A method and system for comprehensive detection of goaf areas using a combination of unmanned aerial vehicles (UAVs) and ground robots.

Abstract

This application relates to the field of mine exploration technology, and in particular to a method and system for comprehensive exploration of goaf areas using a UAV-ground robot collaborative approach. The method includes: acquiring a remote gypsum mine scanning information set; analyzing dust interference and water mist risk information within the goaf area based on the remote gypsum mine scanning information set to obtain an initial risk parameter set for the goaf; analyzing the dynamic coupling relationship between dust generated by the ground robot's movement and its obstruction interference to UAV detection, and water mist generated by the UAV's dust suppression operations and its slippery interference to the ground robot's passage, based on the initial risk parameter set, to obtain a dynamic cross-interference feature set; and generating a collaborative exploration strategy between the UAV and ground robot, aiming to avoid high-risk areas and maximize exploration efficiency, based on the dynamic cross-interference feature set, and outputting a comprehensive stability assessment report for the goaf. This method compensates for blind spots in single-device exploration, improving exploration efficiency and resource utilization.

Description

Technical Field

[0001] This application relates to the field of mine exploration technology, and in particular to a method and system for comprehensive exploration of goaf areas using a combination of unmanned aerial vehicles (UAVs) and ground robots. Background Technology

[0002] In existing technologies, drones can achieve large-scale scanning of goaf areas from a high-altitude perspective. Some solutions use drones to carry dust suppression devices to spray water mist to reduce the interference of dust on detection. Ground robots can obtain detailed data such as wall cracks and structural parameters of goaf areas at close range. Current collaborative detection solutions mostly adopt a static strategy of "planning the path first and then executing it in a fixed manner", simply dividing the detection areas of the two.

[0003] Existing static collaborative detection strategies do not consider the dynamic interaction between UAVs and ground robots during operation. They neither quantify the real-time interference intensity of ground robot dust on UAV signals nor assess the dynamic impact of UAV dust suppression water mist on the safety of ground robot passage. They cannot adapt to the dynamic coupling relationship of "dust distribution-water mist retention-equipment operation" in the goaf area, causing the system to fall into the dilemma of "risk avoidance and detection effectiveness being difficult to balance". Furthermore, the stability assessment report cannot reflect dynamic risk changes, which may mislead safety decisions. Summary of the Invention

[0004] This application provides a method and system for comprehensive detection of goaf areas using a combination of unmanned aerial vehicles (UAVs) and ground robots, in order to solve the above-mentioned problems.

[0005] In a first aspect, this application provides a method for comprehensive detection of goaf areas using a combination of unmanned aerial vehicles (UAVs) and ground robots, the method comprising:

[0006] A remote gypsum mine scanning information set is acquired. Based on this information set, dust interference information and water mist risk information within the gypsum mine goaf are analyzed to obtain an initial risk parameter set for the goaf. Based on this initial risk parameter set, the dynamic coupling relationship between the dust generated by the ground robot's movement and the obstruction interference of the drone's detection, and the slippery interference of the water mist generated by the drone's dust suppression operation on the ground robot's passage, is analyzed to obtain a dynamic cross-interference feature set. Based on this dynamic cross-interference feature set, a collaborative detection strategy between the drone and the ground robot is generated with the goal of avoiding high-risk areas and maximizing detection efficiency. A comprehensive stability assessment report for the goaf is then output.

[0007] The above technical solution utilizes the collaborative detection of drones and ground robots to effectively compensate for the blind spots of single equipment. By combining aerial macroscopic scanning with precise ground detection and analyzing and controlling the cross-interference of dust and water mist between the two, the accuracy of key data in the goaf area is significantly improved, reducing missed and false detections. Collaborative strategies are formulated with the goal of avoiding high-risk areas, clarifying safe action paths, reducing the probability of equipment damage, and reducing the need for personnel to enter the mine, thus ensuring operational safety. By optimizing task allocation and collaborative timing, duplicate detection and rework are avoided, significantly improving detection efficiency and resource utilization. The final comprehensive assessment report integrates multi-source data, clarifies the risk level and location of hidden dangers, and provides a scientific basis for goaf reinforcement, backfilling, and other remediation work, helping to reduce the risk of collapse accidents and ensuring the safety of mine production and the surrounding environment.

[0008] Optionally, based on the remote gypsum mine scanning information set, the dust concentration, dust distribution range, and dust suspension duration in different areas of the goaf are analyzed to determine the light shading intensity and detection signal attenuation of the detection components carried by the UAV, thus obtaining the dust interference information; based on the remote gypsum mine scanning information set, the water mist accumulation density, water mist coverage area, and water mist retention duration in different areas of the goaf are analyzed to determine the degree of friction reduction and slippage probability between the ground robot's walking mechanism and the contact surface, thus obtaining the water mist risk information; based on the dust interference information and the water mist risk information, the spatiotemporal coupling relationship between the two in the goaf is evaluated to generate the initial risk parameter set of the goaf used to characterize the comprehensive risk level of the region.

[0009] Optionally, based on the dust concentration, the degree of attenuation of the transmittance of the visible and infrared spectra of the UAV detection components by dust particles is analyzed to obtain the light blocking intensity under different visibility levels; based on the dust distribution range and the dust suspension duration, the scattering and absorption effects of dust particles on the laser ranging signal and image acquisition signal of the UAV detection components are analyzed to determine the attenuation degree of the effective signal transmission distance and the probability of image clarity reduction, thereby obtaining the signal attenuation degree; integrating the light blocking intensity and the signal attenuation degree, the dust interference information is obtained.

[0010] Optionally, based on the water mist aggregation density, the thickness and uniformity of the water film formed between the ground robot's walking mechanism and the contact surface are analyzed, and the attenuation coefficient of the ground robot's friction force on the contact surface under different water film states is evaluated to obtain the degree of friction reduction; based on the water mist coverage area and the water mist retention time, the overlap ratio between the ground robot's predetermined walking path and the water mist coverage area and the dwell time in the gypsum mine area are analyzed to determine the probability of slippage under different overlap ratios and dwell times to obtain the slippage probability; integrating the degree of friction reduction and the slippage probability, the impact of different areas on the ground robot's passage safety is comprehensively evaluated to obtain the water mist risk information.

[0011] Optionally, based on the light shielding intensity and the signal attenuation degree, the adsorption and sedimentation effect of high water mist accumulation on suspended dust in the UAV dust suppression operation area is analyzed to obtain the suppression effect strength of water mist on dust interference; based on the friction reduction degree and the slip probability, the blowing and drying effect of dust caused by ground robot movement on local water mist is analyzed to obtain the mitigation effect strength of dust on water mist risk; based on the suppression effect strength and the mitigation effect strength, the mutual constraint relationship between dust and water mist risks during UAV dust suppression operation and ground robot movement detection is cross-analyzed to generate the initial risk parameter set of the goaf area for quantifying the dynamic risk level of the area.

[0012] Optionally, based on the intensity of the suppression effect, the adsorption and sedimentation efficiency of water mist sprayed by the UAV on suspended dust in the dust suppression operation area is analyzed to obtain water mist dust suppression efficiency information; based on the water mist dust suppression efficiency information, combined with the intensity of the weakening effect, the degree of interference of dust on water mist adsorption when the ground robot moves in the gypsum mine area is analyzed to obtain dust-water mist dynamic suppression balance characteristics; based on the intensity of the weakening effect, the blowing and drying rate of local water mist caused by dust caused by the movement of the ground robot is analyzed to obtain dust dehumidification efficiency information; based on the dust dehumidification efficiency information, combined with the intensity of the suppression effect, the degree of suppression of the dust diffusion effect by the newly generated water mist from the UAV dust suppression operation is analyzed to obtain water mist-dust dynamic weakening balance characteristics; integrating the dust-water mist dynamic suppression balance characteristics and the water mist-dust dynamic weakening balance characteristics, the dominant alternation law of dust interference and water mist risk under different spatiotemporal conditions is comprehensively analyzed to generate the dynamic cross-interference feature set.

[0013] Optionally, based on the water mist dust suppression efficiency information, the trend of the adsorption efficiency of suspended dust in the goaf area by the UAV sprayed water mist over time is analyzed to obtain the dynamic efficiency curve of water mist dust suppression; based on the weakening effect intensity, the influence of the ground robot's moving speed and path on the intensity of local dust generation is analyzed to obtain the dynamic dust generation characteristics; based on the water mist dust suppression dynamic efficiency curve, combined with the dynamic dust generation characteristics, the interference intensity of dust generated during the ground robot's movement on the water mist adsorption process, and the water mist's ability to suppress newly generated dust are analyzed. The difference between the interference intensity and the suppression ability is used as the dust-water mist dynamic suppression balance index. The closer the dust-water mist dynamic suppression balance index is to zero, the higher the suppression coverage of water mist on dust interference; the dust-water mist dynamic suppression balance index is used as the dust-water mist dynamic suppression balance characteristic.

[0014] Optionally, based on the dust dehumidification efficiency information, the trend of the dust dispersion effect on local water mist caused by changes in the ground robot's moving speed and path over time is analyzed to obtain a dynamic efficiency curve for dust dehumidification; based on the suppression effect intensity, the encapsulation and adsorption efficiency of water mist spraying rate and coverage area on newly added dust particles during UAV dust suppression operations is analyzed, and the weighted sum of the water mist spraying rate and the encapsulation and adsorption efficiency is used as the dynamic response intensity of water mist dust suppression; based on the dynamic efficiency curve for dust dehumidification, combined with the dynamic response intensity of water mist dust suppression, the influence trend of dust generated during the ground robot's movement after being affected by the UAV and the suppression ability of water mist on dust are analyzed to obtain the dynamic weakening balance intensity of water mist-dust; based on the dynamic weakening balance intensity of water mist-dust, combined with the water mist risk information and the dust interference information, the dominant relationship between water mist dust suppression and dust diffusion in the gypsum mine area is determined, and the dynamic weakening balance characteristic of water mist-dust is generated to characterize the balance state of dust suppression and dust weakening.

