, three-dimensional imaging map and detection report and send them back to the handheld terminal; A power supply unit is used to supply power to the self-following ground-penetrating radar device.

2. The self-following ground-penetrating radar detection system according to claim 1, characterized in that, The handheld terminal includes a second positioning module, an interactive display module, a processor unit, and a wireless communication module; the second positioning module uses a UWB positioning tag or an RTK satellite navigation module; the processor unit is used to load a control program to provide a preset interactive interface through the interactive display module for importing detection tasks and displaying the radar profile, the three-dimensional imaging map, or the detection report; the wireless communication module uses a Bluetooth module, a WiFi module, a Zigbee module, or a data network communication module.

3. The self-following ground-penetrating radar detection system according to claim 1, characterized in that, The chassis adopts a frame structure and has multiple fixed points for mounting the shock-absorbing damping components; the shock-absorbing damping components are hydraulic dampers, shock-absorbing springs, polymer elastic dampers, magnetorheological dampers, or electrorheological dampers. The first end of the suspension arm is fixedly connected to the shock absorber by a rotating connector and the angle is adjusted by a hydraulic rod; the second end of the suspension arm is connected to the drive mechanism, which is a wheeled, tracked, or wheel-tracked hybrid type.

4. The self-following ground-penetrating radar detection system according to claim 1, characterized in that, The drive mechanism is either wheel-footed or leg-footed.

5. The self-following ground-penetrating radar detection system according to claim 1, characterized in that, The equipment installation platform adopts a frame structure and is equipped with multiple shock-absorbing connecting seats and equipment installation rails. The equipment installation rails are equipped with a battery compartment, a counterweight adjustment slot, the self-following control module, and a ground-penetrating control unit. The equipment installation platform has a radar host installation position in the center for mounting the ground-penetrating radar. The equipment installation platform is connected and fixed to the self-following mobile platform through an active shock absorption mechanism.

6. The self-following ground-penetrating radar detection system according to claim 5, characterized in that, The active damping mechanism adopts a two-stage damping structure, including: The primary damping layer uses polymer gaskets to absorb high-frequency vibrations; The first end is mounted on the base and deployed on the primary damping layer; A limiting buffer block is fixedly connected to the first end mounting base; A hydraulic damper is fixedly connected to the limiting buffer block. The guide rod of the hydraulic damper is connected to and fixed to the second end mounting base. A shock-absorbing spring is provided around the guide rod to provide support.

7. The self-following ground-penetrating radar detection system according to claim 1, characterized in that, The self-following control module is also used to plan paths for the target area to be detected using the bow-shaped round-trip path algorithm, the cell decomposition full coverage algorithm, the spiral path algorithm, the artificial potential field method, or the improved A algorithm; to plan paths for the target points using the Dijkstra algorithm; to execute the follow command using the visual locking real-time follow and historical trajectory delayed follow control algorithms; and to execute the obstacle avoidance decision using the dynamic window algorithm.

8. The self-following ground-penetrating radar detection system according to claim 1, characterized in that, The self-following control module performs attitude recognition based on the IMU (Inertial Measurement Unit) and generates attitude control commands to control the suspension arm to adjust its attitude, including: The motion data of the equipment installation platform is collected in real time by the IMU inertial measurement unit. After error correction by performing Kalman filtering or particle filtering, the left and right tilt angles and front and back tilt angles of the equipment installation platform are calculated. The left and right tilt angles and the front and back tilt angles are compared with the horizontal target attitude to calculate the correction deviation value; Based on the corrected deviation value and the structural parameters of the suspension arms, the target extension stroke of each suspension arm is calculated, and the attitude control command is generated to control the suspension arms to adjust their attitude to ensure that the equipment installation platform is level.

9. A control method for a self-following bottom-penetrating radar detection system, characterized in that, The method is used to be executed in a self-following ground-penetrating radar device in any one of the self-following ground-penetrating radar detection systems according to any one of claims 1 to 8, and the method includes: Receive detection tasks and detection task parameters sent by the handheld terminal. The detection tasks are used to mark the target area, target path, target location, or record follow instructions. For the detection task that marks the target area, target path, or target point, path planning is performed. The current position is collected by the first positioning module, and the vehicle moves autonomously according to the planned path. For the detection task that records follow instructions, the operator's positioning data sent by the handheld terminal is acquired in real time, and the operator's movement path is recorded. The current position is collected by the first positioning module, and the vehicle moves autonomously according to the operator's movement path. Obstacle detection and dynamic obstacle avoidance are performed based on point cloud data collected by lidar and environmental images collected by visual cameras. Based on the detection mission parameters, the ground-penetrating radar is controlled to emit electromagnetic waves and receive echoes during autonomous driving, while simultaneously collecting radar data and its own current position information. The radar data is filtered, gain adjusted, and background removed in real time using edge computing. Then, it is combined with the user's current location information and sent to the ground station to generate a radar profile, a 3D imaging map, and a detection report, which are then transmitted back to the handheld terminal.

10. The control method for the self-following bottom-penetrating radar detection system according to claim 9, characterized in that, The method further includes: The system receives terrain feedback data sent by the handheld terminal. For flat roads, it executes a rapid detection mode within a first speed range, adjusting the suspension arm to control the chassis height to the lowest possible level for high-speed scanning. For rugged mountains, soft sand, or muddy roads, it executes a fine detection mode within a second speed range, performing attitude recognition based on the IMU (Inertial Measurement Unit) and generating attitude control commands to control the suspension arm for attitude adjustment, ensuring the ground-penetrating radar is level and improving its transmission gain to compensate for signal attenuation. For underground utility tunnels or underground passages, it employs either the rapid detection mode or the fine detection mode. The first speed range is 20-30 km / h, and the second speed range is 5-10 km / h. In addition, in response to remote commands from the handheld terminal, detection is performed at a specified speed, a specified ground-penetrating radar angle, a specified chassis height, and a specified transmission gain.

Images