Traffic geological disaster unmanned aerial vehicle sensor adaptive configuration method and system

Abstract

The present application relates to the technical field of geological disaster monitoring, and discloses a traffic geological disaster unmanned aerial vehicle sensor adaptive configuration method and system. The method comprises the following steps: obtaining disaster characteristics and environmental parameters of a target area, quantifying sensor capability, establishing an adaptation measurement matrix between the disaster characteristics and the sensor capability, generating a sensor working parameter configuration scheme, controlling an unmanned aerial vehicle to perform monitoring and collect data to optimize the weight coefficients of the adaptation measurement matrix. The present application can adaptively adjust the sensor configuration according to the disaster characteristics and the environmental conditions, improve the monitoring accuracy and efficiency, and reduce the monitoring cost.

Description

Technical Field

[0001] This invention relates to the field of geological disaster monitoring technology, and in particular to an adaptive configuration method and system for unmanned aerial vehicle (UAV) sensors for traffic geological disasters. Background Technology

[0002] With the increasing frequency of natural disasters and the continuous advancement of transportation infrastructure construction, monitoring of geological hazards in transportation has become a crucial aspect of ensuring traffic safety. In recent years, drones, with their advantages of mobility, wide coverage, and low cost, have gradually become an important technological means for monitoring geological hazards in transportation. Traditional monitoring of geological hazards in transportation mainly relies on manual on-site investigations or fixed monitoring equipment, which is inefficient and poses safety hazards. Drones, equipped with various types of sensors, can quickly acquire multi-dimensional information from disaster sites, providing vital data support for disaster assessment and emergency decision-making.

[0003] Unmanned aerial vehicle (UAV) monitoring systems are typically equipped with various types of sensors, including optical cameras, infrared thermal imagers, lidar, and multispectral sensors, capable of collecting data on multiple disaster indicators such as surface deformation, hydrological characteristics, and temperature changes. With the development of sensor technology, the technical performance of UAV monitoring systems has continuously improved, and their application scope has become increasingly wide. However, existing UAV monitoring technologies for traffic and geological disasters still have the following shortcomings: Sensor configuration lacks specificity; most systems use fixed sensor combinations, unable to flexibly adjust according to different types of geological disaster characteristics, resulting in low matching between monitoring data and actual disaster characteristics, affecting monitoring effectiveness and judgment accuracy. Sensor operating parameter settings lack environmental adaptability, failing to fully consider the impact of dynamic environmental changes on sensor performance. For example, when factors such as weather conditions and terrain complexity change, sensor operating parameters cannot be automatically adjusted, leading to a decline in monitoring data quality in complex environments. Existing technologies lack a quantitative mapping mechanism between sensor capabilities and disaster characteristics, making it impossible to achieve precise optimization of sensor configuration. Settings often rely on manual experience, lacking scientific theoretical support and adaptive optimization capabilities, making it difficult to meet the needs of efficient and accurate disaster monitoring, especially in complex and changing disaster environments where optimal monitoring performance is difficult to achieve. Summary of the Invention

[0004] This invention provides an adaptive configuration method and system for drone sensors for traffic and geological disasters, which can solve the problems in the prior art.

[0005] A first aspect of the present invention provides an adaptive configuration method for unmanned aerial vehicle (UAV) sensors for traffic and geological disaster relief, comprising:

[0006] Obtain the geological hazard type characteristics and environmental dynamic parameters of the target monitoring area;

[0007] Based on the preset sensor performance constraint rules and the mapping relationship between detection capabilities, the capabilities of various types of sensors carried by the UAV are quantitatively characterized to obtain a set of sensor capability vectors;

[0008] Based on the geological hazard type characteristics, each capability vector in the sensor capability vector set is processed to establish a quantitative adaptation relationship between the hazard feature dimension and the sensor capability dimension, thus obtaining an adaptation metric matrix;

[0009] Based on the adaptation metric matrix and the environmental dynamic parameters, a sensor operating parameter configuration scheme is generated by coordinating solutions through coupled multi-dimensional constraints.

[0010] Based on the sensor operating parameter configuration scheme, the working status of each sensor of the UAV is controlled, and the UAV is driven to perform monitoring tasks along a predetermined route and collect geological disaster detection data. The geological disaster detection data is used to optimize and adjust the weight coefficients of the corresponding disaster feature dimensions in the adaptation metric matrix.

[0011] Based on the preset sensor performance constraint rules and the mapping relationship between detection capabilities, the capabilities of various types of sensors carried by the UAV are quantitatively characterized, resulting in a set of sensor capability vectors, including:

[0012] Extracting the physical characteristic parameters of each sensor from the various types of sensors carried by the drone;

[0013] Based on preset sensor performance constraint rules, the mapping function relationship between physical characteristic parameters and capability characterization quantities is determined, and the physical characteristic parameters are processed to obtain standardized values.

[0014] Based on the mapping function relationship, the standardized values ​​are substituted into the corresponding transformation rules to generate the detection range function, resolution function and data acquisition rate function as capability characterization quantities.

[0015] Based on the positions of the effective detection distance boundary and angular resolution boundary in the physical characteristic parameters within a preset threshold, multiple types of sensors are classified into different capability categories;

[0016] The capability representation quantity is associated and combined with the capability category to construct a sensor capability vector set.

[0017] Based on the geological hazard type characteristics, each capability vector in the sensor capability vector set is processed to establish a quantitative adaptation relationship between the hazard feature dimension and the sensor capability dimension, resulting in an adaptation metric matrix, including:

[0018] Information on the spatial distribution morphology of geological hazards and information identifying the stages of hazard evolution are extracted from the characteristics of the geological hazard types.

[0019] The spatial distribution pattern information of the disaster is converted into spatial scale feature dimension and spatial complexity feature dimension, and the disaster evolution stage identification information is converted into temporal evolution rate feature dimension and stability feature dimension, thus obtaining a disaster feature dimension set;

[0020] Extract the detection range function, resolution function, and data acquisition rate function carried by each capability vector from the set of sensor capability vectors;

[0021] The detection range function is mapped to a spatial coverage capability dimension, the resolution function is mapped to a detail recognition capability dimension, and the data acquisition rate function is mapped to a temporal tracking capability dimension, thus obtaining a set of sensor capability dimensions;

[0022] A matching rule is determined between the disaster feature dimension set and the sensor capability dimension set. Based on the matching rule, the correlation strength between each feature dimension in the disaster feature dimension set and each capability dimension in the sensor capability dimension set is calculated to generate an adaptation metric matrix. Each element in the adaptation metric matrix represents the degree of dependence of a specific disaster feature dimension on a specific sensor capability dimension.

[0023] The spatial distribution morphology information of the disaster is converted into spatial scale feature dimension and spatial complexity feature dimension, and the disaster evolution stage identification information is converted into temporal evolution rate feature dimension and stability feature dimension, resulting in a disaster feature dimension set, including:

[0024] Based on the aforementioned disaster spatial distribution morphology information, mapping rules between geometric boundaries and spatial scale, as well as mapping rules between internal density distribution and spatial complexity, are determined;

[0025] According to the mapping rules, the spatial distribution pattern information of the disaster is converted into spatial scale feature dimension and spatial complexity feature dimension;

[0026] Extract the state transition nodes between different evolution stages and the duration data of each evolution stage from the disaster evolution stage identification information;

[0027] Evolution rate mutation points are calculated based on the time intervals of the state transition nodes, and stability imbalance intervals are identified based on the fluctuation amplitude of the duration data.

[0028] The correspondence between the distribution density of the evolution rate mutation points and the temporal evolution rate feature dimension, and the correspondence between the duration of the stability imbalance interval and the stability feature dimension are determined to obtain the association rules between evolutionary stages and temporal features.

[0029] According to the association rules, the disaster evolution stage identification information is converted into a time evolution rate feature dimension and a stability feature dimension, and the spatial scale feature dimension, the spatial complexity feature dimension, the time evolution rate feature dimension and the stability feature dimension are combined into a disaster feature dimension set.