[0015] Optionally, based on the dynamic suppression balance characteristics of dust and water mist, the changes in the suppression and counter-suppression intensity of water mist adsorption and ground robot-generated dust in the time dimension are analyzed to obtain the evolution trend of the dynamic balance of dust and water mist; based on the dynamic weakening balance characteristics of water mist and dust, the changes in the weakening and counter-weakening intensity of dust dispersion and UAV dust suppression water mist in the spatial dimension are analyzed to obtain the evolution trend of the dynamic balance of water mist and dust; based on the evolution trends of the dynamic balance of dust and water mist and the dynamic balance of water mist and dust, the alternation pattern and duration of the dust interference-dominated stage and the water mist risk-dominated stage in different regions during the detection process are cross-analyzed to obtain the detection dynamic risk-dominated stage sequence; based on the detection dynamic risk-dominated stage sequence, combined with the dust interference information and the water mist risk information, the comprehensive risk level of the detection impact of each region in different time segments is quantified to generate the spatiotemporally refined dynamic cross-interference feature set.

[0016] Secondly, this application provides a comprehensive goaf detection system based on a drone-ground robot collaboration. The system includes: an information acquisition module for acquiring a remote gypsum mine scanning information set, and based on the remote gypsum mine scanning information set, analyzing dust interference information and water mist risk information within the gypsum mine goaf to obtain an initial risk parameter set for the goaf; a cross-analysis module for analyzing, based on the initial risk parameter set for the goaf, the dynamic coupling relationship between the dust generated by the ground robot's movement and the obstruction interference of drone detection, and the slippery interference of water mist generated by the drone's dust suppression operation on the ground robot's passage, to obtain a dynamic cross-interference feature set; and a strategy generation module for generating a drone-ground robot collaborative detection strategy based on the dynamic cross-interference feature set, with the goal of avoiding high-risk areas and maximizing detection efficiency, and outputting a comprehensive goaf stability assessment report. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram illustrating an application scenario provided in one embodiment of this application;

[0019] Figure 2 A flowchart illustrating a method for comprehensive detection of goaf areas using a drone-ground robot collaborative approach, as provided in one embodiment of this application;

[0020] Figure 3 This is a schematic diagram of a comprehensive detection system for goaf areas using a drone-ground robot collaborative approach, provided as an embodiment of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0022] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0023] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0024] Drones can achieve large-scale scanning of goaf areas from a high-altitude perspective. Some solutions use drones to carry dust suppression devices to spray water mist to reduce the interference of dust on the detection. Ground robots can obtain detailed data such as wall cracks and structural parameters of goaf areas at close range. Currently, most collaborative detection solutions adopt a static strategy of "planning the path first and then executing it in a fixed manner" and simply divide the detection areas of the two.

[0025] Based on this, this application provides a method and system for comprehensive detection of goaf areas using a combination of UAVs and ground robots. By coordinating UAVs and ground robots for detection, the system effectively compensates for the blind spots of single equipment detection. It combines aerial macroscopic scanning with precise ground detection and analyzes and controls the cross-interference of dust and water mist between the two, significantly improving the accuracy of key data in goaf areas, reducing missed and false detections, formulating collaborative strategies with the goal of avoiding high-risk areas, clarifying safe action paths, and reducing the probability of equipment damage.

[0026] Figure 1 This application provides an illustration of an application scenario. In collaborative detection of gypsum mines, a collaborative strategy is used to efficiently and comprehensively detect mined-out areas, reduce risk exposure, improve assessment accuracy, provide a reliable basis for mine safety, and enhance overall detection effectiveness and management efficiency.

[0027] Specifically, the method provided in this application can be applied to any server, which interacts with a pre-deployed sensor network to obtain a set of remote gypsum mine scanning information provided by the pre-deployed sensor network. It analyzes and controls the cross-interference between dust and water mist between the two, significantly improves the accuracy of key data in the goaf, reduces missed and false judgments, formulates collaborative strategies with the goal of avoiding high-risk areas, clarifies safe action paths, generates a comprehensive stability assessment report of the goaf, and provides it to the detection personnel, greatly improving detection efficiency and resource utilization.

[0028] For specific implementation details, please refer to the following examples.

[0029] Figure 2 This is a flowchart illustrating a method for comprehensive detection of mined-out areas using a drone-ground robot collaborative approach, as provided in one embodiment of this application. The method of this embodiment can be applied to servers in the above-described scenario. Figure 2 As shown, the method includes:

[0030] S201. Obtain the remote gypsum mine scanning information set. Based on the remote gypsum mine scanning information set, analyze the dust interference information and water mist risk information in the goaf of the gypsum mine to obtain the initial risk parameter set of the goaf.

[0031] The remote gypsum mine scanning information set can be a collection of data obtained by scanning the goaf of a gypsum mine using remote sensing devices, provided by a pre-deployed sensor network. Dust disturbance information can be information on the concentration and distribution of dust within the goaf. Water mist risk information can be information on the level of water mist or humidity within the goaf. The initial risk parameter set for the goaf can be a set of parameters integrating dust disturbance information and water mist risk information.

[0032] Specifically, goaf detection is a crucial aspect of mine safety. However, existing detection methods often overlook environmental factors such as dust and water mist, leading to inaccurate detection data and even safety accidents. In current technologies, the detection process is usually conducted independently, lacking collaborative analysis of multiple risk factors, making it difficult to comprehensively assess the stability of the goaf. By acquiring remote scanning information and analyzing the risks of dust and water mist, an initial risk parameter set can be obtained, providing basic data support for subsequent collaborative detection and improving the reliability and safety of the detection.

[0033] S202. Based on the initial risk parameter set of the goaf, analyze the dynamic coupling relationship between the dust caused by the movement of the ground robot and the obstruction interference of the drone detection, and the water mist generated by the drone dust suppression operation and the slippery interference of the ground robot passage, and obtain the dynamic cross-interference feature set.

[0034] Dynamic coupling can refer to the mutual influence between ground robots and drones during exploration. A dynamic cross-interference feature set can be a set of features that quantifies this coupling relationship.

[0035] Specifically, in collaborative detection, the actions of ground robots and drones can affect each other. Ignoring these cross-interferences can lead to low detection efficiency or equipment damage. For example, the dust generated by the robot's movement may prevent the drone from detecting accurately, while the drone's dust suppression operation may make the ground slippery, causing the robot to slip or get stuck. Current technologies lack analysis of this dynamic coupling relationship, resulting in suboptimal collaborative detection strategies. They often adopt static planning, which cannot adapt to real-time environmental changes. By analyzing the dynamic coupling relationship, a cross-interference feature set can be obtained, enabling subsequent strategies to avoid these interferences and improve collaborative efficiency.

[0036] S203. Based on the dynamic cross-interference feature set, with the goal of avoiding high-risk areas and maximizing detection efficiency, a collaborative detection strategy of UAV-ground robot is generated, and a comprehensive assessment report on the stability of the goaf is output.

[0037] Collaborative detection strategies can include planning the movement paths, task allocation, and time scheduling of drones and ground robots. A comprehensive stability assessment report for goaf areas can be a report that integrates detection data, risk analysis, and stability assessment.

[0038] Specifically, in UAV-Ground Robot Collaborative Exploration, generating collaborative strategies is crucial to ensuring successful exploration. Existing methods are often based on static planning, which cannot adapt to dynamic environmental changes, leading to incomplete exploration coverage or excessive risk exposure. By using dynamic cross-interference feature sets, with the goal of avoiding high risks and maximizing efficiency, adaptive strategies are generated to improve the comprehensiveness and accuracy of exploration. A comprehensive evaluation report is then output to provide decision support for mine management, avoid blind exploration, and ensure efficient resource utilization.

[0039] The method provided in this embodiment combines drones and ground robots for collaborative detection, effectively compensating for blind spots in single-device detection. By integrating aerial macroscopic scanning with precise ground detection and analyzing and controlling the cross-interference of dust and water mist between the two, the accuracy of key data in goaf areas is significantly improved, reducing missed and false detections. Collaborative strategies are formulated with the goal of avoiding high-risk areas, clarifying safe action paths, reducing the probability of equipment damage, and minimizing the need for personnel to enter the mine, thus ensuring operational safety. By optimizing task allocation and collaborative timing, redundant detection and rework are avoided, significantly improving detection efficiency and resource utilization. The final comprehensive assessment report integrates multi-source data, clarifies risk levels and hazard locations, and provides a scientific basis for goaf reinforcement, backfilling, and other remediation work, helping to reduce the risk of collapse accidents and ensuring the safety of mine production and the surrounding environment.

[0040] In some embodiments, based on a remote gypsum mine scanning information set, the dust concentration, dust distribution range, and dust suspension duration in different areas of the goaf are analyzed to determine the light shading intensity and detection signal attenuation of the detection components carried by the UAV, thus obtaining dust interference information. Based on the remote gypsum mine scanning information set, the water mist accumulation density, water mist coverage area, and water mist retention duration in different areas of the goaf are analyzed to determine the degree of friction reduction and slippage probability between the ground robot's walking mechanism and the contact surface, thus obtaining water mist risk information. Based on the dust interference information and water mist risk information, the spatiotemporal coupling relationship between the two in the goaf is evaluated to generate an initial risk parameter set for the goaf to characterize the comprehensive risk level of the region.

[0041] Dust interference information can be a set of data related to the adverse effects of dust in the goaf on UAV detection. Dust concentration can be the mass or quantity of dust particles contained in a unit volume of air in the goaf. Dust distribution range can be the size of the spatial area where dust is concentrated in the goaf. Dust suspension duration can be the duration for which dust particles remain suspended in the air of the goaf. The detection components carried by the UAV can be equipment installed on the UAV for goaf detection. Light obstruction intensity can be the degree to which dust particles obstruct the light received by the UAV detection components. Detection signal attenuation can be the degree to which dust weakens the laser ranging signal and image acquisition signal emitted by the UAV detection components. Water mist risk information can be a set of data related to the adverse effects of water mist in the goaf on the passage of ground robots. Water mist accumulation density can be the mass or quantity of water mist contained in a unit volume of air in the goaf. Water mist coverage area can be the size of the spatial area covered by water mist diffusion in the goaf. Water mist retention time can be the duration for which water mist remains in an accumulated state in a specific area of ​​the goaf. Ground robot walking mechanism can be the components used for movement by the ground robot. The contact surface can be the area where the ground robot's walking mechanism contacts the goaf ground. The degree of friction reduction refers to the reduction in friction between the walking mechanism and the contact surface after water mist forms a water film. The slip probability is the likelihood of the ground robot slipping when walking in a water-mist-covered area. The initial risk parameter set for the goaf can be a set of parameters used to quantify the comprehensive risk level of different areas within the goaf. The spatiotemporal coupling relationship can be the mutual influence and interaction between dust disturbance information and water mist risk information at different spatial locations and time periods within the goaf. The comprehensive regional risk level can be based on the combined impact of dust disturbance and water mist risk.

