Slope profile measurement method and system based on amphibious inspection robot

Abstract

The present invention provides a slope profile measurement method and system based on an amphibious inspection robot, which relates to the field of slope profile measurement, wherein the method comprises: acquiring and processing deformation influence information, initial profile information and profile measurement information of multiple sample slopes; based on the deformation influence information and initial profile information of the current slope, retrieving profile measurement information of multiple target sample slopes for determining multiple profile measurement positions of the current slope; arranging multiple real-time monitoring devices at the profile measurement positions; determining a slope profile inspection scheme of the amphibious inspection robot according to real-time monitoring data collected by the multiple real-time monitoring devices; determining current profile measurement information of the current slope according to slope profile data collected by the amphibious inspection robot based on the slope profile inspection scheme; determining deformation information of the current slope according to the initial profile information and current profile measurement information of the current slope, which has the advantages of improving the efficiency and accuracy of slope profile measurement.

Description

Technical Field

[0001] The present invention relates to the field of slope profile measurement, and in particular to a slope profile measurement method and system based on an amphibious inspection robot. Background Art

[0002] A slope is a slope with a certain slope in order to stabilize facilities such as highway foundations or foundation pits. With the use and construction of various types of infrastructure, the use scenarios of various types of slopes are gradually increasing, such as artificial slope areas built on the side of highways or natural slope areas on the side of highways, protective slopes on the side of water conservancy projects and coal mining projects, and foundation pit slopes of various construction projects that require foundation pit excavation. While slopes ensure safety for people's production and life, they also contain certain risks. Slopes will deform due to various factors. When the quantitative change of deformation causes qualitative change, that is, when slope deformation turns into slope damage, it will cause serious threats to the safety of people's lives and property, including landslides, collapses, dumping, etc. Therefore, it is necessary to pay attention to monitoring the deformation of slopes for disaster warning.

[0003] Slope profile measurement aims to obtain detailed geometric data of slopes through precise measurement technology, so as to provide a reliable basis for slope stability analysis, support design and monitoring. In the existing technology, slope profile data is mainly collected on-site by staff setting up multiple instruments and auxiliary equipment, and then the deformation of the slope profile is manually calculated. However, due to the complex geological conditions of mountain road slopes, large slope areas, and large area occupied by instruments and auxiliary equipment, it is difficult for staff to collect slope data on-site and the operation is complicated.

[0004] Therefore, it is necessary to provide a slope profile measurement method and system based on an amphibious inspection robot to improve the efficiency and accuracy of slope profile measurement. Summary of the invention

[0005] The present invention provides a slope profile measurement system based on an amphibious inspection robot, comprising: a data management module, used to obtain and process deformation influence information, initial profile information and profile measurement information of multiple sample slopes, wherein the profile measurement information includes profile deformation monitoring data of multiple positions; a data retrieval module, used to obtain the deformation influence information and initial profile information of the current slope, and based on the deformation influence information and initial profile information of the current slope, retrieve the profile measurement information of multiple target sample slopes; a real-time monitoring module, used to monitor the profile measurement information of the multiple target sample slopes based on the profile measurement information of the multiple target sample slopes; , determine multiple contour measurement positions of the current slope, and set multiple real-time monitoring devices at the contour measurement positions; an inspection scheduling module, used to determine the slope contour inspection plan of the amphibious inspection robot according to the real-time monitoring data collected by the multiple real-time monitoring devices; a contour measurement module, used to determine the current contour measurement information of the current slope according to the slope contour data collected by the amphibious inspection robot based on the slope contour inspection plan; a deformation monitoring module, used to determine the deformation information of the current slope according to the initial contour information and current contour measurement information of the current slope.

[0006] Further, the data management module processes deformation influence information, initial contour information and contour measurement information of multiple sample slopes, including: calculating the first information similarity of the deformation influence information of any two sample slopes; calculating the second information similarity of the initial contour information of any two sample slopes; calculating the third information similarity of the contour measurement information of any two sample slopes; clustering the multiple sample slopes according to the first information similarity of the deformation influence information of any two sample slopes through a clustering algorithm to determine multiple first clusters; for each of the first clusters, clustering the multiple sample slopes included in the first cluster according to the second information similarity of the initial contour information of any two sample slopes included in the first cluster to determine multiple second clusters included in the first cluster; for each of the second clusters, clustering the multiple sample slopes included in the second cluster according to the third information similarity of the contour measurement information of any two sample slopes included in the second cluster to determine multiple third clusters included in the second cluster.

