Coordinated aerial and ground system for targeted detection and localization of buried objects using radar
Patent Information
- Application Number
- IN202621022044
- Authority / Receiving Office
- IN · IN
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2026-02-24
- Publication Date
- 2026-10-26
- Estimated Expiration
- 2046-02-24
AI Technical Summary
Conventional buried object detection techniques face challenges such as high risk to human operators, inefficient coverage of large or complex terrains, excessive power consumption, and low detection accuracy due to the lack of integration between aerial surveillance and ground-based sensing, leading to fragmented workflows and reliance on human decision-making in hazardous environments.
A coordinated buried object detection system comprising an unmanned ground vehicle with a ground penetrating radar and an unmanned aerial vehicle that integrates aerial surveillance with ground navigation, using a control station to process aerial data for targeted subsurface scanning, reducing human exposure by focusing radar transmission only on identified regions of interest.
Enables accurate, efficient, and safe detection of buried objects by minimizing direct human exposure to hazardous environments, improving detection accuracy, and reducing power consumption through selective scanning.
Abstract
Description
TECHNICAL FIELD
[0001] The present invention relates to the field of a system and method for coordinated buried object detection.BACKGROUND
[0002] Detection and localization of buried objects such as landmines, unexploded ordnance, underground utilities, and concealed hazardous materials is a critical technical challenge in applications including defence, disaster response, infrastructure inspection, and environmental surveying. Conventional buried object detection techniques typically rely on manual probing, handheld ground penetrating radar equipment, or ground-based robotic platforms that perform continuous or pre-programmed subsurface scanning without contextual awareness. Such approaches suffer from several technical limitations, including high risk to human operators, inefficient coverage of large or complex terrains, excessive power consumption due to indiscriminate scanning, and a high incidence of false positives caused by subsurface clutter such as rocks, roots, or soil heterogeneity. Aerial surveillance systems, while capable of rapidly assessing surface conditions over wide areas, are generally limited to visual inspection and lack the capability to directly detect subsurface objects, and existing systems do not effectively integrate aerial observations with targeted ground-based sensing in a coordinated manner. Furthermore, known solutions fail to provide an automated mechanism by which aerially derived information can be processed and translated into precise ground navigation and selective subsurface scanning actions, resulting in fragmented workflows, reduced detection accuracy, and continued dependence on human decision-making in hazardous environments. Accordingly, there exists a technical need for a coordinated detection system that integrates aerial surveillance, intelligent data processing, and targeted ground-based subsurface sensing to enable accurate, efficient, and safe detection of buried objects.SUMMARY
[0003] The present invention relates to a coordinated buried object detection system comprising an unmanned ground vehicle equipped with a locomotion assembly and a ground penetrating radar module, an unmanned aerialvehicle having an imaging unit, a communication interface, and a control station. The unmanned aerial vehicle acquires aerial surveillance data of a target area using a downward-facing imaging unit and transmits the acquired data to the control station through a wireless communication link. The control station processes the aerial data to generate a grid-based coordinate representation of the target area and identifies one or more regions of interest. Based on the grid coordinates, the control station computes real-world distance and direction values and generates navigation parameters for the unmanned ground vehicle. The navigation parameters are transmitted to the unmanned ground vehicle, which navigates toward the selected region of interest using a multi-legged locomotion assembly adapted for stable traversal over uneven terrain.
[0004] Upon confirmation that the unmanned ground vehicle has reached the region of interest, the control station selectively activates the ground penetrating radar module to perform subsurface scanning. The radar module generates continuous-wave radio frequency signals using an oscillator coupled to a tunable resonant circuit and alternately switches a shared antenna between transmission and reception modes. Reflected electromagnetic signals from subsurface objects are processed to extract depth-related information, converted into digital data, and transmitted to the control station for visualization and interpretation. The control station correlates the subsurface scan results with the aerially derived grid coordinates to determine the presence and location of buried objects. The coordinated operation between aerial surveillance, directed ground navigation, and condition-based subsurface scanning enables focused detection while reducing direct human exposure in hazardous environments such as mine-affected areas, disaster zones, and unstable terrain. BRIEF DESCRIPTION OF DRAWINGS
[0005] FIG. 1 illustrates a block diagram for a system for coordinated buried object detection, in accordance with an embodiment of the present disclosure; FIG. 2 illustrates a flow chart for a method for coordinated buried object detection, in accordance with the present disclosure; FIG. 3 illustrates a block diagram of an integrated unmanned aerial vehicle and unmanned ground vehicle-based buried object detection system, in accordance with the present disclosure; FIG. 4 illustrates the signal flow and functional components of the GPR module, in accordance with the present disclosure; FIG. 5 illustrates a block schematic of the control station, in accordance with the present disclosure.DETAILED DESCRIPTION
[0006] The present invention relates to a coordinated buried object detection system configured to detect and localize subsurface objects while minimizing human exposure to hazardous environments. In the best mode of working, the system comprises an unmanned aerial vehicle, an unmanned ground vehicle equipped with a ground penetrating radar, a communication interface, and a control station that collectively enable aerial reconnaissance-driven, targeted subsurface scanning.
[0007] In operation, the system functions in a sequential manner beginning with aerial surveillance of a target area, followed by ground-based navigation and subsurface scanning at selected locations. The aerial and ground platforms are configured to exchange information bidirectionally through a wireless communication link, while the control station serves as a centralized processing and coordination unit. The system is particularly suited for use in hazardous or inaccessible environments such as minefields, disaster zones, construction sites, or underground utility corridors, where direct human involvement poses safety risks.
[0008] The unmanned ground vehicle is implemented as a mobile robotic platform designed to traverse uneven terrain with stability and precision. In the best mode, the ground vehicle employs a hexapod locomotion configuration comprising six articulated legs, each driven by electric actuators such as servo motors or brushless DC motors. This configuration distributes ground contact forces across multiple points, thereby reducing localized pressure and enhancing stability on loose soil, rubble, or irregular surfaces.