[0042] Specifically, in the field of gypsum mine goaf detection, collaborative detection by UAVs and ground robots is the core technological path to achieve comprehensive and efficient detection. However, the complex environmental interference in goaf areas has always been a key bottleneck restricting detection efficiency, and dust and water mist are the two most important and common interference factors. This step addresses the above problems through the following methods: Based on the remote gypsum mine scanning information set (such as the overall spatial data of the goaf obtained by a ground-based 3D laser scanner and the environmental sensor data obtained by UAV high-altitude pre-scanning), a comprehensive approach combining environmental parameter analysis, spectral analysis, tribological simulation, probability statistics, and spatiotemporal coupling analysis is adopted. First, the dust interference information is analyzed: the dust concentration (e.g., 8 mg / m³) in different areas is extracted from the remote scanning information set. 3The dust distribution range (e.g., 30m × 20m) and dust suspension duration (e.g., 3 hours) were analyzed. Spectral analysis was used to assess the attenuation of visible and infrared spectral transmittance of dust particles on the visible light camera, infrared thermal imager, and other detection components mounted on the UAV. The light blocking intensity corresponding to different visibility levels (e.g., 3m visibility) was determined. Simultaneously, combining spatial distribution analysis and signal transmission simulation methods, the scattering effect of dust particles on laser ranging signals and the absorption effect on image acquisition signals were analyzed based on the dust distribution range and suspension duration. The attenuation degree of effective signal transmission distance (e.g., from 80m to 45m) and the probability of image clarity reduction (e.g., a 25% reduction) were determined, and dust interference information was integrated. Subsequently, water mist risk information was analyzed: water mist aggregation density (e.g., 0.4g / m³) in different areas was extracted from the remote scanning data set. 3 The system considers the water mist coverage area (e.g., 25m × 15m) and water mist retention time (e.g., 2 hours). Using a tribomechanical simulation method, it assesses the thickness and uniformity of the water film formed between the ground robot's tracked walking mechanism and the gypsum rock in the goaf based on the water mist aggregation density. This determines the frictional attenuation coefficient under different water film states, yielding the degree of frictional reduction (e.g., 18% reduction). Simultaneously, combining path matching analysis and event probability calculation methods, it determines the degree of frictional attenuation based on the overlap ratio between the water mist coverage area and the ground robot's predetermined walking path (e.g., 40% overlap) and the residence time in that area (e.g., 0.5 hours). The probability of slippage events is determined (e.g., 12%), and water mist risk information is integrated. Finally, spatiotemporal coupling analysis and risk level quantification methods are used to evaluate the interaction between dust interference information and water mist risk information in different spatial locations (e.g., the eastern and western areas of the goaf) and different time periods (e.g., 10:00 AM to 12:00 PM and 3:00 PM to 5:00 PM). For example, water mist accumulation in the eastern part of the goaf in the morning leads to dust deposition, which reduces the intensity of dust interference but increases the risk of slippage of ground robots. This generates an initial risk parameter set for the goaf to characterize the comprehensive risk level of the region (e.g., low risk, medium risk, high risk).

[0043] By using the method provided in this embodiment, system analysis is conducted based on the remote gypsum mine scanning information set, enabling the early identification and quantification of two core interference factors in the goaf: dust and water mist. This avoids the detection equipment from directly entering unknown high-risk areas, effectively reducing the safety risks of equipment damage and mission interruption. Through targeted analysis of the effects of dust on the light obstruction and signal attenuation of the UAV detection components, and the effects of water mist on the friction reduction and slippage of the ground robot, the data obtained in the subsequent collaborative detection process is ensured to have high accuracy and effectiveness, providing reliable data support for the stability assessment of the goaf.

[0044] In some embodiments, based on dust concentration, the degree of attenuation of the transmittance of visible light and infrared spectra of the UAV detection component by dust particles is analyzed to obtain the light blocking intensity under different visibility levels; based on the dust distribution range and dust suspension duration, the scattering and absorption effects of dust particles on the laser ranging signal and image acquisition signal of the UAV detection component are analyzed to determine the attenuation degree of effective signal transmission distance and the probability of image clarity reduction, thus obtaining the signal attenuation degree; integrating the light blocking intensity and the signal attenuation degree, dust interference information is obtained. The light blocking intensity can be the degree of obstruction and attenuation of the visible light and infrared spectrum upon which the UAV detection component relies by dust particles. The signal attenuation degree can be the degree of signal strength reduction, effective transmission distance shortening, and image quality degradation caused by the scattering and absorption effects of dust particles on the laser ranging signal and image acquisition signal emitted by the UAV detection component.

[0045] Specifically, the dust environment within gypsum mine goaf is complex and variable, directly affecting the accuracy and reliability of UAV detection. Excessive dust concentration can cause light obstruction, preventing UAVs from acquiring clear images or infrared data. The distribution range and duration of dust suspension determine the spatiotemporal extent of this interference. Ignoring these factors may lead to misjudgments or missed detections of critical structural defects in high-dust areas, potentially causing safety accidents. This step addresses these issues using the following method: Based on the remote gypsum mine scanning information set, environmental parameter analysis technology is first used to extract the dust concentration (e.g., 1.0 mg / m³) in different areas of the goaf. 3 The system first obtains basic parameters such as dust distribution range (e.g., 120 square meters) and dust suspension duration (e.g., 2.5 hours). Then, for the dust concentration parameter, spectral transmittance analysis is used to simulate the impact of different concentrations of dust particles on the transmission paths of visible light (e.g., wavelength 550 nm) and infrared spectrum (e.g., wavelength 10 μm) upon which the UAV's detection components rely. The degree of spectral transmittance attenuation is calculated, and different visibility levels are then defined (e.g., 6 meters of visibility corresponds to strong obstruction intensity, and 18 meters of visibility corresponds to weak obstruction intensity), thus obtaining the light obstruction intensity. Simultaneously, the extracted dust distribution range and dust suspension duration are analyzed... The parameters were analyzed using signal scattering and absorption simulation technology to construct an interaction model between dust particles and laser ranging signals (e.g., the original effective transmission distance of 45 meters is attenuated to 28 meters after dust scattering and absorption) and image acquisition signals. The interference degree of dust on signal propagation under different distribution ranges and suspension durations was analyzed to determine the attenuation ratio of the effective transmission distance and the probability of image clarity reduction (e.g., 40%), thus obtaining the signal attenuation degree. Finally, through multi-source information fusion, the obtained light blocking intensity and signal attenuation degree were comprehensively correlated and analyzed to form dust interference information that can comprehensively reflect the level of dust interference on UAV detection.

[0046] By analyzing the specific interference effect of dust parameters on UAV detection through the method provided in this embodiment, the dust interference level in different areas of the goaf can be accurately quantified. This provides reliable basic data support for subsequent analysis of the dynamic cross-interference between UAVs and ground robots and the generation of collaborative detection strategies. It effectively avoids the problem of UAV detection data distortion caused by dust interference, improves the accuracy of goaf detection information, and provides a scientific basis for the comprehensive assessment of goaf stability. By clarifying the interference intensity in different areas, UAVs can be guided to avoid high-interference areas in advance, reduce ineffective detection operations, improve detection efficiency, reduce the risk of equipment damage caused by UAV signal interference, and ensure the smooth implementation of collaborative detection operations.

[0047] In some embodiments, based on the water mist aggregation density, the thickness and uniformity of the water film formed between the ground robot's walking mechanism and the contact surface are analyzed, and the attenuation coefficient of the ground robot's friction force on the contact surface under different water film states is evaluated to obtain the degree of friction reduction. Based on the water mist coverage area and water mist retention time, the overlap ratio between the ground robot's predetermined walking path and the water mist coverage area and the dwell time in the gypsum mine area are analyzed to determine the probability of slippage under different overlap ratios and dwell times, and the slippage probability is obtained. By integrating the degree of friction reduction and the slippage probability, the impact of different areas on the ground robot's passage safety is comprehensively evaluated to obtain water mist risk information.

[0048] A ground robot's locomotion mechanism can be the mechanical structure used by a ground robot for movement. The contact surface refers to the interface between the ground robot's locomotion mechanism and the ground in the goaf area. The degree of friction reduction refers to the reduction in friction between the ground robot's locomotion mechanism and the contact surface under water mist conditions compared to a dry state.

[0049] Specifically, in the exploration of gypsum mine goaf areas, water mist is a key risk factor because it directly affects the mobility and stability of ground robots. Parameters such as water mist aggregation density, water mist coverage area, and water mist persistence duration are crucial for quantifying water mist risks, as they collectively determine the formation and persistence of the water film, thus affecting the friction between the robot and the ground. Without detailed analysis of these parameters, the ground robot may slip in wet areas, leading to mission interruption, equipment damage, or even safety accidents. This step addresses these issues by employing a combination of physical property analysis and contact mechanics simulation, based on water mist aggregation density data (e.g., 20 g / m³) extracted from remote gypsum mine scanning information in specific areas. 3This study uses infrared thermal imaging technology to reconstruct the adhesion state of water mist between the ground robot's walking mechanism (e.g., tracked wheels) and the contact surface (e.g., gypsum dust accumulation surface). It analyzes the thickness (e.g., 0.5 mm) and uniformity of the formed water film. Then, an indoor simulation experiment is conducted using a test platform with the same material as the contact surface in the goaf area to simulate the stress on the walking mechanism under different water film thicknesses. Friction changes are measured using a tension sensor to evaluate the corresponding friction attenuation coefficient (e.g., 0.3), thereby determining the degree of friction reduction in the area. Simultaneously, path planning algorithms and risk probability statistics are used to determine the water mist coverage area (e.g., 50 square meters) identified in remote scanning information and the water mist retention time obtained through time-series tracking (e.g., ...). For a period of 2 hours, the predetermined walking path of the ground robot is spatially overlaid with the water mist coverage area to calculate the path overlap ratio (e.g., 60%). Combined with the robot's preset moving speed (e.g., 0.5 m / s), the dwell time in the area is estimated (e.g., 30 minutes). Then, the passage records of the ground robot under similar historical conditions are retrieved to count the number of slip events under the same overlap ratio and dwell time conditions, and the slip probability is calculated (e.g., 25%). Finally, the weighting coefficients of the friction reduction degree and slip probability are determined by the analytic hierarchy process (e.g., 0.6 and 0.4 respectively). The passage risk of different areas is comprehensively assessed by weighted summation to generate water mist risk information including risk level, impact range, and duration.