[0007] Further, the data retrieval module retrieves contour measurement information of multiple target sample slopes based on the deformation influence information and initial contour information of the current slope, including: calculating the fourth information similarity between the deformation influence information of the current slope and the deformation influence information of the cluster center of each of the first clusters; determining a target first cluster from the multiple first clusters based on the fourth information similarity between the deformation influence information of the current slope and the deformation influence information of the cluster center of each of the first clusters; calculating the fifth information similarity between the initial contour information of the current slope and the initial contour information of the cluster center of each of the second clusters included in the target first cluster; determining a target second cluster from the multiple second clusters based on the fifth information similarity between the initial contour information of the current slope and the initial contour information of the cluster center of each of the second clusters included in the target first cluster, and retrieving the contour measurement information of multiple sample slopes included in the target second cluster as the contour measurement information of the multiple target sample slopes.

[0008] Furthermore, the real-time monitoring module determines multiple contour measurement positions of the current slope based on the contour measurement information of the multiple target sample slopes, including: for each of the target sample slopes, determining the contour deformation risk position of the target sample slope based on the contour measurement information of the target sample slope; and determining multiple contour measurement positions of the current slope according to the contour deformation risk position of each of the target sample slopes and the initial contour information of the current slope through a position determination model.

[0009] Furthermore, the inspection scheduling module determines the slope profile inspection plan of the amphibious inspection robot based on the real-time monitoring data collected by the multiple real-time monitoring devices, including: determining multiple slope profile inspection positions based on the real-time monitoring data collected by the multiple real-time monitoring devices; acquiring real-time environmental information based on the multiple slope profile inspection positions; and determining the slope profile inspection plan of the amphibious inspection robot based on the multiple slope profile inspection positions and the real-time environmental information through a plan generation model.

[0010] Furthermore, the inspection scheduling module determines a plurality of slope profile inspection positions based on the real-time monitoring data collected by the plurality of real-time monitoring devices, including: determining a position correlation coefficient between any two of the profile measurement positions; determining the plurality of slope profile inspection positions based on the real-time monitoring data collected by the plurality of real-time monitoring devices and the position correlation coefficient between any two of the profile measurement positions.

[0011] Furthermore, the inspection scheduling module determines the slope profile inspection scheme of the amphibious inspection robot according to the multiple slope profile inspection positions and the real-time environmental information through a scheme generation model, including: S11, establishing a scheme evaluation system; S12, generating multiple candidate slope profile inspection schemes according to the multiple slope profile inspection positions through a scheme generation unit of the scheme generation model; S13, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S14, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S15, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S16, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S17, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S18, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S19, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S20, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S21, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S22, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S23, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S24, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S25, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S26, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and the real-time environmental information; S27, calculating the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system Scheme score of slope profile inspection scheme, screening the multiple candidate slope profile inspection schemes to obtain multiple target candidate slope profile inspection schemes; S15, repeatedly executing S12 to S14 until the number of the target candidate slope profile inspection schemes is greater than the preset number threshold; S16, calculating the global similarity standard deviation of the multiple target candidate slope profile inspection schemes; S17, judging whether the global similarity standard deviation of the multiple target candidate slope profile inspection schemes is greater than the preset global similarity standard deviation threshold, if so, executing S18, if not, executing S12;

[0012] S18. Generate a slope profile inspection plan for the amphibious inspection robot based on multiple target candidate slope profile inspection plans and the plan score of each target candidate slope profile inspection plan through the plan optimization unit of the plan generation model.

[0013] Furthermore, the real-time monitoring device includes at least a contour marker; the contour metrology module determines the current contour metrology information of the current slope according to the slope contour data collected by the amphibious inspection robot based on the slope contour inspection scheme, including: determining the real-time relative position parameters between multiple contour markers according to the slope image collected by the amphibious inspection robot based on the slope contour inspection scheme and the position correlation coefficient of any two of the contour metrology positions.

[0014] Furthermore, the deformation monitoring module determines the deformation information of the current slope based on the initial contour information and current contour measurement information of the current slope, including: determining the initial relative position parameters between multiple contour markers based on the initial contour information of the current slope and the position correlation coefficient of any two of the contour measurement positions; determining the deformation information of the current slope based on the initial relative position parameters between the multiple contour markers and the real-time relative position parameters between the multiple contour markers.