[0009] Movement of the ground vehicle is controlled through embedded processors executing gait generation and inverse kinematics algorithms. These algorithms coordinate the motion of individual legs to achieve controlled forward movement, turning, and positioning while maintaining balance. Sensors such asgyroscopes and accelerometers provide real-time feedback on orientation and movement, enabling closed-loop control of locomotion.
[0010] Mounted on the ground vehicle is a ground penetrating radar module configured to detect subsurface objects by transmitting electromagnetic waves into the ground and receiving reflected signals. In the best mode, the radar module includes a radio frequency signal generator implemented using an oscillator-based circuit. A Colpitts oscillator coupled with a tunable LC tank circuit is used to generate a continuous-wave radio frequency signal, allowing the operating frequency to be adjusted to suit different soil conditions and target object characteristics.
[0011] The radar module employs a shared antenna element connected through a switching arrangement that alternates the antenna between transmission and reception modes. During operation, the transmitted signal penetrates the subsurface and interacts with buried objects. Reflected signals are received by the antenna, mixed with a reference oscillator signal using a mixer circuit, and processed to extract information related to depth and material properties. The processed analog signals are subsequently converted into digital data using an analog-to-digital converter.
[0012] The unmanned aerial vehicle is implemented as a multirotor or fixed-wing platform equipped with a downward-facing imaging unit. The imaging unit may comprise an optical camera and, optionally, additional sensors such as thermal or multispectral cameras. During operation, the aerial vehicle performs a surveillance flight over the target area, capturing images of the terrain from above.
[0013] The aerial vehicle may include positioning systems such as GPS to geo-tag the acquired imagery. The captured images are transmitted wirelessly to the control station for further analysis. Aerial surveillance enables rapid assessment of large areas and provides contextual information that cannot be obtained from ground-based sensing alone.
[0014] The control station operates as the central coordination and processing unit of the system. In the best mode, the control station is implemented using a computing device such as an industrial computer or portable workstationexecuting dedicated control and analysis software. The control station receives aerial surveillance data from the aerial vehicle and processes the data to identify regions within the target area that are likely to contain buried objects.
[0015] Aerial images are converted into a grid-based coordinate representation by dividing each image into discrete grid cells. Pixel coordinates corresponding to selected grid cells are mathematically transformed into real-world distances and directions using known camera parameters, aerial vehicle altitude, and positioning data. Based on this transformation, navigation parameters are generated to guide the ground vehicle toward selected regions of interest.
[0016] In the best mode, artificial intelligence techniques are employed to enhance the identification of regions of interest and interpretation of subsurface data. Machine learning models such as convolutional neural networks are trained using aerial imagery datasets containing labeled examples of terrain features associated with buried objects, including disturbed soil patterns, excavation marks, thermal anomalies, or vegetation irregularities. Training datasets may be obtained from controlled test sites, historical survey data, or simulated environments.
[0017] The trained models analyze incoming aerial imagery and generate probability scores or confidence maps indicating the likelihood of buried objects at different locations. These outputs are used by the control station to prioritize regions for ground-based investigation.
[0018] Similarly, machine learning models may be trained on labeled ground penetrating radar datasets to distinguish true buried objects from natural subsurface clutter such as rocks or roots. These models assist in classifying reflected radar signals and improving detection accuracy.
[0019] Once navigation parameters are generated, the ground vehicle is guided toward a selected region of interest. Motion vectors are calculated by comparing target coordinates derived from aerial analysis with real-time position and orientation data obtained from onboard sensors. Locomotion commands are generated through successive coordinate subtraction and executed by the hexapod locomotion system to position the radar module precisely over the target location.
[0020] Subsurface scanning is initiated only after the ground vehicle reaches the designated target location. This selective activation ensures that radar transmission is focused on specific regions identified through aerial analysis rather than performing continuous or indiscriminate scanning across the entire area. Such targeted operation reduces power consumption, minimizes false detections, and enhances overall system efficiency.
[0021] Reflected radar signals obtained during subsurface scanning are digitized and transmitted to the control station for further processing. Signal processing techniques such as filtering, time-domain analysis, and frequency-domain analysis are applied to generate interpretable representations of subsurface structures. The processed data may be visualized as radar images or overlaid on aerial maps, enabling operators to understand the location and characteristics of detected buried objects.
[0022] The combined use of aerial imagery, ground-based radar sensing, artificial intelligence-based analysis, and coordinated vehicle operation enables accurate detection and localization of buried objects while significantly reducing the need for direct human involvement in hazardous environments.
[0023] FIG. 1 illustrates a block diagram for a system for coordinated buried object detection, in accordance with an embodiment of the present disclosure. Reference numerals used in figures. 100 - system; 102 - unmanned ground vehicle (UGV); 102a - locomotion assembly; 102b - ground penetrating radar (GPR) module; 104 - unmanned aerial vehicle (UAV); 104a - imaging unit; 106 - communication interface; 108 - control station; 200 - method.
[0024] The system (100) comprises an unmanned ground vehicle (UGV) (102), an unmanned aerial vehicle (UAV) (104), a communication interface (106), and a control station (108).
[0025] The unmanned ground vehicle (UGV) (102) comprises a locomotion assembly (102a) and a ground penetrating radar (GPR) module (102b) configured to transmit electromagnetic waves into a subsurface region and receive reflected signals from buried objects.
[0026] In the implementation of the embodiment, the locomotion assembly (102a) of the UGV (102) comprises a hexapod configuration adapted to distribute ground contact forces across multiple legs for stable movement over uneven terrain.
[0027] In the implementation of the embodiment, the GPR module (102b) comprises a radio frequency oscillator-based signal generator configured to operate within a frequency range suitable for detecting metallic and non-metallic buried objects, and wherein the radio frequency oscillator comprises a Colpitts oscillator configured to generate a continuous-wave radar signal.