[0030] Based on the adaptation metric matrix and the environmental dynamic parameters, a sensor operating parameter configuration scheme is generated through collaborative solution by coupling multi-dimensional constraints, including:

[0031] Based on the adaptation metric values ​​corresponding to each disaster feature dimension in the adaptation metric matrix, the disaster monitoring requirement constraints are determined;

[0032] Based on the environmental factors affecting the sensor's operating state in the aforementioned environmental dynamic parameters, the allowable adjustment range of the sensor's operating parameters is determined, thus obtaining environmental adaptability constraints;

[0033] The disaster monitoring requirement constraint and the environmental adaptability constraint are combined to form a multi-dimensional constraint set;

[0034] For the multi-dimensional constraint set, a constraint coupling relationship expression is determined, and a mapping function between the sensor operating parameters and the multi-dimensional constraint set is determined through the constraint coupling relationship expression.

[0035] Within the feasible region defined by the mapping function, the adaptation metric values ​​in the adaptation metric matrix are optimized to determine the sensor operating parameter combination that enables the adaptation metric values ​​to reach the target state.

[0036] The sensor operating parameter combinations are converted into operating mode setting parameters and data acquisition control parameters for each sensor to obtain a sensor operating parameter configuration scheme.

[0037] For the multi-dimensional constraint set, determine the constraint coupling relationship expression, and use the constraint coupling relationship expression to determine the mapping function between the sensor operating parameters and the multi-dimensional constraint set, including:

[0038] The requirement boundaries for sensor operating parameters are determined based on the aforementioned disaster monitoring requirements, and the limitation boundaries for sensor operating parameters are determined based on the aforementioned environmental adaptability constraints.

[0039] Based on the demand boundary and the constraint boundary, identify the constraint conflict area where the range of sensor operating parameters required by the disaster monitoring demand constraints and the range of sensor operating parameters allowed by the environmental adaptability constraints have insufficient overlap;

[0040] For the constraint conflict area, calculate the degree of deviation between the demand boundary and the restriction boundary, and determine the priority coefficients of the disaster monitoring demand constraint and the environmental adaptability constraint based on the degree of deviation;

[0041] Based on the priority coefficients, a weighted combination of the disaster monitoring requirement constraints and the environmental adaptability constraints is performed to determine the constraint coupling relationship expression;

[0042] The constraint coupling relationship expression is converted into the boundary condition equation of the sensor operating parameters. The feasible domain boundary of the sensor operating parameters is obtained by solving the boundary condition equation, and the mapping function between the sensor operating parameters and the multi-dimensional constraint set is determined.

[0043] Based on the sensor operating parameter configuration scheme, the operating status of each sensor of the UAV is controlled, and the UAV is driven to perform monitoring tasks and collect geological disaster detection data along a predetermined route, including:

[0044] The operating mode setting parameters and data acquisition control parameters of each sensor are extracted from the sensor operating parameter configuration scheme. The operating mode setting parameters are converted into sensor start commands and sensor operating state switching commands, and the data acquisition control parameters are converted into data acquisition frequency control commands and data acquisition accuracy control commands.

[0045] The system sends corresponding sensor start-up commands, sensor working state switching commands, data acquisition frequency control commands, and data acquisition accuracy control commands to each sensor mounted on the UAV, so that each sensor enters the working state according to the sensor working parameter configuration scheme.

[0046] The system acquires waypoint location information and flight segment time information for a predetermined flight route. It then associates the waypoint location information and flight segment time information with the data acquisition frequency control commands of each sensor to determine the synchronization correspondence between the flight route advancement nodes and the data acquisition trigger nodes. This synchronization correspondence is used to characterize the timing rules that trigger the corresponding sensor to perform data acquisition actions when the UAV reaches a specific waypoint location.

[0047] The drone is driven to fly along the predetermined route. When the drone reaches the route advancement node defined in the synchronization correspondence, the corresponding data acquisition trigger node is triggered, and the corresponding sensor is controlled to collect the geological disaster detection data.

[0048] A second aspect of the present invention provides an adaptive configuration system for unmanned aerial vehicle (UAV) sensors for traffic geological disasters, comprising:

[0049] The first unit is used to acquire the geological hazard type characteristics and environmental dynamic parameters of the target monitoring area;

[0050] The second unit is used to perform capability quantification and characterization of multiple types of sensors carried by the UAV based on preset sensor performance constraint rules and detection capability mapping relationship, and obtain a set of sensor capability vectors;

[0051] The third unit is used to process each capability vector in the sensor capability vector set according to the geological disaster type characteristics, establish a quantitative adaptation relationship between the disaster feature dimension and the sensor capability dimension, and obtain an adaptation metric matrix;

[0052] The fourth unit is used to generate a sensor operating parameter configuration scheme by coordinating the solution through coupling multi-dimensional constraints based on the adaptation metric matrix and the environmental dynamic parameters.

[0053] The fifth unit is used to control the working status of each sensor of the UAV based on the sensor working parameter configuration scheme, drive the UAV to perform monitoring tasks along a predetermined route and collect geological disaster detection data, and the geological disaster detection data is used to optimize and adjust the weight coefficients of the corresponding disaster feature dimensions in the adaptation metric matrix.

[0054] A third aspect of the present invention provides an electronic device, comprising:

[0055] processor;

[0056] Memory used to store processor-executable instructions;

[0057] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0058] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0059] This invention achieves accurate identification and parameterized characterization of different geological disaster scenarios by acquiring geological disaster type characteristics and environmental dynamic parameters of the target monitoring area, providing a scientific basis for subsequent sensor configuration. By quantitatively characterizing capabilities based on preset sensor performance constraint rules and the mapping relationship between detection capabilities, it achieves standardized measurement of the detection capabilities of multiple types of UAV sensors, solving the technical problem of difficulty in uniformly comparing the performance indicators of different sensors. By establishing a quantitative adaptation relationship between disaster characteristic dimensions and sensor capability dimensions, an adaptation metric matrix is ​​formed, establishing a quantitative mapping mechanism between sensor selection and disaster characteristics, improving the scientific rigor and relevance of sensor configuration. By coupling multi-dimensional constraints and collaboratively solving the adaptation metric matrix with environmental dynamic parameters, a dynamic sensor operating parameter configuration scheme is generated, overcoming the limitations of traditional fixed configuration schemes in dealing with complex and variable geological disaster environments. By using the collected geological disaster detection data to adjust the weight coefficients in the adaptation metric matrix, a closed-loop adaptive optimization mechanism is constructed, enabling continuous improvement in the accuracy of sensor configuration and the effectiveness of disaster detection as the monitoring task progresses.

[0060] In summary, this invention establishes a quantitative adaptation mechanism and a closed-loop optimization feedback mechanism between disaster characteristics and sensor capabilities, thereby enabling intelligent and adaptive configuration of UAV sensors during the monitoring of geological disasters in transportation, and improving the accuracy, efficiency, and flexibility of geological disaster monitoring. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating the adaptive configuration method for unmanned aerial vehicle (UAV) sensors for traffic and geological disasters according to an embodiment of the present invention.

[0062] Figure 2 This is a schematic flowchart illustrating the sensor operating parameter configuration scheme according to an embodiment of the present invention. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0065] Figure 1This is a flowchart illustrating the adaptive configuration method for unmanned aerial vehicle (UAV) sensors for traffic and geological disasters, according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0066] Obtain the geological hazard type characteristics and environmental dynamic parameters of the target monitoring area;

[0067] Based on the preset sensor performance constraint rules and the mapping relationship between detection capabilities, the capabilities of various types of sensors carried by the UAV are quantitatively characterized to obtain a set of sensor capability vectors;

[0068] Based on the geological hazard type characteristics, each capability vector in the sensor capability vector set is processed to establish a quantitative adaptation relationship between the hazard feature dimension and the sensor capability dimension, thus obtaining an adaptation metric matrix;

[0069] Based on the adaptation metric matrix and the environmental dynamic parameters, a sensor operating parameter configuration scheme is generated by coordinating solutions through coupled multi-dimensional constraints.