[0015] The present invention provides a slope profile measurement method based on an amphibious inspection robot, which is applied to the above-mentioned slope profile measurement system based on an amphibious inspection robot, comprising: acquiring and processing deformation influence information, initial profile information and profile measurement information of multiple sample slopes, wherein the profile measurement information includes profile deformation monitoring data of multiple positions; based on the deformation influence information and initial profile information of the current slope, retrieving profile measurement information of multiple target sample slopes; based on the profile measurement information of the multiple target sample slopes, determining multiple profile measurement positions of the current slope; arranging multiple real-time monitoring devices at the profile measurement positions; determining a slope profile inspection scheme of the amphibious inspection robot according to the real-time monitoring data collected by the multiple real-time monitoring devices; determining current profile measurement information of the current slope according to the slope profile data collected by the amphibious inspection robot based on the slope profile inspection scheme; determining deformation information of the current slope according to the initial profile information and current profile measurement information of the current slope.

[0016] Compared with the prior art, the slope profile measurement method and system based on the amphibious inspection robot provided by the present invention have at least the following beneficial effects:

[0017] 1. The data management module can efficiently acquire, process and store the deformation influence information, initial contour information and contour measurement information of multiple sample slopes. This ensures the integrity, accuracy and traceability of the data, and provides a solid foundation for subsequent analysis and decision-making. The data retrieval module can intelligently retrieve the contour measurement information of multiple target sample slopes based on the deformation influence information and initial contour information of the current slope. This intelligent data retrieval method improves the pertinence and efficiency of data use, helps to quickly identify historical cases similar to the current slope, and provides valuable references for subsequent monitoring and inspection. The real-time monitoring module can set up real-time monitoring devices at multiple contour measurement locations to collect slope deformation data in real time. This helps to timely detect small deformations of the slope and provide a valuable time window for early warning and emergency response. The inspection scheduling module can dynamically adjust the slope contour inspection plan of the amphibious inspection robot based on real-time monitoring data. This flexible inspection scheduling method ensures the efficient execution of inspection tasks while reducing unnecessary resource consumption. The contour measurement module can accurately determine the current contour measurement information of the current slope based on the slope contour data collected by the amphibious inspection robot. This provides accurate data support for subsequent deformation analysis and evaluation. The deformation monitoring module can comprehensively and accurately evaluate the deformation of the current slope based on the initial profile information and the current profile measurement information. This helps to timely discover the stability problems of the slope and provide a scientific basis for subsequent maintenance and repair work. By integrating multiple modules, the whole process from data acquisition, processing, analysis to decision support is covered. This provides comprehensive technical support for slope profile measurement work and helps to improve the scientificity and accuracy of decision-making.

[0018] 2. The data management module achieves efficient classification and clustering of sample slopes by calculating the similarity between any two sample slopes in deformation influence information, initial contour information and contour measurement information. This similarity-based clustering method helps to discover the internal connection and law between sample slopes, and provides strong support for subsequent data analysis and decision-making. The data retrieval module adopts a layer-by-layer screening method when retrieving the contour measurement information of the target sample slope based on the deformation influence information and initial contour information of the current slope. First, the target first cluster is determined by calculating the similarity of the deformation influence information between the current slope and the center of each first cluster; then, within the target first cluster, the target second cluster is determined by calculating the similarity of the initial contour information between the current slope and the center of each second cluster. This method ensures that the retrieved target sample slope has a high degree of similarity with the current slope in deformation influence information and initial contour information, thereby improving the pertinence and accuracy of the retrieved data. Through the clustering algorithm and the layer-by-layer screening method, the data retrieval module can quickly screen out sample slopes similar to the target slope from a large number of sample slopes and retrieve their contour measurement information. This method greatly improves the efficiency of data retrieval and reduces unnecessary data processing and transmission time. By clustering the sample slopes, it is possible to identify groups of slopes with similar characteristics, thus avoiding repeated monitoring and inspections. This not only saves manpower, material and financial resources, but also improves the efficiency and accuracy of monitoring and inspections.

[0019] 3. By establishing a scheme evaluation system containing multiple evaluation indicators, the advantages and disadvantages of different candidate inspection schemes can be comprehensively and objectively evaluated. This comprehensive evaluation method helps to find the optimal solution in multiple dimensions and ensure that the generated inspection scheme is both efficient and safe. The entire process is highly automated and relies on the calculation and analysis capabilities of the scheme generation model, reducing the subjectivity of manual intervention and decision-making. Intelligent generation and optimization of inspection schemes improves work efficiency and accuracy. The inspection scheme can be dynamically adjusted according to real-time environmental information and changes in the inspection position of the slope profile. In response to different terrain and climatic conditions, the amphibious inspection robot can flexibly switch operating modes (air-dwelling, water-dwelling, and road-dwelling) to ensure the smooth completion of the inspection task. By screening the target candidate inspection schemes through multiple iterations and calculating the global similarity standard deviation to evaluate the diversity of the schemes, duplication and redundancy are avoided. This method can efficiently converge to the optimal solution while ensuring the diversity and innovation of the inspection scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0021] Figure 1It is a module schematic diagram of a slope profile measurement system based on an amphibious inspection robot according to some embodiments of this specification;