[0028] In the implementation of the embodiment, the ground penetrating radar module comprises a tunable LC tank circuit and a switching arrangement.
[0029] In the implementation of the embodiment, the tunable LC tank circuit is coupled to a Colpitts oscillator, the oscillator being configured to generate a continuous-wave radio frequency signal.
[0030] In the implementation of the embodiment, the switching arrangement is configured to alternately connect a shared antenna element between a transmission mode and a reception mode, wherein reflected subsurface signals are mixed with a reference oscillator signal to extract depth-related information prior to digitization.
[0031] The unmanned aerial vehicle (UAV) (104) comprises an imaging unit (104a) configured to acquire aerial surveillance data of a target area.
[0032] In the implementation of the embodiment, the UAV (104) imaging unit (104a) comprises a downward-facing camera, and the control station (108) is configured to convert the acquired aerial image into a grid-based coordinate system.
[0033] In the implementation of the embodiment, the grid-based coordinate system is used to determine a real-world distance and direction for guiding the movement of the UGV (102) toward a selected region of interest.
[0034] The communication interface (106) is configured to enable bidirectional data exchange between the UAV (104) and the UGV (102).
[0035] In the implementation of the embodiment, the communication interface (106) comprises a wireless radio frequency link enabling real-time transmission of navigation data, sensing data, and operational status information.
[0036] The control station (108) is configured to process the aerial surveillance data to identify regions of interest and generate navigation parameters for the UGV (102).
[0037] In the implementation of the embodiment, the control station (108) is configured to initiate subsurface scanning by the ground penetrating radar only upon confirmation that the unmanned ground vehicle has reached a target location derived from aerial surveillance data, such that radar transmission is selectively activated based on aerially identified regions of interest rather than continuous ground scanning.
[0038] The control station (108) is configured to receive and process subsurface scan data from the GPR module (102b) to determine the presence and location of buried objects.
[0039] The coordinated operation of the UAV (104) and the UGV (102) enables detection and localization of buried objects while reducing direct human exposure to hazardous environments.
[0040] In the implementation of the embodiment, the unmanned ground vehicle (102) comprises a hexapod locomotion assembly configured to traverse uneven terrain while supporting a payload comprising the ground penetrating radar module (102b), a downward-facing thermal sensor, and a positioning module comprising a real-time kinematic global positioning system (RTK-GPS) receiver.
[0041] In the implementation of the embodiment, the unmanned aerial vehicle (104) comprises a multi-rotor aerial platform carrying a dual-axis gimbal, the dual-axis gimbal mounting a high-resolution visual camera and a radiometric thermal camera for simultaneous visible-spectrum and thermal imaging of a target area.
[0042] In the implementation of the embodiment, the communication interface (106) comprises a wireless mesh communication network or a long-range low-power radio link configured to transmit command signals and multi-modal sensor data between the control station (108), the unmanned aerial vehicle (104), and the unmanned ground vehicle (102).
[0043] In the implementation of the embodiment, the unmanned ground vehicle (102) further comprises at least one obstacle detection sensor selected from a LiDAR sensor and an ultrasonic ranging sensor, the obstacle detection sensor being configured to provide terrain or obstacle proximity data for autonomous navigation.
[0044] In the implementation of the embodiment, the control station (108) is configured to execute photogrammetry processing to generate a geo-referenced orthomosaic map of the target area from aerial images acquired by the imaging unit (104a).
[0045] In the implementation of the embodiment, the control station (108) is configured to geo-register ground penetrating radar data, thermal sensor data, and aerial image data into a common spatial coordinate frame.
[0046] In the implementation of the embodiment, the control station (108) is configured to compute a probabilistic threat score for a detected subsurface anomaly based on correlation between ground penetrating radar response, thermal response, and surface visual features.
[0047] In the implementation of the embodiment, the unmanned aerial vehicle (104) is configured to loiter above the unmanned ground vehicle (102) during ground traversal to operate as a communication relay and to perform forward-looking surface or thermal scanning of a projected path of the unmanned ground vehicle (102).
[0048] In the implementation of the embodiment, the control station (108) is configured to plot detected subsurface anomalies as geo-tagged markers on an aerial map and display the markers in real time during movement of the unmanned ground vehicle (102).
[0049] In one exemplary embodiment, the unmanned ground vehicle (UGV) (102) comprises a locomotion assembly (102a) and a ground penetrating radar (GPR) module (102b) mounted on a forward support frame. The unmanned aerial vehicle (UAV) (104) comprises an imaging unit (104a) configured to acquire aerial surveillance data of a target area. A communication interface (106) establishes bidirectional data exchange between the UGV (102) and the UAV (104).A control station (108) implemented as a laptop computer is configured to receive aerial surveillance data from the UAV (104), process the aerial surveillance data to identify regions of interest and generate navigation parameters for the UGV (102), and receive subsurface scan data from the GPR module (102b) to determine presence and location of buried objects.
[0050] In another exemplary embodiment, the control station (108) is implemented as a handheld mobile device selected from a smartphone or a tablet. The UAV (104) transmits aerial surveillance data acquired by the imaging unit (104a) to the control station (108) through the communication interface (106). The control station (108) processes the aerial surveillance data to identify regions of interest and transmits navigation parameters to the UGV (102). The UGV (102) navigates according to the navigation parameters and transmits subsurface scan data obtained from the GPR module (102b) to the control station (108) through the communication interface (106), wherein the control station (108) determines presence and location of buried objects.
[0051] In another exemplary embodiment, the control station (108) comprises a ruggedized field terminal deployed in an outdoor environment. The UAV (104) transmits aerial surveillance data captured by the imaging unit (104a) to the control station (108) via the communication interface (106). The control station (108) processes the aerial surveillance data to identify regions of interest and generate navigation parameters for the UGV (102). The UGV (102) transmits subsurface scan data acquired by the GPR module (102b) to the control station (108) through the communication interface (106), wherein the control station (108) processes the subsurface scan data to determine the presence and location of buried objects.