[0070] Based on the sensor operating parameter configuration scheme, the working status of each sensor of the UAV is controlled, and the UAV is driven to perform monitoring tasks along a predetermined route and collect geological disaster detection data. The geological disaster detection data is used to optimize and adjust the weight coefficients of the corresponding disaster feature dimensions in the adaptation metric matrix.

[0071] In one optional implementation, based on a preset mapping relationship between sensor performance constraints and detection capabilities, the capabilities of various types of sensors carried by the UAV are quantitatively characterized to obtain a set of sensor capability vectors, including:

[0072] Extracting the physical characteristic parameters of each sensor from the various types of sensors carried by the drone;

[0073] Based on preset sensor performance constraint rules, the mapping function relationship between physical characteristic parameters and capability characterization quantities is determined, and the physical characteristic parameters are processed to obtain standardized values.

[0074] Based on the mapping function relationship, the standardized values ​​are substituted into the corresponding transformation rules to generate the detection range function, resolution function and data acquisition rate function as capability characterization quantities.

[0075] Based on the positions of the effective detection distance boundary and angular resolution boundary in the physical characteristic parameters within a preset threshold, multiple types of sensors are classified into different capability categories;

[0076] The capability representation quantity is associated and combined with the capability category to construct a sensor capability vector set.

[0077] First, the physical characteristic parameters of each sensor are extracted from the various types of sensors carried by the UAV. For commonly used photoelectric detection equipment on UAVs, the extracted physical characteristic parameters mainly include sensor model, detection wavelength range, visible / infrared spectral response range, lens focal length, aperture parameters, field of view, detection distance upper and lower limits, angular resolution, and data output frame rate. For example, for visible light cameras, parameters such as CMOS sensor size (e.g., 1 / 2.3 inch), effective pixel count (e.g., 12 million pixels), lens focal length (e.g., 35mm), maximum aperture (e.g., F2.8), and field of view (e.g., 94°) are extracted. For infrared thermal imagers, parameters such as detector type (e.g., uncooled microbolometer), thermal sensitivity (e.g., 50mK), wavelength response range (e.g., 8-14μm), and spatial resolution (e.g., 640×512 pixels) are extracted.

[0078] Based on pre-defined sensor performance constraints, a mapping function relationship between physical characteristic parameters and capability characterization quantities is determined. The physical characteristic parameters are then standardized by dimensional unification to obtain standardized values. Mapping function relationship models are established for different types of sensors. For example, the detection range function of a visible light camera is related to the lens focal length, sensor size, and aperture size, and can be expressed as a function of the detection distance and these parameters. The resolution function is related to the sensor pixel density, optical system quality, and signal processing capability. During dimensional unification, different physical quantities are standardized to a standard range, such as standardizing the detection distance to the [0,1] range. Specifically, this is done by calculating (actual value - minimum value) / (maximum value - minimum value). A similar method can be used to standardize angular resolution, making the performance parameters of different types of sensors comparable.

[0079] Based on the mapping function relationship, standardized values ​​are substituted into the corresponding transformation rules to generate detection range function, resolution function, and data acquisition rate function as capability characterization quantities. The detection range function can be expressed as a functional relationship between the effective detection distance and target characteristics (such as size and reflectivity). For example, under standard atmospheric conditions, a certain type of optoelectronic pod has a maximum detection range of 8 kilometers for a 5m × 5m target. The resolution function describes the sensor's spatial resolution capability at different distances, such as the smallest target size that can be resolved at a distance of 5 kilometers being 0.5 meters. The data acquisition rate function reflects the time efficiency of the sensor in acquiring effective data, including factors such as sampling frequency and data processing latency. For example, a camera system can output a 1080P high-definition video stream at a rate of 60 frames per second.

[0080] Based on the positions of the effective detection range boundary and angular resolution boundary in the physical characteristic parameters within preset thresholds, various types of sensors are classified into different capability categories. Sensors can be categorized by detection range into short-range (0-2 km), medium-range (2-10 km), and long-range (over 10 km). They can also be categorized by angular resolution into low-resolution (>1 mrad), medium-resolution (0.1-1 mrad), and high-resolution (<0.1 mrad). Combining these two dimensions, nine basic capability categories can be formed, such as short-range-high-resolution, long-range-medium-resolution, etc. For composite sensors with multispectral detection capabilities, further spectral dimension classification can be added, such as combinations like visible light + near-infrared, mid-wave infrared + far-wave infrared, etc.

[0081] By associating and combining capability representations with capability categories, a set of sensor capability vectors is constructed. Each sensor's capability vector can be represented as a multi-dimensional vector, whose components include standardized detection range function values, resolution function values, data acquisition rate function values, and the encoding of its capability category. For example, the capability vector of a certain optoelectronic pod can be represented as [0.85, 0.72, 0.65, 2, 3], where the first three values ​​represent the standardized detection range, resolution, and data acquisition rate capability values, respectively, and the last two values ​​indicate that the sensor belongs to the medium-range to high-resolution category. For multiple sensors mounted on an UAV platform, the capability vectors of each sensor can be combined to form a capability vector set, serving as a quantitative representation of the overall platform's perception capability.

[0082] In practical applications, sensor capability vector sets can be used for UAV mission planning and resource allocation. For example, when performing target search missions, the most suitable sensor combination can be selected from the capability vector set based on mission requirements and environmental conditions. For targets requiring high-precision identification, high-resolution sensors are prioritized; while for large-area searches, sensors with a wider detection range are given priority. Through this intelligent decision-making based on capability quantification, the mission performance and resource utilization efficiency of UAV systems can be significantly improved.

[0083] In one optional implementation, the capability vectors in the sensor capability vector set are processed according to the geological hazard type characteristics to establish a quantitative adaptation relationship between the hazard feature dimension and the sensor capability dimension, resulting in an adaptation metric matrix, including:

[0084] Information on the spatial distribution morphology of geological hazards and information identifying the stages of hazard evolution are extracted from the characteristics of the geological hazard types.

[0085] The spatial distribution pattern information of the disaster is converted into spatial scale feature dimension and spatial complexity feature dimension, and the disaster evolution stage identification information is converted into temporal evolution rate feature dimension and stability feature dimension, thus obtaining a disaster feature dimension set;

[0086] Extract the detection range function, resolution function, and data acquisition rate function carried by each capability vector from the set of sensor capability vectors;

[0087] The detection range function is mapped to a spatial coverage capability dimension, the resolution function is mapped to a detail recognition capability dimension, and the data acquisition rate function is mapped to a temporal tracking capability dimension, thus obtaining a set of sensor capability dimensions;

[0088] A matching rule is determined between the disaster feature dimension set and the sensor capability dimension set. Based on the matching rule, the correlation strength between each feature dimension in the disaster feature dimension set and each capability dimension in the sensor capability dimension set is calculated to generate an adaptation metric matrix. Each element in the adaptation metric matrix represents the degree of dependence of a specific disaster feature dimension on a specific sensor capability dimension.

[0089] In the design of a geological disaster monitoring system, to achieve optimal adaptation between sensors and geological disaster characteristics, it is necessary to establish a quantitative adaptation relationship between disaster characteristics and sensor capabilities. This implementation method details the complete process from disaster characteristics to the generation of an adaptation metric matrix.

[0090] First, key information is extracted based on the characteristics of geological disaster types. Taking landslide disasters as an example, spatial distribution morphology information is extracted from historical monitoring data, including the geometric dimensions (length, width, and depth) of the landslide body and its shape complexity (simple slope, multi-step, or complex type). Simultaneously, information identifying the disaster evolution stage is extracted, such as deformation rate change curves, cumulative displacement, and acceleration characteristics, to indicate whether the landslide is in a potential deformation stage, a stable deformation stage, or an accelerated deformation stage.