[0022] Figure 2 It is a schematic diagram of a process for determining a slope profile inspection scheme of an amphibious inspection robot according to some embodiments of this specification;

[0023] Figure 3 It is a flow chart of a slope profile measurement method based on an amphibious inspection robot according to some embodiments of this specification. DETAILED DESCRIPTION

[0024] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0025] Figure 1 is a schematic diagram of a module of a slope profile measurement system based on an amphibious inspection robot according to some embodiments of this specification, such as Figure 1 As shown, the slope profile measurement system based on the amphibious inspection robot may include a data management module, a data retrieval module, a real-time monitoring module, an inspection scheduling module, a profile measurement module and a deformation monitoring module.

[0026] The data management module can be used to obtain and process deformation influence information, initial contour information and contour measurement information of multiple sample slopes.

[0027] The deformation influencing information of the sample slope may include information of multiple factors related to the slope deformation. For example, the deformation influencing information of the sample slope may include geological structure (e.g., fold morphology of the slope section, rock formation occurrence, degree of development of faults and joint fissures, and characteristics of new tectonic movement, etc.), groundwater information (e.g., hydrostatic pressure, buoyancy, and hydrodynamic pressure, etc.), climate conditions (e.g., temperature, humidity, rainfall, wind speed, etc.), and other information.

[0028] It is understandable that the fold morphology, rock formation occurrence, degree of development of faults and joint fissures, and characteristics of neotectonic movement of the slope section will significantly affect the stability of the slope. Areas with complex geological structures, such as areas with strong folds, numerous faults, and developed rock fissures, have a higher risk of slope deformation. The rise or fall of groundwater levels, the flow of groundwater, and the quality of fault permeability may all lead to softening, loss, or deformation of the slope rock mass. Rainfall can cause the softening and loss of slope soil, while drought and high temperature may aggravate the weathering of the slope rock mass. Sudden changes in temperature will aggravate the weathering of the slope rock mass, leading to natural slope cutting or natural peeling, and changing the shape and slope of the slope.

[0029] The initial contour information may include the contour information of the sample slope that has not been deformed after construction. Through technologies such as 3D laser scanning or drone remote sensing, the 3D contour data of the slope can be obtained, and then a 3D model can be constructed to intuitively display the contour morphological characteristics of the slope.

[0030] The profile metrology information includes profile deformation monitoring data at multiple locations. Specifically, profile deformation monitoring devices may be set at multiple locations of the sample slope to collect profile deformation monitoring data at multiple locations of the sample slope. The profile deformation monitoring device may include at least vibration sensors, tilt sensors, strain gauges and other equipment.

[0031] Preferably, the data management module processes deformation influence information, initial contour information and contour measurement information of multiple sample slopes, including:

[0032] Calculate the first information similarity of the deformation influence information of any two sample slopes;

[0033] Calculate the second information similarity of the initial contour information of any two sample slopes;

[0034] Calculate the third information similarity of the contour measurement information of any two sample slopes;

[0035] Clustering the plurality of sample slopes according to the first information similarity of the deformation influence information of any two sample slopes by a clustering algorithm (for example, a hierarchical clustering algorithm, etc.) to determine a plurality of first clusters;

[0036] For each first cluster, clustering the plurality of sample slopes included in the first cluster according to the second information similarity of the initial contour information of any two sample slopes included in the first cluster by using a clustering algorithm to determine a plurality of second clusters included in the first cluster;

[0037] For each second cluster, a clustering algorithm is used to cluster the multiple sample slopes included in the second cluster according to the third information similarity of the contour measurement information of any two sample slopes included in the second cluster, so as to determine multiple third clusters included in the second cluster.

[0038] Specifically, since the data of different factors may have different dimensions and value ranges, it is necessary to standardize the deformation influence information of any two sample slopes for subsequent calculations. The standardization method can adopt Z-score standardization, Min-Max standardization, etc. The deformation influence information of each sample slope is regarded as a high-dimensional vector. Use the similarity calculation method in the vector space model, such as cosine similarity, Euclidean distance, Manhattan distance, etc., to calculate the first information similarity of the deformation influence information of any two sample slopes.

[0039] The data management module can extract the contour line of the sample slope from the three-dimensional model of the sample slope. The extracted contour line is used to reflect the contour characteristics of the sample slope. The extracted contour line is subjected to metrological analysis to calculate geometric parameters such as the length, curvature, and area of ​​the contour line. The second information similarity of the initial contour information of the sample slope is calculated according to the following formula:

[0040] ,

[0041] in, For the The initial contour information of the sample slope is The second information similarity of the initial contour information of the sample slopes, For the The sample slope is in The values ​​of the geometric parameters, For the The value of the nth geometric parameter of the sample slope, is the total number of geometric parameters.