[0052] In a further exemplary embodiment, the control station (108) comprises a local operator device and a remote processing server. The UAV (104) transmits aerial surveillance data captured by the imaging unit (104a) to the control station (108) through the communication interface (106). The control station (108) forwards the aerial surveillance data to the remote processing server for processing to identify regions of interest and generate navigation parameters for the UGV (102). The UGV (102) transmits subsurface scan data obtained from the GPR module (102b) to the control station (108) through the communication interface (106), wherein the subsurface scan data is processed by the remote processing server to determine presence and location of buried objects and the processed results are returned to the control station (108) for display.
[0053] In one implementation, the buried objects detected by the ground penetrating radar module (102b) comprise hazardous and non-hazardous subsurface targets including landmines, improvised explosive devices, underground utility networks comprising pipes, cables, or conduits, and buried artefacts of archaeological or forensic relevance. The system (100) is thereby configured to support detection and localisation of both man-made and naturally concealed objects present within the subsurface region.
[0054] In one implementation, the control station (108) is configured to classify reflected subsurface scan data received from the ground penetrating radar module (102b) based on signal characteristics corresponding to different categories of buried objects, including landmines, improvised explosive devices, underground utility networks, and buried artefacts, to assist an operator in differentiating hazardous targets from non-hazardous subsurface structures.
[0055] FIG. 2 illustrates a flow chart for a method for coordinated buried object detection, in accordance with the present disclosure.
[0056] In step of acquiring (202) aerial surveillance data, the unmanned aerial vehicle (104) captures downward-facing images of a target area and transmits the aerial data to the control station (108) through the communication interface (106).
[0057] In step of processing (204) the aerial surveillance data, the control station (108) identifies one or more regions of interest and generates navigation information including distance and direction values corresponding to the selected region of interest.
[0058] In step of transmitting (206) the navigation information, the generated navigation parameters are communicated from the control station (108) to the unmanned ground vehicle (102) via the communication interface (106).
[0059] In step of navigating (208) the unmanned ground vehicle (102), the unmanned ground vehicle (102) moves toward the region of interest based on the received navigation information.
[0060] In step of performing (210) subsurface scanning, a ground penetrating radar mounted on the unmanned ground vehicle (102) transmits electromagnetic waves into a subsurface region and receives reflected signals from buried objects after the unmanned ground vehicle (102) reaches the target location.
[0061] In step of processing (212) reflected radar signals, the reflected signals are converted into digital data and processed at the control station (108) to determine the presence and location of buried objects.
[0062] In the implementation of the embodiment, the method (200) further comprises the step of acquiring aerial surveillance data, which comprises capturing a downward-facing image of the target area.
[0063] In the implementation of the embodiment, the step of generating navigation information comprises an aerial image into discrete grid cells, and converting pixel coordinates of a selected grid cell into real-world distance values.
[0064] In the implementation of the embodiment, the step of generating navigation information comprises generating a motion vector for the unmanned ground vehicle based on a difference between gyroscope reference coordinates and the converted real-world distance values, such that locomotion of the unmanned ground vehicle is achieved through successive coordinate subtraction.
[0065] In the implementation of the embodiment, wherein subsurface scanning is initiated only after the UGV (102) reaches a target location corresponding to the region of interest.
[0066] In the implementation of the embodiment, wherein the step of processing reflected radar signals comprises converting analog radar signals into digital data for imaging and visualization.
[0067] In the implementation of the embodiment, wherein acquiring aerial surveillance data comprises autonomously flying the unmanned aerial vehicle (104) along predefined waypoint trajectories in a coverage pattern selected from a lawnmower pattern and a grid pattern.
[0068] In the implementation of the embodiment, wherein processing the aerial surveillance data comprises generating a geo-referenced orthomosaic map using photogrammetric reconstruction of multiple aerial images.
[0069] In the implementation of the embodiment, wherein navigating the unmanned ground vehicle (102) comprises autonomous traversal along GPS-derived waypoints with obstacle avoidance based on proximity data received from a LiDAR sensor or an ultrasonic sensor.
[0070] In the implementation of the embodiment, the method (200) further comprising fusing ground penetrating radar data, thermal sensor data, and aerial image data to compute a probabilistic likelihood of a buried object at a detected location.
[0071] In the implementation of the embodiment, wherein the method (200) further comprising operating the unmanned aerial vehicle (104) in a loitering mode above the unmanned ground vehicle (102) to provide communication relay and forward scanning of the ground path during subsurface scanning.
[0072] In the implementation of the embodiment, wherein the method (200) further comprising geo-tagging detected subsurface anomalies and plotting the geo-tagged anomalies onto a real-time aerial map displayed at the control station (108).
[0073] In the implementation of the embodiment, wherein the method (200) further comprising validating detection performance by comparing detected target locations with pre-mapped buried reference targets and computing detection probability, false alarm rate, and area coverage rate.
[0074] FIG. 3 illustrates a block diagram of an integrated unmanned aerial vehicle and unmanned ground vehicle-based buried object detection system, in accordance with the present disclosure.
[0075] The system (100) comprises two coordinated subsystems, namely an unmanned aerial vehicle (UAV) (104) for aerial surveillance and an unmanned ground vehicle (UGV) (102) configured in a hexapod locomotion arrangement (102a) for ground-based sensing. The UAV (104) and the UGV (102) are independently powered by respective battery packs to enable wireless mobility and autonomous operation in field environments. The UAV (104) comprises an imagingunit (104a) mounted in a downward-facing orientation for acquiring aerial surveillance data of a target area. The UAV (104) is communicatively coupled to a control station (108) through a communication interface (106), wherein the communication interface (106) comprises a wireless radio frequency link enabling bidirectional exchange of control commands, operational status data, and sensing data between the UAV (104), the UGV (102), and the control station (108).