[0091] Next, the extracted spatial distribution morphology information of the disaster is converted into quantitative feature dimensions. The spatial scale feature dimension represents the size of the disaster body with normalized values, such as 0.2 for small landslides (10^2-10^3 cubic meters), 0.5 for medium-sized landslides (10^4-10^6 cubic meters), and 0.8 for large landslides (above 10^6 cubic meters). The spatial complexity feature dimension is quantified by the geometric complexity index, such as 0.3 for a single sliding surface, 0.6 for a multi-level sliding surface, and 0.9 for a network of composite sliding surfaces.

[0092] Simultaneously, the disaster evolution stage identification information is converted into a time evolution rate feature dimension and a stability feature dimension. The time evolution rate is normalized using displacement rate, such as 0.9 for a slow type (1 meter / day). The stability feature dimension is based on the degree of fluctuation of the displacement time series, quantified using the coefficient of variation. A coefficient of variation less than 0.1 indicates high stability (value 0.1), a coefficient of variation between 0.1 and 0.5 indicates medium stability (value 0.5), and a coefficient of variation greater than 0.5 indicates low stability (value 0.9). Through the above transformation, a disaster feature dimension set containing four dimensions is formed.

[0093] In terms of sensor capability vector processing, key function parameters are extracted from the sensor capability vector set. Taking GPS receivers, tiltmeters, and InSAR sensors as examples, their detection range functions are extracted. For instance, the monitoring range of a GPS point is a single point (within a radius of 5 meters), the monitoring range of a tiltmeter is a local area (10-50 meters), and the monitoring range of InSAR is a wide area (several kilometers to tens of kilometers). Simultaneously, resolution functions are extracted. For example, the displacement resolution of GPS is at the millimeter level (1-5 millimeters), the angular resolution of a tiltmeter is 0.01 degrees, and the deformation resolution of InSAR is at the centimeter level (1-5 centimeters). Furthermore, data acquisition rate functions are extracted. For example, GPS can achieve sampling once per second, a tiltmeter can achieve sampling once per minute, and InSAR typically samples once per day to once per month.

[0094] The extracted sensor parameters are mapped to standardized capability dimensions. Spatial coverage capability is achieved by standardizing the detection range function; for example, point monitoring is mapped to 0.2, local area monitoring to 0.5, and wide-area monitoring to 0.9. Detail recognition capability is achieved through inverse standardization of the resolution function; for example, millimeter-level resolution is mapped to 0.9, centimeter-level to 0.5, and decimeter-level to 0.2. Temporal tracking capability is based on the data acquisition rate function; for example, real-time (second-level) is mapped to 0.9, near real-time (minute-level) to 0.6, and periodic (day-level) to 0.3. After mapping, a set of sensor capability dimensions is formed.

[0095] Matching rules were established between the disaster feature dimension set and the sensor capability dimension set. For the spatial scale feature dimension, the matching weight with the spatial coverage capability dimension was set to 0.8, and the matching weight with the detail recognition capability dimension was set to 0.4. For the spatial complexity feature dimension, the matching weight with the detail recognition capability dimension was set to 0.9, and the matching weight with the spatial coverage capability dimension was set to 0.3. For the temporal evolution rate feature dimension, the matching weight with the time series tracking capability dimension was set to 0.9, and the matching weight with the detail recognition capability dimension was set to 0.5. For the stability feature dimension, the matching weight with the time series tracking capability dimension was set to 0.7, and the matching weight with the detail recognition capability dimension was set to 0.4.

[0096] Based on the established matching rules, the correlation strength between the disaster feature dimension and the sensor capability dimension is calculated. A weighted correlation calculation method is used to calculate the correlation strength value for each pair of feature and capability dimensions. For example, for the spatial scale feature dimension (value 0.5) and the spatial coverage capability dimension (value 0.5), the correlation strength is calculated as 0.5 × 0.5 × 0.8 = 0.2. Through similar calculations, a complete fit metric matrix is ​​obtained, where each element reflects the degree of dependence of a specific disaster feature dimension on a specific sensor capability dimension.

[0097] Once the adaptation metric matrix is ​​generated, it can be used to guide sensor selection and deployment optimization. For example, for landslides that are large in scale (spatial scale eigenvalue 0.8) and structurally complex (spatial complexity eigenvalue 0.7), the adaptation metric matrix indicates that sensor combinations with strong spatial coverage and high detail recognition capabilities should be prioritized, such as InSAR combined with a local GPS network. Conversely, for landslides with fast evolution rates (temporal evolution rate eigenvalue 0.9) and low stability (stability eigenvalue 0.8), sensors with strong time-series tracking capabilities, such as high-frequency GPS monitoring systems, should be prioritized.

[0098] The above establishes a quantitative adaptation relationship between disaster characteristics and sensor capabilities, providing a scientific basis for the optimized design of geological disaster monitoring systems.

[0099] In one optional implementation, the spatial distribution morphology information of the disaster is converted into spatial scale feature dimension and spatial complexity feature dimension, and the disaster evolution stage identification information is converted into temporal evolution rate feature dimension and stability feature dimension, resulting in a disaster feature dimension set, including:

[0100] Based on the aforementioned disaster spatial distribution morphology information, mapping rules between geometric boundaries and spatial scale, as well as mapping rules between internal density distribution and spatial complexity, are determined;

[0101] According to the mapping rules, the spatial distribution pattern information of the disaster is converted into spatial scale feature dimension and spatial complexity feature dimension;

[0102] Extract the state transition nodes between different evolution stages and the duration data of each evolution stage from the disaster evolution stage identification information;

[0103] Evolution rate mutation points are calculated based on the time intervals of the state transition nodes, and stability imbalance intervals are identified based on the fluctuation amplitude of the duration data.

[0104] The correspondence between the distribution density of the evolution rate mutation points and the temporal evolution rate feature dimension, and the correspondence between the duration of the stability imbalance interval and the stability feature dimension are determined to obtain the association rules between evolutionary stages and temporal features.

[0105] According to the association rules, the disaster evolution stage identification information is converted into a time evolution rate feature dimension and a stability feature dimension, and the spatial scale feature dimension, the spatial complexity feature dimension, the time evolution rate feature dimension and the stability feature dimension are combined into a disaster feature dimension set.

[0106] First, based on the spatial distribution morphology information of disasters, mapping rules between geometric boundaries and spatial scale, as well as mapping rules between internal density distribution and spatial complexity, are determined. For earthquake disasters, geometric boundaries can be constructed using parameters such as focal depth, epicenter area, and affected area, and then quantified into spatial scale values. By analyzing factors such as the spatial variability of seismic wave energy attenuation rate, aftershock distribution density, and surface damage degree, a mapping relationship between internal density distribution and spatial complexity is established. For example, when the focal depth is 15 km and the radius of the affected area is 50 km, its spatial scale characteristic value is determined using the regional area calculation formula. Simultaneously, if the dispersion of magnitude distribution at monitoring points within the region is high, it indicates greater spatial complexity.

[0107] Based on the above mapping rules, the spatial distribution pattern information of disasters is converted into spatial scale feature dimension and spatial complexity feature dimension. The spatial scale feature dimension can be divided into five levels: local micro-scale (0-0.2), local small (0.2-0.4), regional medium (0.4-0.6), regional large (0.6-0.8), and cross-regional super-large (0.8-1.0). The spatial complexity feature dimension is also divided into five levels: homogeneous simple type (0-0.2), low complexity type (0.2-0.4), medium complexity type (0.4-0.6), high complexity type (0.6-0.8), and extremely complex chaotic type (0.8-1.0). Through normalization processing, the calculated original spatial parameters are mapped to the corresponding dimension intervals to form a standardized feature representation.