[0050] The method provided in this embodiment quantifies and accurately assesses the risks of ground robot passage through systematic analysis of water mist aggregation density, coverage area, and retention time, avoiding the subjectivity and bias of existing experience-based judgments. The combination of physical simulation and statistical analysis ensures the scientific validity of the calculation of friction reduction and slippage probability, providing a reliable risk basis for subsequent development of UAV-ground robot collaborative detection strategies. The generated water mist risk information can accurately guide the robot to avoid high-risk areas and optimize its walking path, effectively reducing the risk of robot slippage, overturning, and other malfunctions, ensuring the continuity of close-range contact detection tasks. Simultaneously, this analysis process connects remote data with on-site execution, enabling the collaborative detection system to dynamically adapt to changes in the water mist environment of the goaf, improving the overall safety and efficiency of the detection task.

[0051] In some embodiments, based on the light shading intensity and the signal attenuation degree, the adsorption and sedimentation effect of high water mist accumulation in the UAV dust suppression operation area on suspended dust is analyzed to obtain the suppression effect strength of water mist on dust interference; based on the friction reduction degree and the slip probability, the blowing and drying effect of dust caused by ground robot movement on local water mist is analyzed to obtain the mitigation effect strength of dust on water mist risk; based on the suppression effect strength and the mitigation effect strength, the mutual constraint relationship between dust and water mist risks during UAV dust suppression operation and ground robot movement detection is cross-analyzed to generate an initial risk parameter set for goaf areas used to quantify the dynamic risk level of the area.

[0052] The intensity of the suppression effect can be defined as the strength of the adsorption and settling effect of water mist sprayed by drones on suspended dust in the goaf. The intensity of the mitigation effect can be defined as the strength of the dispersing and drying effect of water mist on the goaf caused by dust generated during the movement of ground robots. Drone dust suppression operations refer to the operation of drones equipped with dust suppression equipment spraying water mist on high-dust areas within the goaf to suppress dust diffusion and reduce detection interference. Ground robot mobile detection refers to the movement of ground robots within the goaf along a preset path. The interrelationship between dust and water mist risks can be defined as a dynamic balance between dust interference and water mist risks within the goaf, where "water mist suppresses dust, and dust weakens water mist," with the effects of both alternating and dominating as spatiotemporal conditions change.

[0053] Specifically, in the detection of gypsum mine goaf areas, dust interference and water mist risks do not exist independently, but rather influence each other through spatiotemporal coupling. For example, water mist generated by drone dust suppression operations may inhibit dust diffusion, but simultaneously increase the risk of slippage for ground robots. Conversely, dust stirred up by ground robot movement may disperse the water mist, reducing the risk of slippage, but may exacerbate dust obscuring interference with drones. If dust or water mist risks are assessed in isolation, this dynamic interaction will be overlooked, leading to inaccurate risk parameters and consequently affecting the formulation of collaborative detection strategies. For instance, it may result in the incorrect allocation of operating areas for drones and ground robots, increasing the probability of collisions or failures. This step addresses the aforementioned issues using the following methods: An effect assessment method is employed. Based on dust interference information such as the intensity of light obstruction (visibility 3 meters) and the signal attenuation (e.g., 40% reduction in effective transmission distance of laser ranging signals), combined with data showing a 75% decrease in dust concentration and an increase in visibility to 9 meters after the drone sprayed water mist in the area, the adsorption and sedimentation effect of the water mist on suspended dust is analyzed, quantifying a suppression effect strength of 0.75 (representing a 75% reduction in dust interference). Simultaneously, based on water mist risk information such as a friction attenuation coefficient of 0.5 (friction reduction coefficient) and a 25% slip probability, combined with data showing a 60% decrease in water mist accumulation density and a water film thickness reduction from 2 mm to 0.8 mm after the ground robot moved in the area, the blowing and drying effect of dust on the water mist is analyzed, quantifying a effect of 0.6 (representing a 60% reduction). The intensity of the mitigation effect of water mist risk was determined. Subsequently, a cross-analysis and comprehensive quantification method was adopted, with the intensity of suppression effect and the intensity of mitigation effect as the core indicators. Combined with the spatial distribution characteristics of the area located in the eastern part of the goaf, overlapping 80% with the robot's predetermined path, and the temporal variation pattern of weak air flow and 20% longer water mist retention time from 10:00 to 12:00 in the morning, the mutual constraint relationship between dust interference and water mist risk during the dust suppression operation of the UAV and the mobile detection of the ground robot was cross-analyzed (for example, when the intensity of suppression effect is 0.75, which is higher than the intensity of mitigation effect is 0.6, the regional risk is mainly the passage risk caused by water mist; when the intensity of mitigation effect rises to 0.8, it turns into the dust detection interference). This dynamic relationship was transformed into quantitative parameters including medium risk (risk value 0.55), an affected area of ​​60 square meters, and a duration of 30 minutes, and finally the initial risk parameter set of the goaf area was generated.

[0054] The method provided in this embodiment comprehensively evaluates the coupling effect of dust and water mist, generates an accurate set of risk parameters, improves the comprehensiveness and reliability of risk identification in goaf areas, and enables UAV-ground robot collaborative detection to better avoid high-risk areas, optimize detection paths, thereby improving overall detection efficiency and safety, and reducing equipment failures and data errors.

[0055] In some embodiments, based on the intensity of the suppression effect, the adsorption and sedimentation efficiency of water mist sprayed by the UAV on suspended dust in the dust suppression operation area is analyzed to obtain water mist dust suppression efficiency information; based on the water mist dust suppression efficiency information, combined with the intensity of the weakening effect, the degree of interference of dust on the water mist adsorption effect when the ground robot moves in the gypsum mine area is analyzed to obtain dust-water mist dynamic suppression balance characteristics; based on the intensity of the weakening effect, the blowing and drying rate of local water mist caused by dust caused by the movement of the ground robot is analyzed to obtain dust dehumidification efficiency information; based on the dust dehumidification efficiency information, combined with the intensity of the suppression effect, the degree of suppression of the dust diffusion effect by the newly generated water mist from the UAV dust suppression operation is analyzed to obtain water mist-dust dynamic weakening balance characteristics; integrating the dust-water mist dynamic suppression balance characteristics and the water mist-dust dynamic weakening balance characteristics, the dominant alternation law of dust interference and water mist risk under different spatiotemporal conditions is comprehensively analyzed to generate a dynamic cross-interference feature set.

[0056] The dust-water mist dynamic suppression balance characteristic can be a characteristic index reflecting the dynamic balance between dust interference and water mist suppression by combining the dust suppression efficiency of water mist with the degree of interference of dust on water mist adsorption. Dust dehumidification efficiency information can be a set of information formed by analyzing the dispersion and drying rate of local water mist in the goaf caused by dust generated by ground robot movement. The water mist-dust dynamic weakening balance characteristic can be a characteristic index reflecting the dynamic balance between water mist risk and dust weakening by combining the dust dehumidification efficiency with the degree of suppression of dust diffusion by water mist. The dynamic cross-interference characteristic set can be a comprehensive characteristic set used to quantify the dynamic interference risk in the goaf by integrating the dust-water mist dynamic suppression balance characteristic and the water mist-dust dynamic weakening balance characteristic, combined with the alternation pattern of their dominant effects under different spatiotemporal conditions.

[0057] Specifically, in the comprehensive exploration of gypsum mine goaf areas, the operational behaviors of drones and ground robots will influence each other. Ignoring the dynamic spatiotemporal coupling of dust and water mist generated by both will lead to decreased detection efficiency, increased equipment operational risks, and even mission failure. For example, while drone dust suppression can reduce dust concentration, excessive water mist increases the risk of ground robot slippage. Conversely, while ground robot movement can partially remove water mist, the resulting dust may interfere with drone detection signals. This step addresses these issues using the following method: First, based on the determined suppression effect strength (e.g., a quantization value of 0.65, representing a moderate suppression ability of water mist on dust), a ring... The environmental effect tracking and analysis method involves equipping a drone with dust concentration sensors and water mist flow sensors to monitor changes in dust concentration at different spray rates (e.g., 5 L / min, 8 L / min) within a dust suppression operation area (e.g., a 10m × 10m area in the middle of a goaf). The method analyzes the adsorption and settling efficiency of water mist on suspended dust (e.g., at a spray rate of 8 L / min, the dust settling efficiency reaches 70% within 10 minutes), thus generating information on the dust suppression effectiveness of water mist. Subsequently, combined with the weakening effect intensity (e.g., a quantified value of 0.4, representing a relatively weak weakening ability of dust on water mist), an interactive comparative evaluation method is used to record the ground robot's stationary and moving states (e.g., moving speed 0).At a speed of 4 m / s, the efficiency of water mist adsorption of dust in this area was measured. A comparison revealed that the water mist adsorption efficiency decreased from 70% to 55% after the robot moved. This analysis categorized the interference of dust on water mist adsorption, yielding dynamic suppression balance characteristics between dust and water mist. Simultaneously, based on the intensity of the mitigation effect, a dynamic concentration monitoring method was employed. Humidity sensors were deployed around the ground robot to record the rate of change in local water mist concentration under different movement paths (e.g., straight lines, turns). For example, during straight-line movement, the water mist concentration decreased by 6% every 15 minutes. This analysis analyzed the rate of dust dispersion and drying of local water mist, generating information on dust dehumidification efficiency. Furthermore, combined with the intensity of the suppression effect, a diffusion range comparison analysis was used. Image acquisition components on the UAV captured the dust diffusion range during the ground robot's movement. The changes in the dust diffusion range before and after the UAV sprayed additional water mist (e.g., a spray rate of 6 L / min) (e.g., shrinking from 8 m to 4 m) were observed, analyzing the new... The degree of suppression of dust dispersion effect by water mist was increased, and the dynamic weakening balance characteristics of water mist and dust were obtained. Finally, a spatiotemporal feature integration method was used to integrate the two balance characteristics. The area was divided into grid units according to the spatial layout of the goaf (e.g., 4m×4m) and time intervals according to the detection progress (e.g., every 20 minutes). The alternating pattern of the dominant roles of dust interference and water mist risk within each spatiotemporal unit was analyzed (e.g., in the eastern grid area from 20-40 minutes, frequent robot turns led to the dominant risk of dust, while in the western grid area from 40-60 minutes, continuous dust suppression by drones led to the dominant risk of water mist). A dynamic cross-interference feature set was generated. Then, a risk level quantification model was used to convert the qualitative descriptions such as "dominant risk type" and "risk change amplitude" in the feature set into specific parameters (e.g., dust dominant risk level is 3, risk change rate is 0.2 levels / minute), ultimately forming the dynamic cross-interference feature set.