[0076] In operation, the UAV (104) performs aerial reconnaissance of the target area and captures real-time image and video data using the imaging unit (104a). The acquired aerial surveillance data is transmitted to an area mapping system executed at the control station (108), wherein the area mapping system processes the incoming video feed to scan the target area and generate a live visual representation with moderate latency. The aerial surveillance data is further provided to a graphical user interface executed at the control station (108), wherein the graphical user interface converts the incoming video data into a grid-based coordinate representation of the target area. The grid-based coordinate system assigns discrete spatial coordinates to selected surface regions suspected to correspond to subsurface anomalies, thereby defining one or more regions of interest for subsequent ground-based investigation.
[0077] Based on the grid-based coordinate representation generated from the aerial surveillance data, the control station (108) computes navigation parameters for the UGV (102), including real-world distance values and directional vectors corresponding to a selected region of interest. The navigation parameters are transmitted to the UGV (102) through the communication interface (106). The UGV (102) executes locomotion commands using the hexapod locomotion assembly (102a), wherein multiple articulated legs distribute ground contact forces to enable stable traversal over uneven, soft, or obstructed terrain. The UGV (102) navigates toward the defined region of interest under closed-loop coordination with the control station (108), which monitors positional feedback and operational status transmitted by the UGV (102).
[0078] Upon confirmation that the UGV (102) has reached the target location corresponding to the region of interest comprising a selected gridcoordinate derived from aerial mapping, the control station (108) initiates subsurface scanning by activating the ground penetrating radar (GPR) module (102b) mounted on the UGV (102). The GPR module (102b) transmits electromagnetic signals into a subsurface region of the target area and receives reflected signals corresponding to buried objects or subsurface discontinuities. The reflected radar signals are generated as analog electrical signals and are routed to an analog-to-digital conversion stage coupled to the GPR module (102b). The analog-to-digital converter digitizes the radar return signals to generate digital subsurface scan data suitable for further computational processing.
[0079] The digitized subsurface scan data is transmitted from the UGV (102) to the control station (108) through the communication interface (106). The control station (108) executes signal processing operations on the received digital data to generate spectrograms and subsurface images indicative of buried objects and anomalies. The signal processing may be implemented using software-based processing modules executed on a computing platform, such as a personal computer, to perform frequency-domain analysis and imaging of the subsurface region. The processed subsurface detection results are rendered on a display interface associated with the control station (108), thereby enabling visualization and interpretation of buried object presence and location in correlation with the grid-based aerial coordinate representation of the target area.
[0080] The control station (108) further manages coordinated operation of the UAV (104) and the UGV (102) through separate transmitter-receiver communication links provided by the communication interface (106), wherein aerial surveillance, grid-based mapping, navigation parameter generation, ground movement control, selective activation of the GPR module (102b), and subsurface data acquisition are synchronized as part of a unified operational workflow. The coordinated execution of aerial mapping and targeted ground-based subsurface sensing enables detection and localization of buried objects while reducing direct human exposure to hazardous environments, thereby facilitating safer land assessment, rehabilitation, and operational planning.
[0081] FIG. 4 illustrates the signal flow and functional components of the GPR module, in accordance with the present disclosure.
[0082] The ground penetrating radar module (102b) of the system (100) comprises a power supply stage configured to receive direct current power from a battery source mounted on the unmanned ground vehicle (102). The power supply stage provides regulated electrical power for waveform synthesis and operation of radio frequency components of the ground penetrating radar module (102b). The ground penetrating radar module (102b) comprises a radio frequency oscillator configured as a Colpitts oscillator coupled to a tunable LC tank circuit, wherein the oscillator generates a continuous-wave radio frequency signal within a frequency range selected for subsurface penetration and reflection from buried objects. The tunable LC tank circuit is configured to enable frequency adjustment based on soil characteristics, penetration depth requirements, and target material properties.
[0083] The radio frequency signal generated by the Colpitts oscillator is coupled to an antenna element through an actuating switch configured to alternately connect the antenna element between a transmission mode and a reception mode. During the transmission mode, the antenna element radiates electromagnetic waves into the subsurface region of a target area. The transmitted electromagnetic waves propagate through soil and underground materials and interact with buried objects, wherein a portion of the electromagnetic energy is reflected back toward the antenna element as a target echo in the electromagnetic domain. During the reception mode, the actuating switch couples the antenna element to a receiver path of the ground penetrating radar module (102b), thereby enabling reception of the reflected electromagnetic signals corresponding to subsurface targets.
[0084] The received electromagnetic signals captured by the antenna element are converted into corresponding electrical signals and provided to a mixer circuit within the ground penetrating radar module (102b). The mixer circuit is configured to combine the received electrical signals with a reference oscillator signal derived from the Colpitts oscillator to generate an intermediate signal containing frequency and phase information associated with the reflected electromagnetic waves. The extracted frequency and phase components areindicative of depth-related characteristics and material-dependent response of the buried objects. The output of the mixer circuit is provided as an analog signal to an analog-to-digital conversion stage coupled to the ground penetrating radar module (102b).
[0085] The analog-to-digital conversion stage digitizes the analog mixer output to generate digital subsurface scan data representative of the reflected electromagnetic signals. The digitized data is transmitted from the unmanned ground vehicle (102) to the control station (108) through the communication interface (106) for further computational processing. The control station (108) executes signal processing operations on the received digital data to perform spectral analysis, spectrogram generation, and subsurface imaging, thereby enabling visualization of buried objects and subsurface anomalies. The processed output is rendered on a display interface associated with the control station (108) for interpretation and analysis by an operator. The described signal flow and functional integration of the power supply stage, Colpitts oscillator with tunable LC tank circuit, actuating switch, antenna element, mixer circuit, and analog-to-digital conversion stage within the ground penetrating radar module (102b) enables controlled transmission and reception of electromagnetic waves for subsurface detection in coordinated operation with the unmanned aerial vehicle (104) and the control station (108).