[0108] Next, the state transition nodes between different evolution stages and the duration of each stage are extracted from the disaster evolution stage identifier information. Taking flood disaster as an example, its evolution can be divided into three main stages: the water level rise period, the peak period, and the receding period. The transition points of each stage are identified through water level observation data. The specific method is as follows: the state transition nodes are determined by the abrupt change points of the water level change rate, the start and end times of each stage are recorded, and the duration of each stage is calculated. For example, during a river flood, monitoring data shows that the water level rise period lasts for 48 hours, the peak period lasts for 12 hours, and the receding period lasts for 72 hours, and the specific timestamps of each transition node are recorded.

[0109] Evolution rate abrupt changes are calculated based on the time intervals between state transition nodes. Stability imbalance intervals are identified based on the fluctuation amplitude of duration data. Evolution rate abrupt changes are determined by the rate of change of the time interval between adjacent transition nodes; when the rate of change exceeds a set threshold (e.g., 30%), it is marked as an abrupt change. Stability imbalance intervals are identified by analyzing historical statistical data on duration; when the duration of a certain period deviates from the historical average by a certain amount (e.g., 1.5 times the standard deviation), it is determined to be a stability imbalance interval.

[0110] The correspondence between the distribution density of evolution rate mutation points and the temporal evolution rate characteristic dimension, as well as the correspondence between the duration of stability imbalance intervals and the stability characteristic dimension, was determined. This yielded the association rules between evolutionary stages and temporal characteristics. The temporal evolution rate characteristic dimension was divided into five levels: slow evolution (0-0.2), low-speed evolution (0.2-0.4), medium-speed evolution (0.4-0.6), rapid evolution (0.6-0.8), and explosive evolution (0.8-1.0). The stability characteristic dimension was divided into five levels: highly stable (0-0.2), basically stable (0.2-0.4), fluctuating (0.4-0.6), unstable (0.6-0.8), and chaotic (0.8-1.0).

[0111] According to the association rules, the disaster evolution stage identification information is converted into the time evolution rate feature dimension and the stability feature dimension. For debris flow disasters, if the time interval between the cumulative rainfall trigger threshold and the formation of debris flow is less than 50% of the historical average, the evolution rate feature value is mapped to the rapid evolution range (0.6-0.8); if the duration of the debris flow movement stage fluctuates more than 40% of the historical record, the stability feature value is mapped to the unstable range (0.6-0.8).

[0112] Finally, the spatial scale feature dimension, spatial complexity feature dimension, temporal evolution rate feature dimension, and stability feature dimension are combined into a disaster feature dimension set, which is represented by a four-dimensional vector, where each component is a standardized value in the interval [0,1]. For example, the feature dimension set of a certain landslide disaster is shown to be a combination of local small-scale (spatial scale), high complexity (spatial complexity), rapid evolution (temporal evolution rate), and fluctuating (stability) characteristics.

[0113] By combining the above four feature dimensions, a unified characteristic representation of different types of disasters can be achieved, providing a scientific basis for subsequent disaster risk assessment, emergency response decision-making, and the formulation of disaster prevention and mitigation measures. This set of feature dimensions has strong adaptability; the calculation parameters and thresholds of each dimension can be adjusted according to the specific disaster type, enabling effective representation of various natural disasters such as earthquakes, floods, typhoons, and debris flows.

[0114] In one optional implementation, a sensor operating parameter configuration scheme is generated by co-solving the adaptation metric matrix and the environmental dynamic parameters through coupled multi-dimensional constraints, including:

[0115] Based on the adaptation metric values ​​corresponding to each disaster feature dimension in the adaptation metric matrix, the disaster monitoring requirement constraints are determined;

[0116] Based on the environmental factors affecting the sensor's operating state in the aforementioned environmental dynamic parameters, the allowable adjustment range of the sensor's operating parameters is determined, thus obtaining environmental adaptability constraints;

[0117] The disaster monitoring requirement constraint and the environmental adaptability constraint are combined to form a multi-dimensional constraint set;

[0118] For the multi-dimensional constraint set, a constraint coupling relationship expression is determined, and a mapping function between the sensor operating parameters and the multi-dimensional constraint set is determined through the constraint coupling relationship expression.

[0119] Within the feasible region defined by the mapping function, the adaptation metric values ​​in the adaptation metric matrix are optimized to determine the sensor operating parameter combination that enables the adaptation metric values ​​to reach the target state.

[0120] The sensor operating parameter combinations are converted into operating mode setting parameters and data acquisition control parameters for each sensor to obtain a sensor operating parameter configuration scheme.

[0121] In disaster monitoring systems, the appropriate configuration of sensor operating parameters is crucial for improving monitoring efficiency. Based on the adaptation metric matrix and dynamic environmental parameters, a sensor operating parameter configuration scheme suitable for the current monitoring environment can be generated through collaborative solution by coupling multi-dimensional constraints.

[0122] Figure 2 This is a schematic flowchart illustrating the sensor operating parameter configuration scheme according to an embodiment of the present invention. Figure 2 As shown, firstly, based on the adaptation metric values ​​corresponding to each disaster feature dimension in the adaptation metric matrix, the disaster monitoring requirement constraints are determined. The adaptation metric matrix includes the monitoring sensitivity requirements for different disaster features. For example, for floods, high accuracy in monitoring water level changes is required; for landslides, high accuracy in monitoring soil moisture and displacement is required. Specifically, threshold parameters corresponding to each disaster feature, such as minimum monitoring accuracy and minimum sampling frequency, are extracted from the adaptation metric matrix to form hard constraints on monitoring requirements. For example, if a certain area has a high risk of flooding, and the adaptation metric value for water level monitoring accuracy in the adaptation metric matrix is ​​0.85, this can be transformed into a constraint that the accuracy of the water level sensor is not less than ±2cm.

[0123] Next, based on the environmental factors affecting the sensor's operating state in the environmental dynamic parameters, the allowable adjustment range of the sensor's operating parameters is determined, resulting in environmental adaptability constraints. Environmental dynamic parameters include external conditions such as temperature, humidity, and electromagnetic interference, which affect the sensor's operating state and reliability. By analyzing the sensitivity of each sensor to environmental factors, a correspondence between sensor operating parameters and environmental factors is established. For example, the influence curve of temperature on sensor measurement accuracy is used, and based on this, a safe boundary for parameter adjustment is determined. For instance, when the ambient temperature is between -10℃ and 40℃, the sampling frequency of a displacement sensor is allowed to be adjusted between 2-10Hz, while outside this temperature range, the sampling frequency needs to be limited to between 2-5Hz to ensure data reliability.

[0124] Subsequently, the constraints of disaster monitoring requirements and environmental adaptability are combined to form a multi-dimensional constraint set. This process requires identifying the interaction between the two types of constraints and analyzing the compatibility of the constraints. In practice, each constraint can be represented as a constraint region in the sensor parameter space, and the feasible region that satisfies all constraints can be obtained by finding the intersection. For example, if a meteorological station simultaneously monitors heavy rain and flash floods, the monitoring requirement constraint requires that the sampling interval of the rainfall sensor not exceed 5 minutes, while the environmental constraint indicates that the sampling interval should not be less than 3 minutes under strong electromagnetic interference, then the final constraint range is 3-5 minutes.

[0125] For a multi-dimensional constraint set, a constraint coupling relationship expression is determined. This expression then establishes a mapping function between the sensor's operating parameters and the multi-dimensional constraint set. The constraint coupling relationship expression describes the interaction and constraint relationships between various constraints, taking into account the importance and priority of different constraints. For example, for a hydrological monitoring sensor, a coupling relationship can be established between sampling frequency (f), accuracy (a), power consumption (p), and ambient temperature (t): when the temperature t decreases, to ensure that the accuracy a remains constant, the power consumption p needs to be increased or the sampling frequency f needs to be decreased. Based on these relationships, a mapping function from sensor parameters to constraint satisfaction is constructed. This function maps the sensor's operating parameter space to an evaluation space of constraint satisfaction.