[0058] The method provided in this embodiment can accurately capture the dynamic interaction between dust and water mist during collaborative detection by UAVs and ground robots, overcoming the limitations of existing static risk assessments that cannot adapt to changes in environment and operational status. The generated dynamic cross-interference feature set can provide a comprehensive and dynamic risk basis for subsequent collaborative detection strategies, avoiding the "one-sided" problem caused by single interference analysis, and effectively reducing the probability of UAV detection data failure and ground robot traffic accidents. At the same time, by analyzing the alternating patterns of the dominant role of interference under different spatiotemporal conditions, it can guide UAVs and ground robots to adjust operational parameters (such as water mist spray volume and movement speed) as needed, improve the resource utilization efficiency of collaborative detection, ensure the safety and comprehensiveness of goaf detection operations, and provide reliable basic support for subsequent comprehensive stability assessment of goaf areas.

[0059] In some embodiments, based on water mist dust suppression efficiency information, the adsorption efficiency of water mist sprayed by UAVs on suspended dust in the goaf area is analyzed over time to obtain a dynamic efficiency curve for water mist dust suppression; based on the weakening effect intensity, the influence of the ground robot's moving speed and path on the intensity of local dust generation is analyzed to obtain dynamic dust generation characteristics; based on the dynamic efficiency curve for water mist dust suppression, combined with the dynamic dust generation characteristics, the interference intensity of dust generated during the ground robot's movement on the water mist adsorption process, as well as the water mist's ability to suppress newly generated dust, are analyzed. The difference between the interference intensity and the suppression ability is used as the dust-water mist dynamic suppression balance index. The closer the dust-water mist dynamic suppression balance index is to zero, the higher the suppression coverage of water mist on dust interference; the dust-water mist dynamic suppression balance index is used as the dust-water mist dynamic suppression balance characteristic.

[0060] Movement can refer to the changes in parameters during the positional shift of the ground robot within the gypsum mine area along a preset or dynamically adjusted trajectory. Disturbance level can refer to the degree to which dust particles hinder or weaken the process of water mist adsorbing dust, representing the decrease in water mist adsorption efficiency. The dust-water mist dynamic suppression balance characteristic can be a characteristic parameter of the dynamic balance between the water mist's ability to suppress dust and the dust's ability to interfere with water mist adsorption. The water mist dust suppression dynamic efficiency curve can be a curve reflecting the change in adsorption efficiency over time after the UAV sprays water mist, with time as the horizontal axis and the water mist's dust adsorption efficiency as the vertical axis. Dust dynamic generation characteristics can be a feature reflecting the correlation between the ground robot's movement speed, movement path, and local dust generation intensity. Movement speed can be the distance the ground robot moves per unit time. Movement path can be the walking trajectory of the ground robot within the gypsum mine area. Dust generation intensity can be the amount of dust generated by the ground robot's movement per unit time. Disturbance intensity can be a numerical value quantifying the ability of dust to interfere with the water mist adsorption process. Suppression ability can be a numerical value quantifying the ability of water mist to adsorb and settle newly generated dust. The dust-water mist dynamic suppression balance index can be a quantitative indicator obtained by calculating the difference between the interference intensity and the suppression capability.

[0061] Specifically, when drones and ground robots collaboratively probe gypsum mine goaf areas, a complex dynamic coupling relationship exists between dust interference and water mist risk. If the degree of interference from dust on water mist adsorption is not precisely quantified, it may lead to poor dust suppression or low detection efficiency. Specifically, dust interference in the initial state of the goaf can obscure the drone's detection signal. While the water mist sprayed by the drone can adsorb dust, the movement of the ground robot can cause dust to disperse, thus interfering with the water mist adsorption process and creating a counterproductive effect. If this dynamic interaction is not monitored and adjusted in real time, it will reduce the coverage of water mist dust suppression, increasing detection risk. This step addresses the aforementioned issues using the following methods: First, targeting the initial dust disturbance in the gypsum mine goaf, a drone is used to conduct water mist dust suppression operations. Information on the water mist dust suppression efficiency is obtained, and a combination of real-time monitoring and data fitting is employed to analyze the changing trend of the adsorption efficiency of the drone-sprayed water mist on suspended dust in the goaf over time, generating a dynamic efficiency curve for water mist dust suppression. For example, in the initial spraying stage (0-5 minutes), the water mist adsorption efficiency rapidly increases from 0.15 to 0.75. Within 5-20 minutes, due to the gradual settling of the water mist, the adsorption efficiency slowly decreases to 0.3. Subsequently, based on the intensity of the weakening effect, a parametric variable analysis method is used to study the ground robot's moving speed (e.g., when the moving speed is 0.3 m / s, the dust generation intensity is 0.2 g / s; when the moving speed increases to 0.6 m / s, the dust generation intensity increases to 0.5 g / s) and moving path (e.g., when moving along a leveling path along the goaf sidewall, the dust generation intensity is 0.25 g / s). The influence of dust generation intensity (0.6 g / s) on local dust generation intensity when moving along the dust accumulation path at the bottom of the goaf is obtained to obtain the dynamic dust generation characteristics. Then, the dynamic efficiency curve of water mist dust suppression is coupled with the dynamic dust generation characteristics for analysis. By collecting parameters such as dust concentration and water mist density in real time, comparative analysis is used to determine the interference intensity (e.g., interference intensity of 0.4) of dust generated during the movement of the ground robot (e.g., dust generated by adjusting the path through the dust accumulation area) on the water mist adsorption process. At the same time, the suppression ability of these newly added dusts under the current water mist adsorption efficiency (e.g., the suppression ability is 0.42) is analyzed. The difference between the interference intensity and the suppression ability (i.e., 0.42-0.4=0.02) is used as the dust-water mist dynamic suppression balance index. Finally, the dust-water mist dynamic suppression balance index is used as the dust-water mist dynamic suppression balance characteristic.

[0062] The method provided in this embodiment utilizes data analysis and mathematical modeling to achieve real-time quantification of the dynamic interaction between dust and water mist, enabling drones and ground robots to adaptively adjust operating parameters, improve dust suppression and detection safety, and reduce energy waste and risk events.

[0063] In some embodiments, based on dust dehumidification efficiency information, the trend of the dust dispersion effect on local water mist caused by changes in the ground robot's moving speed and path over time is analyzed to obtain a dynamic efficiency curve for dust dehumidification. Based on the suppression effect intensity, the encapsulation and adsorption efficiency of water mist spraying rate and coverage area on newly added dust particles during UAV dust suppression operations is analyzed, and the weighted sum of water mist spraying rate and encapsulation and adsorption efficiency is used as the dynamic response intensity of water mist dust suppression. Based on the dynamic efficiency curve for dust dehumidification, combined with the dynamic response intensity of water mist dust suppression, the influence trend of dust generated during the ground robot's movement after being affected by the UAV and the suppression ability of water mist on dust are analyzed to obtain the dynamic weakening balance intensity of water mist-dust. Based on the dynamic weakening balance intensity of water mist-dust, combined with water mist risk information and dust interference information, the dominant relationship between water mist dust suppression and dust diffusion in the gypsum mine area is determined, and a dynamic weakening balance characteristic of water mist-dust is generated to characterize the balance state of dust suppression and dust weakening.

[0064] The newly added water mist can refer to the water mist newly sprayed during the drone dust suppression operation. The dust diffusion effect can refer to the phenomenon and impact of dust dispersion and distribution in the goaf area under the combined action of ground robot movement and drone airflow. The degree of suppression can refer to the degree to which the newly added water mist constrains and weakens the dust diffusion effect. The dynamic efficiency curve of dust dehumidification can be a curve reflecting the change in the local water mist dispersion effect of dust over time, with time as the horizontal axis and the dust dispersion intensity of water mist as the vertical axis. The water mist spraying rate can be the amount of water mist sprayed by the drone per unit time. The encapsulation and adsorption efficiency can be the degree of effectiveness of water mist in encapsulating and adsorbing newly added dust particles. The dynamic response intensity of water mist dust suppression can be the weighted sum of the comprehensive water mist spraying rate and encapsulation and adsorption efficiency, reflecting the immediate response capability of water mist dust suppression to dust. The dynamic weakening balance intensity of water mist-dust can be an indicator that quantifies the strength of the interaction between water mist dust suppression and dust diffusion. The dynamic weakening balance characteristics of water mist-dust can be a set of characteristics characterizing the balance state between water mist dust suppression and dust diffusion.