[0086] FIG. 5 illustrates a block schematic of the control station, in accordance with the present disclosure. The control station (108) of the system (100) operates as a centralized decision-making and coordination unit for managing communication, data processing, and operational control between the unmanned aerial vehicle (104) and the unmanned ground vehicle (102). The control station (108) comprises bidirectional transmitter-receiver communication links forming part of the communication interface (106) for establishing a wireless data exchange channel between the unmanned aerial vehicle (104) and the control station (108), and between the unmanned ground vehicle (102) and the control station (108). The communication interface (106) is configured to transmit flight-related control signals to the unmanned aerial vehicle (104) and to receive aerial surveillance datacaptured by the imaging unit (104a) of the unmanned aerial vehicle (104). The control station (108) further comprises a computing platform implemented as a personal computer configured to execute processing routines for image analysis, navigation parameter generation, and subsurface data interpretation.
[0087] In operation, the control station (108) receives digital image data wirelessly transmitted from the unmanned aerial vehicle (104) through the communication interface (106). The received aerial surveillance data is rendered on a display interface of the computing platform to enable visual monitoring by an operator. The control station (108) executes an image processing module configured to segment the received digital image into a grid-based coordinate representation, thereby defining discrete grid cells corresponding to spatial regions of the target area. A pixel-to-distance conversion routine is applied to transform pixel coordinates associated with a selected grid cell into real-world coordinate values. Based on the converted real-world coordinates, the control station (108) performs target coordinate selection to define a region of interest for ground-based investigation.
[0088] The selected target coordinates generated by the control station (108) are transmitted to the unmanned ground vehicle (102) through the communication interface (106). The unmanned ground vehicle (102) navigates to the target location using the locomotion assembly (102a) based on the received navigation parameters. Upon completion of subsurface scanning by the ground penetrating radar module (102b), the unmanned ground vehicle (102) transmits digitized radar data back to the control station (108) through the communication interface (106). The control station (108) executes radargram generation and signal processing routines on the received subsurface scan data to generate visual representations indicative of buried objects and subsurface anomalies. The processed radargram output is displayed on the display interface of the computing platform for interpretation and analysis by the operator.
[0089] The control station (108) thereby provides synchronized coordination of aerial surveillance, grid-based mapping, target coordinate selection, navigation command generation, and subsurface data processing within a unifiedcontrol architecture. The control station (108) operates in a semi-automated manner under operator supervision, wherein the operator may intervene to select regions of interest, initiate scanning operations, and review processed detection outputs, while the control station (108) maintains real-time communication and data exchange with both the unmanned aerial vehicle (104) and the unmanned ground vehicle (102) through the communication interface (106) as part of the coordinated buried object detection system (100).
[0090] In a best mode of implementation, a coordinated buried object detection system (100) is deployed in a hazardous or inaccessible terrain such as a suspected landmine field, collapsed structure zone, or disaster-affected region. The system (100) comprises an unmanned ground vehicle (UGV) (102) having a locomotion assembly (102a) configured as a hexapod walking mechanism and a ground penetrating radar (GPR) module (102b) mounted on an adjustable forward-facing support frame. The hexapod locomotion assembly (102a) may be implemented using six independently actuated servo or brushless DC motor-driven legs, each leg comprising hip, knee, and foot joints with force-distributing pads to enable stable traversal over uneven soil, gravel, debris, and vegetation. In one example, each leg is actuated by digitally controlled servomotors with closed-loop feedback using rotary encoders, and coordinated gait patterns are generated by an onboard locomotion controller integrated within the UGV (102).
[0091] In operation, an unmanned aerial vehicle (UAV) (104) is first deployed above the target area to acquire aerial surveillance data. The UAV (104) comprises a downward-facing imaging unit (104a), such as a high-resolution RGB camera or a multispectral camera mounted on a stabilized gimbal, and an onboard inertial measurement unit (IMU) for attitude estimation. The UAV (104) transmits the captured aerial image frames and associated positional metadata to a control station (108) via a wireless communication interface (106), which may be implemented using a real-time radio frequency data link such as a Wi-Fi transceiver, long-range telemetry radio, or LTE / 5G modem. The control station (108), implemented as a ruggedized computing terminal or portable ground control unit, executes image processing routines to convert the aerial imagery into a grid-based coordinate system by dividing the image into discrete grid cells corresponding to real-world ground segments. Pixel coordinates of a selected grid cell representing a suspected region of interest are transformed into real-world distances and directional vectors using camera calibration parameters, altitude data of the UAV (104), and known field-of-view geometry.
[0092] Based on the grid-based coordinate conversion, the control station (108) generates navigation parameters for the UGV (102) in the form of motion vectors. In one exemplary implementation, the navigation parameters are computed by subtracting the current gyroscope reference coordinates of the UGV (102), obtained from an onboard inertial sensor, from the target real-world coordinate values derived from the aerial image. The resulting vector is translated into sequential locomotion commands corresponding to forward movement distance, angular orientation change, and stepping gait pattern parameters for the hexapod locomotion assembly (102a). These navigation parameters are transmitted in real time to the UGV (102) via the communication interface (106), enabling closed-loop guidance of the UGV (102) toward the selected region of interest.
[0093] Upon receiving navigation commands, the UGV (102) initiates movement using coordinated actuation of the hexapod legs, with successive coordinate subtraction being performed after each incremental movement step to correct positional drift. Once the UGV (102) reaches the target location as confirmed by onboard position sensors and feedback acknowledgment transmitted to the control station (108), the control station (108) issues an activation command to initiate subsurface scanning by the GPR module (102b). This selective activation ensures that radar transmission is performed only at aerially identified regions of interest rather than continuous scanning across the entire terrain, thereby conserving power and reducing unnecessary electromagnetic exposure.