[0126] Within the feasible region defined by the mapping function, the adaptation metric values ​​in the adaptation metric matrix are optimized to determine the sensor operating parameter combination that brings the adaptation metric values ​​to the target state. Optimization calculations can employ multi-objective optimization methods, such as genetic algorithms and particle swarm optimization, to maximize overall monitoring efficiency. Optimization objectives include maximizing monitoring coverage, minimizing energy consumption, and maximizing data reliability. For example, in flash flood monitoring scenarios, the overall monitoring capability of the sensor network can be defined as the weighted sum of the adaptation metrics of each node. By adjusting the operating parameter combination of each sensor, this weighted sum is maximized while ensuring that energy consumption remains within acceptable limits.

[0127] Finally, the sensor operating parameters are combined and converted into operating mode setting parameters and data acquisition control parameters for each sensor, resulting in a sensor operating parameter configuration scheme. This conversion process must consider the specific interface specifications and control protocols of each sensor, mapping the optimized parameter values ​​to the instruction set recognizable by the sensor. Operating mode setting parameters include the sensor's operating state (e.g., normal, sleep, low power consumption) and signal processing mode; data acquisition control parameters include sampling frequency, sampling accuracy, and data buffer size. For example, for a certain type of displacement sensor, a parameter configuration scheme containing "operating mode = normal monitoring, sampling frequency = 5Hz, accuracy level = high, data upload cycle = 30 seconds" can be generated and converted into corresponding communication commands sent to the sensor.

[0128] Through the above, the operating parameters of the sensor network can be adjusted in real time according to the dynamic changes in current disaster risks and environmental conditions, thereby realizing the intelligence and adaptability of the disaster monitoring system, improving the effectiveness and reliability of monitoring data, and providing more accurate data support for disaster early warning and emergency decision-making.

[0129] In one optional implementation, a constraint coupling relationship expression is determined for the multi-dimensional constraint set, and a mapping function between the sensor operating parameters and the multi-dimensional constraint set is determined through the constraint coupling relationship expression, including:

[0130] The requirement boundaries for sensor operating parameters are determined based on the aforementioned disaster monitoring requirements, and the limitation boundaries for sensor operating parameters are determined based on the aforementioned environmental adaptability constraints.

[0131] Based on the demand boundary and the constraint boundary, identify the constraint conflict area where the range of sensor operating parameters required by the disaster monitoring demand constraints and the range of sensor operating parameters allowed by the environmental adaptability constraints have insufficient overlap;

[0132] For the constraint conflict area, calculate the degree of deviation between the demand boundary and the restriction boundary, and determine the priority coefficients of the disaster monitoring demand constraint and the environmental adaptability constraint based on the degree of deviation;

[0133] Based on the priority coefficients, a weighted combination of the disaster monitoring requirement constraints and the environmental adaptability constraints is performed to determine the constraint coupling relationship expression;

[0134] The constraint coupling relationship expression is converted into the boundary condition equation of the sensor operating parameters. The feasible domain boundary of the sensor operating parameters is obtained by solving the boundary condition equation, and the mapping function between the sensor operating parameters and the multi-dimensional constraint set is determined.

[0135] In a specific implementation, the requirement boundaries of the sensor's operating parameters are first determined based on the constraints of disaster monitoring needs. For example, in a forest fire monitoring scenario, the monitoring requirements stipulate a temperature detection range of -10℃ to 85℃, a data sampling frequency of no less than 5 times / minute, and a signal transmission distance of at least 500 meters. These requirements constitute the requirement boundaries of the sensor's operating parameters, represented as the parameter vector Pn=(Tn_min, Tn_max, Fn_min, Dn_min)=(-10℃, 85℃, 5 times / minute, 500 meters).

[0136] Simultaneously, the limiting boundaries of the sensor's operating parameters are determined based on environmental adaptability constraints. In a forest environment, these constraints include a temperature limit of -20℃ to 60℃, a maximum power supply of 2 watts, and a battery life requirement of at least 6 months. These limitations constitute the limiting boundaries of the sensor's operating parameters, represented as the parameter vector Pe = (Te_min, Te_max, Pe_max, Be_min) = (-20℃, 60℃, 2 watts, 6 months).

[0137] Based on the demand boundary and the constraint boundary, areas of constraint conflict can be identified. Specifically, there is a conflict in the upper limit of the temperature detection range, with the demand boundary requiring 85°C while the environmental constraint boundary only allows 60°C. Furthermore, high-frequency sampling leads to increased power consumption, which in turn affects battery life, creating a potential conflict point.

[0138] For the constraint conflict region, the deviation between the demand boundary and the limit boundary is calculated. The deviation of the upper temperature limit can be expressed as (Tn_max-Te_max) / Te_max=(85-60) / 60=41.7%. Regarding the relationship between sampling frequency and power consumption, a power consumption model is established, assuming a linear relationship between sampling frequency F and power consumption P: P=k×F+b, where k and b are constants. Experiments determine k=0.3 watts / (sampling / minute) and b=0.5 watts. When the sampling frequency is 5 times / minute, the power consumption is 0.3×5+0.5=2 watts, which just reaches the limit boundary, and the deviation is 0%.

[0139] Priority coefficients are determined based on the degree of deviation, and a priority evaluation function is set. When the deviation is small, demand constraints are satisfied first, while environmental constraints are considered first when the deviation is large. For the temperature parameter, since the deviation exceeds 40%, environmental constraints are given higher weight, and priority coefficients αT=0.3 (demand constraint) and βT=0.7 (environmental constraint) are set. For the sampling frequency parameter, with a deviation of 0%, priority coefficients αF=0.6 and βF=0.4 are set.

[0140] Based on priority coefficients, a weighted combination of disaster monitoring demand constraints and environmental adaptability constraints is used to determine the constraint coupling relationship expression. The coupling expression for the temperature parameter is T_min=max(αT×Tn_min, βT×Te_min)=max(0.3×(-10), 0.7×(-20))=max(-3, -14)=-3℃, T_max=min(αT×Tn_max, βT×Te_max)=min(0.3×85, 0.7×60)=min(25.5, 42)=25.5℃. The coupling expression for the sampling frequency is F_min=αF×Fn_min=0.6×5=3 times / minute, P_max=βF×Pe_max=0.4×2=0.8 watts.

[0141] The constraint coupling relationship expression is converted into boundary condition equations for the sensor's operating parameters. The temperature parameter boundary condition is -3℃≤T≤25.5℃, and the sampling frequency and power consumption relationship is 0.3×F+0.5≤0.8W, which yields F≤1 time / minute. However, since the minimum sampling requirement is 3 times / minute, the conflict needs to be resolved by improving energy efficiency or increasing battery capacity.

[0142] Finally, the feasible domain boundary of the sensor's operating parameters was determined, and a mapping function between the sensor's operating parameters and the multi-dimensional constraint set was established. The temperature detection range mapping function is T(x) = -3 + (25.5 + 3) × x, x ∈ [0, 1]. The normalized parameter x is mapped to the actual temperature range. The sampling frequency is set to a fixed value of 3 times / minute, and the power consumption design is optimized so that P = 0.3 × 3 + b', where b' < 0.2 watts.

[0143] Through the above, a mapping relationship between sensor operating parameters and a multi-dimensional set of constraints was established, providing a scientific basis for sensor parameter configuration in disaster monitoring systems and realizing optimal parameter selection under constraints.

[0144] In one optional implementation, the operating status of each sensor of the UAV is controlled based on the sensor operating parameter configuration scheme, driving the UAV to perform monitoring tasks and collect geological disaster detection data along a predetermined route, including:

[0145] The operating mode setting parameters and data acquisition control parameters of each sensor are extracted from the sensor operating parameter configuration scheme. The operating mode setting parameters are converted into sensor start commands and sensor operating state switching commands, and the data acquisition control parameters are converted into data acquisition frequency control commands and data acquisition accuracy control commands.

[0146] The system sends corresponding sensor start-up commands, sensor working state switching commands, data acquisition frequency control commands, and data acquisition accuracy control commands to each sensor mounted on the UAV, so that each sensor enters the working state according to the sensor working parameter configuration scheme.