[0065] Specifically, during the exploration of gypsum mine goaf areas, there is a complex dynamic coupling relationship between dust interference (interference caused by robots in the initial state of space) and water mist dust suppression (dust suppression measures taken by drones when dust interference occurs). If only the effects of dust or water mist are considered in isolation, it is impossible to accurately reflect their interaction in the real environment (e.g., dust dispersion leads to a decrease in the detection efficiency of drones, or water mist accumulation increases the risk of slippage for ground robots). This step addresses the above problems through the following methods: For dust interference caused by the movement of ground robots in the initial spatial state, firstly, based on dust dehumidification efficiency information, trend analysis is used to analyze the changing trend of the dust dispersion effect on local water mist caused by changes in the ground robot's moving speed (e.g., from 0.5m / s to 1m / s) and moving path (e.g., from straight walking to curved detour) over time, generating a dynamic dust dehumidification efficiency curve; To cope with this dust interference, the drone initiates water mist dust suppression operations. At this time, based on the suppression effect intensity, quantitative calculation methods are used to analyze the encapsulation and adsorption efficiency (e.g., 80%) of newly added dust particles by the water mist spraying rate (e.g., 5L / min) and coverage area (e.g., 10 square meters) during the drone dust suppression operation, and the two are weighted according to preset weights (e.g., 0.4 and 0.6 respectively) through weighted calculation methods. The weighted sum is calculated as the dynamic response intensity of water mist dust suppression. Subsequently, considering that the combined action of the ground robot and the drone may lead to dust dispersion, a comprehensive analysis method is adopted based on the aforementioned dynamic efficiency curve of dust dehumidification, combined with the dynamic response intensity of water mist dust suppression. The diffusion trend of dust generated during the movement of the ground robot after being affected by the drone (e.g., diffusion from a local area of ​​5 square meters to 15 square meters) and the ability of water mist to suppress dust (e.g., the dust diffusion range can be controlled within 8 square meters) are analyzed to obtain the dynamic weakening balance intensity of water mist-dust. Finally, a cross-evaluation method is adopted, based on this balance intensity combined with water mist risk information (e.g., the probability of slippage in a certain area is 20%) and dust interference information (e.g., the signal attenuation degree in a certain area is 30%), to determine the dominant role of water mist dust suppression and dust dispersion in a specific area of ​​gypsum mine (e.g., the eastern area of ​​the goaf) (e.g., dust dispersion plays a major role in the early stage of operation, and water mist dust suppression becomes the major role after 10 minutes of operation), and a water mist-dust dynamic weakening balance feature is generated to characterize the weakening balance state of the two.

[0066] The method provided in this embodiment monitors the movement speed and path changes of the ground robot in real time, and generates a dynamic efficiency curve for dust removal and dehumidification. At the same time, the encapsulation and adsorption efficiency is calculated based on the water mist spraying rate and coverage area of ​​the UAV, and the dynamic response intensity of water mist dust suppression is further weighted to obtain the dynamic response intensity of water mist dust suppression. By using the coupling analysis of curve and intensity data, the dynamic balance relationship between the diffusion trend of dust under the influence of UAV airflow and the water mist suppression capability is evaluated. Finally, the dynamic weakening balance characteristics of water mist and dust are generated. By quantifying the interaction between dust suppression and dust, the adaptability to dynamic risks during collaborative detection is significantly improved, and the detection efficiency is avoided due to local dust diffusion or water mist accumulation. This enhances the stability and reliability of the system in complex environments.

[0067] In some embodiments, based on the dynamic suppression balance characteristics of dust and water mist, the changes in the suppression and counter-suppression intensity of water mist adsorption and ground robot-generated dust in the temporal dimension are analyzed to obtain the dynamic balance evolution trend of dust and water mist; based on the dynamic weakening balance characteristics of water mist and dust, the changes in the weakening and counter-weakening intensity of dust dispersion and UAV dust suppression water mist in the spatial dimension are analyzed to obtain the dynamic balance evolution trend of water mist and dust; based on the dynamic balance evolution trends of dust and water mist and dust, the alternation pattern and duration of the dust interference-dominated stage and the water mist risk-dominated stage in different regions during the detection process are cross-analyzed to obtain the detection dynamic risk-dominated stage sequence; based on the detection dynamic risk-dominated stage sequence, combined with dust interference information and water mist risk information, the comprehensive risk level of detection impact of each region in different time segments is quantified to generate a spatiotemporally refined dynamic cross-interference feature set.

[0068] The detection dynamic risk-dominant phase sequence can be a combination of phases dominated by dust interference or water mist risk in each region, arranged chronologically. The comprehensive risk level of detection impact can be a classification of the detection risk for each spatiotemporal segment, combining the levels of dust interference and water mist risk.

[0069] Specifically, in the process of UAV-ground robot collaborative exploration of gypsum mine goaf areas, dust interference and water mist risks do not exist in isolation, but rather exhibit a complex dynamic coupling relationship. Relying solely on single dust-water mist dynamic suppression balance characteristics or water mist-dust dynamic weakening balance characteristics cannot fully reflect the true risk state of the exploration scenario. This step addresses the above problem through the following method: Based on the acquired dust-water mist dynamic suppression balance characteristics (e.g., a balance index of 0.15 in a certain area, close to zero, indicating good water mist suppression effect), time-series analysis is used, combined with real-time monitoring data of dust concentration in the goaf area (e.g., a concentration of 8 mg / m³ at the start of exploration). 3 It dropped to 5 mg / m³ after 5 minutes. 3 Ten minutes later, the blood glucose level rose to 6 mg / m³ due to the robot's movement. 3The study analyzed the temporal changes of the adsorption strength of water mist on dust (e.g., adsorption strength is 0.8 in 0-5 minutes and decreases to 0.5 in 5-10 minutes) and the counter-inhibition strength of dust generated by the ground robot on water mist adsorption (e.g., counter-inhibition strength is 0.3 when the robot moves at a speed of 0.4 m / s and increases to 0.4 when the speed increases to 0.5 m / s). This yielded the dynamic equilibrium evolution trend of dust-water mist (e.g., 0-5 minutes: water mist adsorption is stronger than dust counter-inhibition → 5-10 minutes: the two strengths are close). Subsequently, based on the dynamic weakening equilibrium characteristics of water mist-dust (e.g., the equilibrium strength in this area is 0.65), a spatial distribution modeling method was used, combined with spatial structure data obtained from the 3D scanning of the goaf (e.g., the area is 10 meters long and 8 meters wide, with 3 narrow passages). This analysis examined the dispersing effect of dust generated by the ground robot at different locations on water mist (e.g., the dispersing effect covers 3 square meters in narrow passages and 5 square meters in open areas) and the counter-weakening effect of water mist sprayed by the drone at the corresponding locations (e.g., in narrow passages). The spatial distribution differences of the water mist attenuation intensity (0.7 in the central area and 0.5 in the open area) were analyzed to obtain the dynamic equilibrium evolution trend of water mist and dust (e.g., in the area center: water mist attenuation is stronger than dust dispersion → at the area edge: dust dispersion is stronger than water mist attenuation). Then, a cross-coupling analysis method was used to combine the temporal and spatial evolution trends. For each sub-area of ​​the goaf (e.g., divided into three sub-areas A, B, and C, each with an area of ​​approximately 25 square meters), the dominant risk type was analyzed at different time segments (e.g., 0-5 min, 5-10 min, 10-15 min). For example, in area A, the dust concentration was 7 mg / m³ in the 0-5 min period. 3 The water mist aggregation density is 0.2 g / m³. 3 Dust interference was the primary factor; within 5-10 minutes, the water mist aggregation density increased to 0.4 g / m³. 3 Dust concentration dropped to 4 mg / m³ 3 Water mist risk becomes dominant; within 10-15 minutes, dust concentration rises to 5 mg / m³ due to robot movement. 3 The water mist aggregation density decreased to 0.3 g / m³. 3Dust interference once again dominates, thus identifying the dominant phase sequence of detection dynamic risks in each region. Finally, based on this sequence, and combining previously acquired dust interference information (e.g., dust suspension duration in region B is 20 minutes) and water mist risk information (e.g., water mist retention duration in region B is 15 minutes), a risk assessment model is constructed using the analytic hierarchy process (AHP) to quantify the comprehensive risk level of detection impact at different time segments in each region: for example, region B is at medium risk (level 3) from 0-5 min, high risk (level 4) from 5-10 min, medium risk (level 3) from 10-15 min, and low risk (level 2) from 15-20 min. Ultimately, the spatiotemporal risk information of all regions is integrated to generate a spatiotemporally refined dynamic cross-interference feature set containing multi-dimensional information of "region-time-dominant risk-risk level".

[0070] By integrating two balance features and analyzing the alternation of dominant effects through the method provided in this embodiment, the generated dynamic cross-interference feature set achieves a refined "spatiotemporal dual-dimensional" characterization of goaf detection risks, avoiding risk misjudgment caused by single-dimensional or local balance analysis. The regionalized and temporal risk information provided, as well as the spatiotemporally refined risk quantification results, also provide detailed basic data for the comprehensive stability assessment report of goaf, effectively improving the scientificity and reliability of the assessment report, and thus ensuring the safety, efficiency, and integrity of the collaborative detection process.

[0071] Figure 3 This application provides a schematic diagram of the structure of a UAV-ground robot collaborative comprehensive detection system for goaf areas, as shown in one embodiment. Figure 3 As shown, the UAV-Ground Robot Collaborative Comprehensive Detection System 300 of this embodiment includes: an information acquisition module 301, a cross-analysis module 302, and a strategy generation module 303.

[0072] Information acquisition module 301 is used to acquire a remote gypsum mine scanning information set, and based on the remote gypsum mine scanning information set, analyze the dust interference information and water mist risk information in the goaf of the gypsum mine to obtain the initial risk parameter set of the goaf.

[0073] The cross-analysis module 302 is used to analyze the dynamic coupling relationship between the dust caused by the movement of the ground robot and the occupant detection of the UAV, and the water mist generated by the dust suppression operation of the UAV and the slippery interference of the ground robot, based on the initial risk parameter set of the goaf area, and to obtain a dynamic cross-interference feature set.

[0074] The strategy generation module 303 is used to generate a collaborative detection strategy of UAV-ground robot based on the dynamic cross-interference feature set, with the goal of avoiding high-risk areas and maximizing detection efficiency, and output a comprehensive assessment report on the stability of the goaf.