[0094] The GPR module (102b) comprises a radio frequency oscillator-based signal generator configured to operate within a frequency range suitable for detecting metallic and non-metallic buried objects. In one exemplary embodiment, the signal generator comprises a Colpitts oscillator coupled to a tunable LC tank circuit to generate a continuous-wave radio frequency signal in the range of 500MHz to 3 GHz. The LC tank circuit may include variable capacitors or electronically tunable varactors to adjust the transmission frequency based on soil conditions and required penetration depth. A switching arrangement, such as a PIN diode-based RF duplexer or RF relay, alternately connects a shared antenna element between a transmission mode and a reception mode. During transmission, the antenna radiates electromagnetic waves into the subsurface region, and during reception, reflected subsurface signals are routed to a low-noise RF amplifier and mixer circuit.
[0095] The reflected signals are mixed with a reference oscillator signal to extract depth-related beat frequency information, which is then filtered and amplified using an RF amplifier circuit. The analog output of the GPR module (102b) is converted into digital data using an analog-to-digital converter mounted on the UGV (102). The digitized radar data is transmitted via the communication interface (106) to the control station (108), where signal processing algorithms, such as time-domain filtering, frequency-domain analysis, and threshold-based object detection routines, are applied to determine the presence and estimated location of buried objects. In one example, the processed subsurface data is visualized as a depth profile or two-dimensional subsurface image on a display interface of the control station (108), enabling an operator to identify buried metallic objects, voids, or anomalies.
[0096] The control station (108) further correlates the detected subsurface object location with the grid-based aerial coordinate system, thereby generating a combined aerial-subsurface mapping of the target area. The combined mapping enables precise localization of buried objects relative to surface landmarks captured by the UAV (104). Operational status data, including UGV position, UAV telemetry, GPR activation status, and detection confidence levels, are continuously exchanged between the UAV (104), UGV (102), and control station (108) through the communication interface (106), thereby maintaining synchronized coordinated operation.
[0097] In the best mode of implementation, the present invention is realized as a coordinated buried object detection system (100) comprising an unmannedground vehicle (UGV) (102), an unmanned aerial vehicle (UAV) (104), a communication interface (106), and a control station (108). The system is configured to operate in a sequential and coordinated manner wherein aerial surveillance is first performed to identify regions of interest, followed by targeted ground-based subsurface scanning at the identified regions. The coordinated operation ensures precise detection and localization of buried objects while minimizing direct human exposure to hazardous environments such as minefields, unstable terrain, or subsurface utility zones.
[0098] In the best mode, the unmanned ground vehicle (102) comprises a locomotion assembly (102a) implemented as a hexapod configuration. The hexapod locomotion assembly includes six articulated legs, each driven by multiple servo or brushless DC motors controlled by an onboard microcontroller or embedded processor such as an ARM-based controller or equivalent. The hexapod structure distributes ground contact forces across multiple legs, thereby reducing ground pressure and enabling stable traversal over uneven terrain, loose soil, rubble, or vegetation without triggering pressure-sensitive buried objects. Gait patterns such as tripod gait or wave gait are implemented through inverse kinematics algorithms stored in onboard memory and executed by the UGV control electronics.
[0099] The UGV (102) further comprises a ground penetrating radar (GPR) module (102b). In the best mode, the GPR module includes a radio frequency oscillator-based signal generator implemented using a Colpitts oscillator circuit. The oscillator is coupled to a tunable LC tank circuit, allowing the operating frequency to be adjusted within a range suitable for detecting both metallic and non-metallic buried objects, such as landmines, plastic explosives, pipelines, or underground utilities.
[00100] A shared antenna element is connected to the oscillator through a switching arrangement, such as a PIN diode or RF relay, which alternates the antenna between transmission and reception modes. During operation, the oscillator generates a continuous-wave radio frequency signal that is transmitted into the subsurface region. Reflected electromagnetic signals returning from subsurface anomalies are received by the same antenna, mixed with a reference oscillatorsignal using a mixer circuit, and processed to extract depth-related and material-dependent information prior to digitization by an analog-to-digital converter (ADC).
[00101] In the best mode, the unmanned aerial vehicle (104) comprises a multirotor or fixed-wing aerial platform equipped with an imaging unit (104a). The imaging unit includes a downward-facing optical camera, optionally combined with additional sensors such as thermal or multispectral cameras. The UAV is configured to perform aerial surveillance of a target area by following a predefined flight path or autonomously generated mapping trajectory.
[00102] The acquired aerial images are transmitted in real time to the control station (108) through the communication interface (106). The UAV may also include onboard positioning modules such as GPS or RTK-GPS to geo-tag the captured imagery, thereby enabling accurate spatial correlation between aerial observations and ground-level locations.
[00103] The communication interface (106) is implemented in the best mode as a wireless radio frequency link, such as Wi-Fi, LoRa, Zigbee, or a dedicated RF telemetry system. The interface enables bidirectional data exchange between the UAV (104), the UGV (102), and the control station (108). Navigation data, aerial surveillance data, subsurface sensing data, and operational status information are exchanged in real time. The communication interface ensures synchronized operation of aerial and ground platforms and supports low-latency command and data transmission necessary for coordinated detection.
[00104] The control station (108) functions as the central processing and decision-making unit of the system. In the best mode, the control station is implemented as a ruggedized computing platform or portable workstation executing dedicated software modules. The control station processes aerial surveillance data received from the UAV to identify one or more regions of interest corresponding to potential buried object locations.
[00105] The aerial images are converted into a grid-based coordinate system, wherein the image is divided into discrete grid cells. Pixel coordinates of selected grid cells are mathematically converted into real-world distance and directionalvalues using camera calibration parameters and UAV altitude data. Based on this conversion, navigation parameters are generated and transmitted to the UGV to guide it toward the selected region of interest.
[00106] In the best mode, artificial intelligence (AI) techniques are employed at the control station to enhance region-of-interest identification. A trained machine learning model, such as a convolutional neural network (CNN), is used to analyze aerial imagery and detect visual patterns indicative of buried object likelihood, including soil disturbances, terrain anomalies, thermal variations, or surface features associated with prior excavation.