[0147] The system acquires waypoint location information and flight segment time information for a predetermined flight route. It then associates the waypoint location information and flight segment time information with the data acquisition frequency control commands of each sensor to determine the synchronization correspondence between the flight route advancement nodes and the data acquisition trigger nodes. This synchronization correspondence is used to characterize the timing rules that trigger the corresponding sensor to perform data acquisition actions when the UAV reaches a specific waypoint location.

[0148] The drone is driven to fly along the predetermined route. When the drone reaches the route advancement node defined in the synchronization correspondence, the corresponding data acquisition trigger node is triggered, and the corresponding sensor is controlled to collect the geological disaster detection data.

[0149] The system extracts the operating mode setting parameters and data acquisition control parameters for each sensor from the sensor operating parameter configuration scheme. The operating mode setting parameters include sensor on / off status, operating frequency range, and power level settings; the data acquisition control parameters include sampling rate, resolution, and sensitivity threshold. For example, for the onboard hyperspectral camera, the operating mode setting parameters include spectral range selection (e.g., visible light, near-infrared, or mid-infrared) and imaging mode (continuous imaging or triggered imaging); the data acquisition control parameters include imaging resolution (e.g., 1280×720 or 1920×1080) and acquisition frequency (e.g., 2 frames per second or 5 frames per second).

[0150] The extracted operating mode settings are converted into specific sensor start-up commands and sensor operating state switching commands. Taking the radar system as an example, when the operating mode is set to "high-precision surface detection mode," the converted start-up commands include setting the radar transmit power to 200 watts, the operating frequency to 10 GHz, and the scanning angle range to ±30 degrees. The operating state switching commands include a power management command sequence to switch from standby mode to full-power operating mode. Simultaneously, data acquisition control parameters are converted into data acquisition frequency control commands and data acquisition accuracy control commands. For example, for a thermal imager, setting the data acquisition frequency to sample once every 3 seconds is converted into a corresponding timed trigger command; setting the accuracy control to a temperature resolution of 0.1℃ is converted into a sensor sensitivity adjustment command.

[0151] The system sends corresponding control commands to each sensor onboard the UAV. These commands, generated via the UAV's internal communication bus (such as a CAN bus or serial communication interface), include sensor start-up commands, operating state switching commands, data acquisition frequency control commands, and data acquisition accuracy control commands, which are then sent to the corresponding sensor's control unit. A specific communication protocol is used during transmission to ensure correct delivery and execution of the commands. For example, when sending an operating command to the onboard multispectral camera, an initialization sequence is sent first, followed by parameter configuration commands (such as setting the band selection to four bands: red, green, blue, and near-infrared, with a sampling interval of 5 seconds), and finally, a trigger signal to start operation.

[0152] Waypoint location information typically includes latitude and longitude coordinates, altitude, and arrival time; flight segment time information includes dynamic parameters such as flight time and speed changes between adjacent waypoints. For example, in a landslide hazard monitoring mission, the planned flight route includes ten key waypoints. Each waypoint is recorded with coordinates such as "N30°25'47.8, E114°12'36.5, altitude 320 meters" and the estimated arrival time at that waypoint is "300 seconds after mission start".

[0153] The waypoint location information and flight segment time information are correlated with the data acquisition frequency control commands of each sensor to determine the synchronization correspondence between the flight path advancement nodes and the data acquisition trigger nodes. This synchronization correspondence is stored in the form of a spatiotemporal mapping table, which clearly defines which sensors should be triggered to perform data acquisition operations when the UAV reaches a specific location coordinate or time point. For example, when the UAV reaches the top of the dangerous rock mass (waypoint P3), it is set to trigger the high-resolution camera to continuously acquire image data for 30 seconds at the highest sampling rate (10 frames per second), while simultaneously activating the lidar to perform a high-precision 3D scan; when flying to the debris flow gully area (the flight segment from waypoints P4 to P5), the thermal imager and multispectral camera are alternately activated at a frequency of once every 5 seconds to acquire data.

[0154] The drone is propelled along a predetermined flight path. Upon reaching a path advancement node defined in the synchronization correspondence, a corresponding data acquisition action is triggered. The drone's flight control system continuously compares the current position with the path advancement node. When it determines that a specific path advancement node has been reached (position error within a preset threshold, such as a horizontal error of less than 2 meters and a vertical error of less than 1 meter), it immediately sends a data acquisition trigger signal to the corresponding sensor. For example, when the drone flies to a position 500 meters above the landslide crown, the synthetic aperture radar is simultaneously triggered to enter high-resolution operating mode and begin acquiring surface deformation data; at the same time, the gravity anomaly detector is activated to acquire underground density distribution data to assess the internal structural characteristics of the landslide.

[0155] In practical applications, the configuration of sensor operating parameters can be tailored to the characteristics of different geological hazard types (such as landslides, debris flows, and ground subsidence). For example, for monitoring active landslides, high-precision lidar and synthetic aperture radar are prioritized to increase sampling density; while for monitoring debris flow hazard areas, the data acquisition frequency of multispectral imaging and thermal infrared sensors is emphasized to capture abnormal changes in surface moisture and temperature. The collected geological hazard detection data is saved in real time to the storage system carried by the UAV and selectively transmitted back to the ground monitoring station via data link for subsequent disaster risk assessment and early warning.

[0156] The traffic geological disaster drone sensor adaptive configuration system of this invention includes:

[0157] The first unit is used to acquire the geological hazard type characteristics and environmental dynamic parameters of the target monitoring area;

[0158] The second unit is used to perform capability quantification and characterization of multiple types of sensors carried by the UAV based on preset sensor performance constraint rules and detection capability mapping relationship, and obtain a set of sensor capability vectors;

[0159] The third unit is used to process each capability vector in the sensor capability vector set according to the geological disaster type characteristics, establish a quantitative adaptation relationship between the disaster feature dimension and the sensor capability dimension, and obtain an adaptation metric matrix;

[0160] The fourth unit is used to generate a sensor operating parameter configuration scheme by coordinating the solution through coupling multi-dimensional constraints based on the adaptation metric matrix and the environmental dynamic parameters.

[0161] The fifth unit is used to control the working status of each sensor of the UAV based on the sensor working parameter configuration scheme, drive the UAV to perform monitoring tasks along a predetermined route and collect geological disaster detection data, and the geological disaster detection data is used to optimize and adjust the weight coefficients of the corresponding disaster feature dimensions in the adaptation metric matrix.

[0162] A third aspect of the present invention provides an electronic device, comprising:

[0163] processor;

[0164] Memory used to store processor-executable instructions;