[0075] Optionally, when the information acquisition module 301 analyzes the dust interference information and water mist risk information in the goaf area based on the remote gypsum mine scanning information set to obtain the initial risk parameter set of the goaf area, it is specifically used for:

[0076] Based on the remote gypsum mine scanning information set, the dust concentration, dust distribution range, and dust suspension duration in different areas of the goaf are analyzed to determine the light shading intensity and detection signal attenuation of the detection components carried by the UAV, thus obtaining the dust interference information. Based on the remote gypsum mine scanning information set, the water mist accumulation density, water mist coverage area, and water mist retention duration in different areas of the goaf are analyzed to determine the degree of friction reduction and slippage probability between the ground robot's walking mechanism and the contact surface, thus obtaining the water mist risk information. Based on the dust interference information and the water mist risk information, the spatiotemporal coupling relationship between the two in the goaf is evaluated, and the initial risk parameter set of the goaf used to characterize the comprehensive risk level of the region is generated.

[0077] Optionally, when the information acquisition module 301 analyzes the dust concentration, dust distribution range, and dust suspension duration in different areas of the goaf based on the remote gypsum mine scanning information set, and obtains the dust interference information by analyzing the light shading intensity and detection signal attenuation of the detection components carried by the UAV, the module is specifically used for:

[0078] Based on the dust concentration, the degree of attenuation of the transmittance of visible light and infrared spectrum of the UAV detection component by dust particles is analyzed to obtain the light blocking intensity under different visibility levels; based on the dust distribution range and the dust suspension duration, the scattering and absorption effects of dust particles on the laser ranging signal and image acquisition signal of the UAV detection component are analyzed to determine the attenuation degree of effective signal transmission distance and the probability of image clarity reduction, and the signal attenuation degree is obtained; by integrating the light blocking intensity and the signal attenuation degree, the dust interference information is obtained.

[0079] Optionally, when the information acquisition module 301 analyzes the degree of weakening of friction and slippage probability between the ground robot's walking mechanism and the contact surface based on the remote gypsum mine scanning information set, the effect of water mist accumulation density, water mist coverage area, and water mist retention time in different areas of the goaf on the water mist risk information, it is specifically used for:

[0080] Based on the water mist aggregation density, the thickness and uniformity of the water film formed between the ground robot's walking mechanism and the contact surface are analyzed. The attenuation coefficient of the ground robot's friction force on the contact surface under different water film states is evaluated to obtain the degree of friction reduction. Based on the water mist coverage area and the water mist retention time, the overlap ratio between the ground robot's predetermined walking path and the water mist coverage area, as well as the dwell time in the gypsum mine area, are analyzed to determine the probability of slippage under different overlap ratios and dwell times, thus obtaining the slippage probability. By integrating the degree of friction reduction and the slippage probability, the impact of different areas on the ground robot's passage safety is comprehensively evaluated to obtain the water mist risk information.

[0081] Optionally, when the information acquisition module 301 assesses the spatiotemporal coupling relationship between the dust interference information and the water mist risk information in the goaf area based on the dust interference information and the water mist risk information, and generates the initial risk parameter set of the goaf area to characterize the comprehensive risk level of the area, it is specifically used for:

[0082] Based on the light shielding intensity and the signal attenuation degree, the adsorption and sedimentation effect of high water mist accumulation on suspended dust in the UAV dust suppression operation area is analyzed to obtain the suppression effect of water mist on dust interference. Based on the friction reduction degree and the slip probability, the blowing and drying effect of dust caused by ground robot movement on local water mist is analyzed to obtain the mitigation effect of dust on water mist risk. Based on the suppression effect strength and the mitigation effect strength, the mutual constraint relationship between dust and water mist risks during UAV dust suppression operation and ground robot movement detection is cross-analyzed to generate the initial risk parameter set of the goaf area for quantifying the dynamic risk level of the area.

[0083] Optionally, when the cross-analysis module 302 analyzes the dynamic coupling relationship between the dust generated by the ground robot's movement and the obstruction interference of the UAV's detection, and the slippery interference of the water mist generated by the UAV's dust suppression operation on the ground robot's passage, based on the initial risk parameter set of the goaf area, to obtain a dynamic cross-interference feature set, it is specifically used for:

[0084] Based on the intensity of the suppression effect, the adsorption and sedimentation efficiency of water mist sprayed by the UAV on suspended dust in the dust suppression operation area is analyzed to obtain water mist dust suppression efficiency information. Based on the water mist dust suppression efficiency information, combined with the intensity of the weakening effect, the degree of interference of dust on water mist adsorption when the ground robot moves in the gypsum mine area is analyzed to obtain dust-water mist dynamic suppression balance characteristics. Based on the intensity of the weakening effect, the blowing and drying rate of local water mist caused by dust caused by the movement of the ground robot is analyzed to obtain dust dehumidification efficiency information. Based on the dust dehumidification efficiency information, combined with the intensity of the suppression effect, the degree of suppression of the dust diffusion effect by the newly generated water mist from the UAV dust suppression operation is analyzed to obtain water mist-dust dynamic weakening balance characteristics. Integrating the dust-water mist dynamic suppression balance characteristics and the water mist-dust dynamic weakening balance characteristics, the dominant alternation law of dust interference and water mist risk under different spatiotemporal conditions is comprehensively analyzed to generate the dynamic cross-interference feature set.

[0085] Optionally, when the cross-analysis module 302 analyzes the degree of interference between dust and water mist adsorption during the movement of a ground robot in a gypsum mine area based on the water mist dust suppression effectiveness information and the intensity of the weakening effect, and obtains the dynamic suppression balance characteristics of dust-water mist, it is specifically used for:

[0086] Based on the water mist dust suppression efficiency information, the adsorption efficiency of water mist sprayed by the UAV on suspended dust in the goaf area over time is analyzed to obtain the dynamic efficiency curve of water mist dust suppression. Based on the weakening effect intensity, the influence of the ground robot's moving speed and path on the intensity of local dust generation is analyzed to obtain the dynamic dust generation characteristics. Based on the water mist dust suppression dynamic efficiency curve and the dynamic dust generation characteristics, the interference intensity of dust generated during the ground robot's movement on the water mist adsorption process and the water mist's ability to suppress newly generated dust are analyzed. The difference between the interference intensity and the suppression ability is used as the dust-water mist dynamic suppression balance index. The closer the dust-water mist dynamic suppression balance index is to zero, the higher the suppression coverage of water mist on dust interference. The dust-water mist dynamic suppression balance index is used as the dust-water mist dynamic suppression balance characteristic.

[0087] Optionally, when the cross-analysis module 302 analyzes the degree of suppression of the dust diffusion effect by the newly generated water mist from the UAV dust suppression operation based on the dust dehumidification efficiency information and the suppression effect intensity, and obtains the dynamic weakening balance characteristics of water mist-dust, it is specifically used for:

[0088] Based on the dust dehumidification efficiency information, the trend of the dust dispersion effect on local water mist caused by changes in the ground robot's moving speed and path is analyzed over time to obtain the dynamic efficiency curve of dust dehumidification. Based on the suppression effect intensity, the encapsulation and adsorption efficiency of water mist spraying rate and coverage area on newly added dust particles during UAV dust suppression operations is analyzed, and the weighted sum of the water mist spraying rate and the encapsulation and adsorption efficiency is taken as the dynamic response intensity of water mist dust suppression. Based on the dynamic efficiency curve of dust dehumidification, combined with the dynamic response intensity of water mist dust suppression, the influence trend of dust generated during the ground robot's movement after being affected by the UAV and the suppression ability of water mist on dust are analyzed to obtain the dynamic weakening balance intensity of water mist-dust.

[0089] Based on the dynamic weakening balance intensity of water mist and dust, combined with the water mist risk information and the dust interference information, the dominant relationship between water mist dust suppression and dust diffusion in the gypsum mine area is determined, and the dynamic weakening balance characteristic of water mist and dust is generated to characterize the balance state of dust suppression and dust weakening.

[0090] Optionally, when the strategy generation module 303 integrates the dynamic suppression balance features of dust and water mist with the dynamic weakening balance features of water mist and dust, and comprehensively analyzes the alternating patterns of the dominant roles of dust interference and water mist risk under different spatiotemporal conditions to generate the dynamic cross-interference feature set, it is specifically used for:

[0091] Based on the dynamic suppression balance characteristics of dust and water mist, the changes in the suppression and counter-suppression intensity of water mist adsorption and ground robot dust in the temporal dimension are analyzed to obtain the evolution trend of the dynamic balance of dust and water mist. Based on the dynamic weakening balance characteristics of water mist and dust, the changes in the weakening and counter-weakening intensity of dust dispersion and UAV dust suppression water mist in the spatial dimension are analyzed to obtain the evolution trend of the dynamic balance of water mist and dust. Based on the evolution trends of the dynamic balance of dust and water mist and the dynamic balance of water mist and dust, the alternation pattern and duration of the dust interference-dominant stage and the water mist risk-dominant stage in different regions during the detection process are cross-analyzed to obtain the detection dynamic risk-dominant stage sequence. Based on the detection dynamic risk-dominant stage sequence, combined with the dust interference information and the water mist risk information, the comprehensive risk level of the detection impact of each region in different time segments is quantified to generate the spatiotemporally refined dynamic cross-interference feature set.

[0092] The system in this embodiment can be used to execute the methods of any of the above embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

Claims

1. A method for comprehensive detection of goaf areas using a drone-ground robot collaborative approach, characterized in that, include: Acquire a remote gypsum mine scanning information set, and based on the remote gypsum mine scanning information set, analyze the dust interference information and water mist risk information in the goaf of the gypsum mine to obtain the initial risk parameter set of the goaf; Based on the initial risk parameter set of the goaf, the dynamic coupling relationship between the dust caused by the movement of the ground robot and the obstruction interference of the drone detection, and the water mist generated by the drone dust suppression operation and the slippery interference of the ground robot passage is analyzed, and a dynamic cross-interference feature set is obtained. Based on the dynamic cross-interference feature set, with the goal of avoiding high-risk areas and maximizing detection efficiency, a collaborative detection strategy of UAV-ground robot is generated, and a comprehensive assessment report on the stability of the goaf area is output.