[00107] Exemplary training datasets include labeled aerial images of known minefields, underground utility corridors, construction sites, or simulated buried-object environments. The datasets may include annotated regions representing positive and negative examples of buried-object indicators. The trained AI model outputs probability maps or confidence scores for each grid cell, which are used by the control station to prioritize regions of interest for ground-based scanning.
[00108] Upon receiving navigation parameters, the UGV (102) navigates toward the identified region of interest. In the best mode, onboard inertial sensors such as gyroscopes and accelerometers provide real-time orientation and position data. Motion vectors are generated by computing the difference between current gyroscope reference coordinates and target coordinates derived from the aerial grid-based mapping. Locomotion is achieved through successive coordinate subtraction and closed-loop control of the hexapod legs, enabling precise positioning of the GPR module over the target location.
[00109] The subsurface scanning is initiated only after the UGV confirms arrival at the target location derived from aerial surveillance data. This selective activation ensures that radar transmission is focused on aerially identified regions of interest rather than continuous or blind ground scanning. The GPR module transmits electromagnetic waves into the subsurface and receives reflected signals corresponding to buried objects or anomalies.
[00110] Reflected radar signals are converted from analog to digital form and transmitted to the control station for processing. In the best mode, signal processingincludes filtering, time-frequency analysis, and generation of radargrams or spectrograms. AI-based classification models, such as support vector machines or neural networks trained on labeled GPR datasets, may be used to distinguish true buried objects from clutter such as rocks or roots.
[00111] The processed data is visualized as intuitive images or overlays on the aerial map, enabling operators to understand detection results without requiring specialized radar expertise.
[00112] In another best mode, the control station (108) is configured to determine target coordinates and locomotion parameters for the unmanned ground vehicle (102) exclusively from image processing of the aerial surveillance data and associated computational algorithms, such that navigation and coordination of the UAV (104) and the UGV (102) are performed without reliance on global navigation satellite system (GNSS) signals, thereby enabling operation in GPS-denied or GPS-restricted environments.
Claims
1. A system for coordinated buried object detection, the system (100) comprising: an unmanned ground vehicle (UGV) (102) comprising a locomotion assembly (102a) and a ground penetrating radar (GPR) module (102b) configured to transmit electromagnetic waves into a subsurface region and receive reflected signals from buried objects; an unmanned aerial vehicle (UAV) (104) comprising an imaging unit (104a) configured to acquire aerial surveillance data of a target area; a communication interface (106) configured to enable bidirectional data exchange between the UAV (104) and the UGV (102); and a control station (108) configured to: process the aerial surveillance data to identify regions of interest and generate navigation parameters for the UGV (102); and receive and process subsurface scan data from the GPR module (102b) to determine the presence and location of buried objects; wherein coordinated operation of the UAV (104) and the UGV (102) enables detection and localization of buried objects while reducing direct human exposure to hazardous environments.
2. The system (100) as claimed in claim 1, wherein the locomotion assembly (102a) of the UGV (102) comprises a hexapod configuration adapted to distribute ground contact forces across multiple legs for stable movement over uneven terrain.
3. The system (100) as claimed in claim 1, wherein the GPR module (102b) comprises a radio frequency oscillator-based signal generator configured to operate within a frequency range suitable for detecting metallic and non-metallic buried objects, and wherein the radio frequency oscillator comprises a Colpitts oscillator configured to generate a continuous-wave radar signal; and wherein the ground penetrating radar module comprises: a tunable LC tank circuit coupled to a Colpitts oscillator, the oscillator being configured to generate a continuous-wave radio frequency signal, and a switching arrangement configured to alternately connect a shared antenna element between a transmission mode and a reception mode, wherein reflected subsurface signals are mixed with a reference oscillator signal to extract depth-related information prior to digitization.
4. The system (100) as claimed in claim 1, wherein the UAV (104) imaging unit (104a) comprises a downward-facing camera, and wherein the control station (108) is configured to convert the acquired aerial image into a grid-based coordinate system, and wherein the grid-based coordinate system is used to determine a real-world distance and direction for guiding the movement of the UGV (102) toward a selected region of interest.
5. The system (100) as claimed in claim 1, wherein wherein the control station (108) is configured to determine target coordinates and locomotion parameters for the unmanned ground vehicle (102) exclusively from image processing of the aerial surveillance data and associated computational algorithms, such that navigation and coordination of the UAV (104) and the UGV (102) are performed without reliance on global navigation satellite system (GNSS) signals, thereby enabling operation in GPS-denied or GPS-restricted environments..
6. The system (100) as claimed in claim 1, wherein the control station (108) is configured to initiate subsurface scanning by the ground penetrating radar only upon confirmation that the unmanned ground vehicle has reached a target location derived from aerial surveillance data, such that radar transmission is selectively activated based on aerially identified regions of interest rather than continuous ground scanning.
7. A method (200) for detecting buried objects, the method (200) comprising: acquiring (202) aerial surveillance data of a target area using an unmanned aerial vehicle (UAV) (104); processing (204) the aerial surveillance data to identify one or more regions of interest and generate navigation information; transmitting (206) the navigation information to an unmanned ground vehicle (UGV) (102); navigating (208) the UGV (102) to the region of interest based on the navigation information; performing (210) subsurface scanning of the region of interest using a ground penetrating radar mounted on the UGV (102); and processing (212) reflected radar signals to determine the presence and location of buried objects.
8. The method (200) as claimed in claim 7, wherein: the step of acquiring aerial surveillance data comprises capturing a downward-facing image of the target area; the step of generating navigation information comprises: dividing an aerial image into discrete grid cells, converting pixel coordinates of a selected grid cell into real-world distance values, and generating a motion vector for the unmanned ground vehicle based on a difference between gyroscope reference coordinates and the converted real-world distance values, such that locomotion of the unmanned ground vehicle is achieved through successive coordinate subtraction.
9. The method (200) as claimed in claim 7, wherein subsurface scanning is initiated only after the UGV (102) reaches a target location corresponding to the region of interest.