[0165] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0166] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0167] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0168] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An adaptive configuration method for UAV sensors for traffic and geological disaster prevention, characterized in that, include: Obtain the geological hazard type characteristics and environmental dynamic parameters of the target monitoring area; Based on the preset sensor performance constraint rules and the mapping relationship between detection capabilities, the capabilities of various types of sensors carried by the UAV are quantitatively characterized to obtain a set of sensor capability vectors; Based on the geological hazard type characteristics, each capability vector in the sensor capability vector set is processed to establish a quantitative adaptation relationship between the hazard feature dimension and the sensor capability dimension, resulting in an adaptation metric matrix, including: From the geological hazard type characteristics, spatial distribution morphology information and hazard evolution stage identification information are extracted. The spatial distribution morphology information is converted into spatial scale feature dimension and spatial complexity feature dimension, and the hazard evolution stage identification information is converted into temporal evolution rate feature dimension and stability feature dimension, thus obtaining a set of hazard feature dimensions. Extract the detection range function, resolution function, and data acquisition rate function carried by each capability vector from the set of sensor capability vectors; The detection range function is mapped to a spatial coverage capability dimension, the resolution function is mapped to a detail recognition capability dimension, and the data acquisition rate function is mapped to a temporal tracking capability dimension, thus obtaining a set of sensor capability dimensions; A matching rule is determined between the disaster feature dimension set and the sensor capability dimension set. Based on the matching rule, the correlation strength between each feature dimension in the disaster feature dimension set and each capability dimension in the sensor capability dimension set is calculated to generate an adaptation metric matrix. Each element in the adaptation metric matrix represents the degree of dependence of a specific disaster feature dimension on a specific sensor capability dimension. Based on the adaptation metric matrix and the environmental dynamic parameters, a sensor operating parameter configuration scheme is generated through collaborative solution by coupling multi-dimensional constraints, including: Based on the adaptation metric values ​​corresponding to each disaster feature dimension in the adaptation metric matrix, the disaster monitoring requirement constraints are determined; Based on the environmental factors affecting the sensor's operating state in the aforementioned environmental dynamic parameters, the allowable adjustment range of the sensor's operating parameters is determined, thus obtaining environmental adaptability constraints; The disaster monitoring requirement constraint and the environmental adaptability constraint are combined to form a multi-dimensional constraint set; For the multi-dimensional constraint set, a constraint coupling relationship expression is determined, and a mapping function between the sensor operating parameters and the multi-dimensional constraint set is determined through the constraint coupling relationship expression. Within the feasible region defined by the mapping function, the adaptation metric values ​​in the adaptation metric matrix are optimized to determine the sensor operating parameter combination that enables the adaptation metric values ​​to reach the target state. The sensor operating parameter combination is converted into operating mode setting parameters and data acquisition control parameters for each sensor to obtain a sensor operating parameter configuration scheme. Based on the sensor operating parameter configuration scheme, the working status of each sensor of the UAV is controlled, and the UAV is driven to perform monitoring tasks along a predetermined route and collect geological disaster detection data. The geological disaster detection data is used to optimize and adjust the weight coefficients of the corresponding disaster feature dimensions in the adaptation metric matrix.

2. The method according to claim 1, characterized in that, Based on the preset sensor performance constraint rules and the mapping relationship between detection capabilities, the capabilities of various types of sensors carried by the UAV are quantitatively characterized, resulting in a set of sensor capability vectors, including: Extracting the physical characteristic parameters of each sensor from the various types of sensors carried by the drone; Based on preset sensor performance constraint rules, the mapping function relationship between physical characteristic parameters and capability characterization quantities is determined, and the physical characteristic parameters are processed to obtain standardized values. Based on the mapping function relationship, the standardized values ​​are substituted into the corresponding transformation rules to generate the detection range function, resolution function and data acquisition rate function as capability characterization quantities. Based on the positions of the effective detection distance boundary and angular resolution boundary in the physical characteristic parameters within a preset threshold, multiple types of sensors are classified into different capability categories; The capability representation quantity is associated and combined with the capability category to construct a sensor capability vector set.

3. The method according to claim 1, characterized in that, The spatial distribution morphology information of the disaster is converted into spatial scale feature dimension and spatial complexity feature dimension, and the disaster evolution stage identification information is converted into temporal evolution rate feature dimension and stability feature dimension, resulting in a disaster feature dimension set, including: Based on the aforementioned disaster spatial distribution morphology information, mapping rules between geometric boundaries and spatial scale, as well as mapping rules between internal density distribution and spatial complexity, are determined; According to the mapping rules, the spatial distribution pattern information of the disaster is converted into spatial scale feature dimension and spatial complexity feature dimension; Extract the state transition nodes between different evolution stages and the duration data of each evolution stage from the disaster evolution stage identification information; Evolution rate mutation points are calculated based on the time intervals of the state transition nodes, and stability imbalance intervals are identified based on the fluctuation amplitude of the duration data. The correspondence between the distribution density of the evolution rate mutation points and the temporal evolution rate feature dimension, and the correspondence between the duration of the stability imbalance interval and the stability feature dimension are determined to obtain the association rules between evolutionary stages and temporal features. According to the association rules, the disaster evolution stage identification information is converted into a time evolution rate feature dimension and a stability feature dimension, and the spatial scale feature dimension, the spatial complexity feature dimension, the time evolution rate feature dimension and the stability feature dimension are combined into a disaster feature dimension set.

4. The method according to claim 1, characterized in that, For the multi-dimensional constraint set, determine the constraint coupling relationship expression, and use the constraint coupling relationship expression to determine the mapping function between the sensor operating parameters and the multi-dimensional constraint set, including: The requirement boundaries for sensor operating parameters are determined based on the aforementioned disaster monitoring requirements, and the limitation boundaries for sensor operating parameters are determined based on the aforementioned environmental adaptability constraints. Based on the demand boundary and the constraint boundary, identify the constraint conflict area where the range of sensor operating parameters required by the disaster monitoring demand constraints and the range of sensor operating parameters allowed by the environmental adaptability constraints have insufficient overlap; For the constraint conflict area, calculate the degree of deviation between the demand boundary and the restriction boundary, and determine the priority coefficients of the disaster monitoring demand constraint and the environmental adaptability constraint based on the degree of deviation; Based on the priority coefficients, a weighted combination of the disaster monitoring requirement constraints and the environmental adaptability constraints is performed to determine the constraint coupling relationship expression; The constraint coupling relationship expression is converted into the boundary condition equation of the sensor operating parameters. The feasible domain boundary of the sensor operating parameters is obtained by solving the boundary condition equation, and the mapping function between the sensor operating parameters and the multi-dimensional constraint set is determined.

5. The method according to claim 1, characterized in that, Based on the sensor operating parameter configuration scheme, the operating status of each sensor of the UAV is controlled, and the UAV is driven to perform monitoring tasks and collect geological disaster detection data along a predetermined route, including: The operating mode setting parameters and data acquisition control parameters of each sensor are extracted from the sensor operating parameter configuration scheme. The operating mode setting parameters are converted into sensor start commands and sensor operating state switching commands, and the data acquisition control parameters are converted into data acquisition frequency control commands and data acquisition accuracy control commands. The system sends corresponding sensor start-up commands, sensor working state switching commands, data acquisition frequency control commands, and data acquisition accuracy control commands to each sensor mounted on the UAV, so that each sensor enters the working state according to the sensor working parameter configuration scheme. The system acquires waypoint location information and flight segment time information for a predetermined flight route. It then associates the waypoint location information and flight segment time information with the data acquisition frequency control commands of each sensor to determine the synchronization correspondence between the flight route advancement nodes and the data acquisition trigger nodes. This synchronization correspondence is used to characterize the timing rules that trigger the corresponding sensor to perform data acquisition actions when the UAV reaches a specific waypoint location. The drone is driven to fly along the predetermined route. When the drone reaches the route advancement node defined in the synchronization correspondence, the corresponding data acquisition trigger node is triggered, and the corresponding sensor is controlled to collect the geological disaster detection data.

6. An adaptive configuration system for unmanned aerial vehicle (UAV) sensors for traffic and geological disaster prevention, used to implement the method as described in any one of claims 1-5, characterized in that, include: The first unit is used to acquire the geological hazard type characteristics and environmental dynamic parameters of the target monitoring area; The second unit is used to perform capability quantification and characterization of multiple types of sensors carried by the UAV based on preset sensor performance constraint rules and detection capability mapping relationship, and obtain a set of sensor capability vectors; The third unit is used to process each capability vector in the sensor capability vector set according to the geological disaster type characteristics, establish a quantitative adaptation relationship between the disaster feature dimension and the sensor capability dimension, and obtain an adaptation metric matrix; The fourth unit is used to generate a sensor operating parameter configuration scheme by coordinating the solution through coupling multi-dimensional constraints based on the adaptation metric matrix and the environmental dynamic parameters. The fifth unit is used to control the working status of each sensor of the UAV based on the sensor working parameter configuration scheme, drive the UAV to perform monitoring tasks along a predetermined route and collect geological disaster detection data, and the geological disaster detection data is used to optimize and adjust the weight coefficients of the corresponding disaster feature dimensions in the adaptation metric matrix.

7. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 5.

Images