2. The method according to claim 1, characterized in that, Based on the remote gypsum mine scanning information set, the dust disturbance information and water mist risk information in the gypsum mine goaf are analyzed to obtain the initial risk parameter set of the goaf, including: Based on the remote gypsum mine scanning information set, the dust concentration, dust distribution range and dust suspension duration in different areas of the goaf are analyzed to determine the light shading intensity and detection signal attenuation of the detection components carried by the UAV, and the dust interference information is obtained. Based on the remote gypsum mine scanning information set, the water mist accumulation density, water mist coverage area and water mist retention time in different areas of the goaf are analyzed to determine the degree of weakening of friction between the ground robot walking mechanism and the contact surface and the probability of slippage, so as to obtain the water mist risk information. Based on the dust interference information and the water mist risk information, the spatiotemporal coupling relationship between the two in the goaf is assessed, and an initial risk parameter set for the goaf is generated to characterize the comprehensive risk level of the region.

3. The method according to claim 2, characterized in that, Based on the remote gypsum mine scanning information set, the dust concentration, dust distribution range, and dust suspension duration in different areas of the goaf are analyzed to determine the light shading intensity and detection signal attenuation of the detection components carried by the UAV, thereby obtaining the dust interference information, including: Based on the dust concentration, the degree of attenuation of the transmittance of dust particles on the visible light and infrared spectrum of the UAV detection components was analyzed to obtain the light blocking intensity under different visibility levels. Based on the dust distribution range and the dust suspension duration, the scattering and absorption effects of dust particles on the laser ranging signal and image acquisition signal of the UAV detection component are analyzed to determine the attenuation degree of the effective transmission distance of the signal and the probability of image clarity reduction, and the signal attenuation degree is obtained. By integrating the light blocking intensity and the signal attenuation, the dust interference information is obtained.

4. The method according to claim 2, characterized in that, Based on the remote gypsum mine scanning information set, the degree of weakening of friction and slippage probability between the ground robot's walking mechanism and the contact surface is analyzed by considering the water mist accumulation density, water mist coverage area, and water mist retention duration in different areas of the goaf, thus obtaining the water mist risk information, including: Based on the water mist aggregation density, the thickness and uniformity of the water film formed between the ground robot's walking mechanism and the contact surface are analyzed, and the attenuation coefficient of the ground robot's friction force on the contact surface under different water film states is evaluated to obtain the degree of friction force reduction. Based on the water mist coverage area and the water mist retention time, the overlap ratio between the ground robot's predetermined walking path and the water mist coverage area and the dwell time in the gypsum mine area are analyzed to determine the probability of slippage under different overlap ratios and dwell times, and thus obtain the slippage probability. By integrating the degree of friction reduction with the probability of slippage, the impact of different areas on the safety of ground robot passage is comprehensively evaluated to obtain the water mist risk information.

5. The method according to claim 4, characterized in that, The process of assessing the spatiotemporal coupling relationship between the dust disturbance information and the water mist risk information in the goaf, and generating an initial risk parameter set for the goaf to characterize the overall risk level of the region, includes: Based on the light blocking intensity and the signal attenuation degree, the adsorption and sedimentation effect of high water mist accumulation in the drone dust suppression operation area on suspended dust is analyzed, and the intensity of the water mist's suppression effect on dust interference is obtained. Based on the degree of friction reduction and the slippage probability, the dust generated by the movement of the ground robot has a blowing and drying effect on local water mist, and the strength of the dust reduction effect on water mist risk is obtained. Based on the intensity of the suppression effect and the intensity of the weakening effect, the mutual constraints between dust and water mist risks during the dust suppression operation of UAVs and the mobile detection of ground robots are cross-analyzed to generate the initial risk parameter set of the goaf area for quantifying the dynamic risk level of the region.

6. The method according to claim 5, characterized in that, Based on the initial risk parameter set of the goaf area, the dynamic coupling relationship between the dust generated by the movement of the ground robot and the obstruction interference of the drone's detection, and the water mist generated by the drone's dust suppression operation and the slippery interference of the ground robot's passage is analyzed, resulting in a dynamic cross-interference feature set, including: Based on the intensity of the suppression effect, the adsorption and sedimentation efficiency of water mist sprayed by the UAV on suspended dust in the dust suppression operation area was analyzed to obtain information on the dust suppression efficiency of water mist. Based on the information on the dust suppression efficiency of water mist and the intensity of the weakening effect, the degree of interference of dust on the adsorption of water mist when the ground robot moves in the gypsum mine area is analyzed, and the dynamic suppression balance characteristics of dust and water mist are obtained. Based on the intensity of the weakening effect, the rate at which dust caused by the movement of the ground robot disperses and dries local water mist is analyzed to obtain information on dust dehumidification efficiency. Based on the dust dehumidification efficiency information and the intensity of the suppression effect, the degree of suppression of the dust diffusion effect by the newly generated water mist from the drone dust suppression operation is analyzed, and the dynamic weakening balance characteristics of water mist and dust are obtained. By integrating the dynamic suppression balance characteristics of dust and water mist with the dynamic weakening balance characteristics of water mist and dust, and comprehensively analyzing the alternating patterns of the dominant roles of dust interference and water mist risk under different spatiotemporal conditions, the dynamic cross-interference feature set is generated.

7. The method according to claim 6, characterized in that, Based on the water mist dust suppression efficiency information and the intensity of the weakening effect, the degree of interference between dust and water mist adsorption when the ground robot moves in the gypsum mine area is analyzed, and the dynamic suppression balance characteristics of dust and water mist are obtained, including: Based on the aforementioned water mist dust suppression efficiency information, the adsorption efficiency of water mist sprayed by UAV on suspended dust in the goaf area over time was analyzed to obtain the dynamic efficiency curve of water mist dust suppression. Based on the strength of the weakening effect, the influence of the ground robot's moving speed and moving path on the intensity of local dust generation is analyzed to obtain the dynamic generation characteristics of dust. Based on the dynamic efficiency curve of water mist dust suppression, combined with the dynamic generation characteristics of dust, the interference intensity of dust generated during the movement of the ground robot on the water mist adsorption process, as well as the ability of water mist to suppress newly generated dust, are analyzed. The difference between the interference intensity and the suppression ability is used as the dust-water mist dynamic suppression balance index. The closer the dust-water mist dynamic suppression balance index is to zero, the higher the suppression coverage of water mist on dust interference. The dust-water mist dynamic suppression balance index is used as the dust-water mist dynamic suppression balance feature.

8. The method according to claim 6, characterized in that, Based on the dust dehumidification efficiency information and the intensity of the suppression effect, the degree of suppression of the dust diffusion effect by the newly generated water mist from the drone dust suppression operation is analyzed, and the dynamic weakening balance characteristics of water mist and dust are obtained, including: Based on the dust dehumidification efficiency information, the trend of the dust dispersion effect on local water mist caused by changes in the ground robot's moving speed and path is analyzed over time, and the dynamic efficiency curve of dust dehumidification is obtained. Based on the intensity of the suppression effect, the efficiency of water mist spraying rate and coverage area in the dust suppression operation of UAVs for the encapsulation and adsorption of newly added dust particles is analyzed, and the weighted sum of the water mist spraying rate and the encapsulation and adsorption efficiency is taken as the dynamic response intensity of water mist dust suppression. Based on the dynamic efficiency curve of dust dehumidification and the dynamic response intensity of water mist dust suppression, the influence trend of dust generated during the movement of the ground robot after being affected by the drone and the suppression ability of water mist on dust are analyzed to obtain the dynamic weakening balance intensity of water mist-dust. Based on the dynamic weakening balance intensity of water mist and dust, combined with the water mist risk information and the dust interference information, the dominant relationship between water mist dust suppression and dust diffusion in the gypsum mine area is determined, and the dynamic weakening balance characteristic of water mist and dust is generated to characterize the balance state of dust suppression and dust weakening.

9. The method according to claim 6, characterized in that, The integration of the dynamic suppression balance characteristics of dust and water mist and the dynamic weakening balance characteristics of water mist and dust, and the comprehensive analysis of the alternating patterns of the dominant roles of dust interference and water mist risk under different spatiotemporal conditions, generates the dynamic cross-interference feature set, including: Based on the dynamic inhibition balance characteristics of dust and water mist, the changes in the inhibition and counter-inhibition intensity of water mist adsorption and dust generated by ground robot movement in the time dimension are analyzed to obtain the evolution trend of the dynamic balance of dust and water mist. Based on the dynamic weakening balance characteristics of water mist and dust, the weakening and counter-weakening intensity changes of dust blowing effect and UAV dust suppression water mist in the spatial dimension are analyzed to obtain the evolution trend of dynamic balance between water mist and dust. Based on the dynamic equilibrium evolution trends of dust-water mist and water mist-dust, the alternation patterns and durations of the dust interference-dominant stage and the water mist risk-dominant stage in different regions during the detection process are cross-analyzed to obtain the detection dynamic risk-dominant stage sequence. Based on the dynamic risk-dominant phase sequence of the detection, combined with the dust interference information and the water mist risk information, the comprehensive risk level of the detection impact of each region in different time segments is quantified, and the spatiotemporally refined dynamic cross-interference feature set is generated.

10. A comprehensive detection system for goaf areas using a drone-ground robot collaborative approach, characterized in that, The method applied to any one of claims 1-9 includes: The information acquisition module is used to acquire a remote gypsum mine scanning information set, and based on the remote gypsum mine scanning information set, analyze the dust interference information and water mist risk information in the goaf of the gypsum mine to obtain the initial risk parameter set of the goaf. The cross-analysis module is used to analyze the dynamic coupling relationship between the dust caused by the movement of the ground robot and the obstruction interference of the drone detection, and the water mist generated by the drone's dust suppression operation and the slippery interference of the ground robot's passage, based on the initial risk parameter set of the goaf area, and to obtain a dynamic cross-interference feature set. The strategy generation module is used to generate a collaborative detection strategy of UAV-ground robot based on the dynamic cross-interference feature set, with the goal of avoiding high-risk areas and maximizing detection efficiency, and output a comprehensive assessment report on the stability of the goaf.

Images