[0042] The data management module can extract key features from the profile deformation monitoring data, such as vibration amplitude, frequency, tilt angle, strain value, etc. These features should be able to reflect the deformation of the slope. The deformation area of ​​the sample slope and the degree of deformation of each deformation area are determined according to the key features, and then the third information similarity of the profile measurement information of the two sample slopes is calculated according to the deformation area of ​​the sample slope and the degree of deformation of each deformation area.

[0043] The data retrieval module can be used to obtain the deformation influence information and initial contour information of the current slope, and based on the deformation influence information and initial contour information of the current slope, retrieve contour measurement information of multiple target sample slopes.

[0044] As a preferred embodiment, it specifically includes:

[0045] Calculating fourth information similarity between deformation influence information of the current slope and deformation influence information of the cluster center of each first cluster;

[0046] determining a target first cluster from the plurality of first clusters based on fourth information similarity between the deformation influence information of the current slope and the deformation influence information of the cluster center of each first cluster;

[0047] Calculating fifth information similarity between initial contour information of the current slope and initial contour information of the cluster center of each second cluster included in the target first cluster;

[0048] Based on the fifth information similarity between the initial contour information of the current slope and the initial contour information of the cluster center of each second cluster included in the target first cluster, the target second cluster is determined from the multiple second clusters, and the contour measurement information of the multiple sample slopes included in the target second cluster is retrieved as the contour measurement information of the multiple target sample slopes.

[0049] Specifically, the method of calculating the fourth information similarity is similar to the method of calculating the first information similarity, which is not repeated here. The first cluster to which the cluster center whose fourth information similarity is greater than the fourth information similarity threshold belongs can be used as the target first cluster.

[0050] The method of calculating the fifth information similarity is similar to the method of calculating the second information similarity, which is not repeated here. The second cluster to which the cluster center of the target first cluster belongs and whose fifth information similarity is greater than the fifth information similarity threshold may be used as the target second cluster.

[0051] The real-time monitoring module can be used to determine multiple contour measurement positions of the current slope based on contour measurement information of multiple target sample slopes, and to set multiple real-time monitoring devices at the contour measurement positions.

[0052] Specifically, the real-time monitoring device may include vibration sensors, tilt sensors, strain gauges and other equipment.

[0053] Preferably, the real-time monitoring device may further include contour markers.

[0054] Preferably, the real-time monitoring module determines multiple contour measurement positions of the current slope based on contour measurement information of multiple target sample slopes, including:

[0055] For each target sample slope, based on the contour measurement information of the target sample slope, the contour deformation risk position of the target sample slope is determined;

[0056] The position determination model is used to determine multiple contour measurement positions of the current slope according to the contour deformation risk position of each target sample slope and the initial contour information of the current slope, wherein the position determination model can be a convolutional neural network (CNN) model.

[0057] Specifically, for each target sample slope, the position where deformation occurs in the target sample slope can be determined based on the contour deformation monitoring data of multiple positions, and it can be used as the contour deformation risk position of the target sample slope. The contour deformation risk positions of the target sample slope are clustered according to the distance between any two contour deformation risk positions of the target sample slope through a clustering algorithm to determine the risk area of ​​the target sample slope. Based on the initial contour information of the target sample slope, the risk characteristics of each risk area of ​​the target sample slope are extracted (for example, geological structure characteristics, groundwater characteristics, geometric parameters (for example, slope angle), etc.).

[0058] The position determination model is used to determine the risk area of ​​the current slope and the deformation risk value of each risk area according to the risk characteristics of each risk area of ​​each target sample slope, the initial contour information of the current slope, and the deformation impact information. According to the deformation risk value of each risk area, the density of the contour deformation risk positions set in the risk area is determined, and the contour deformation risk positions of the risk area are determined according to the density. The contour deformation risk positions are evenly distributed in the risk area according to the density.

[0059] For example, the density of the contour deformation risk position set in the risk area can be calculated according to the following formula:

[0060] ,

[0061] in, To be set in The density of the contour deformation risk position of each risk area, For the The deformation risk value of each risk area is is the preset deformation risk value, The default density.

[0062] The inspection scheduling module can be used to determine the slope profile inspection plan of the amphibious inspection robot based on the real-time monitoring data collected by multiple real-time monitoring devices.

[0063] Specifically, the amphibious inspection robot has a unique mechanical structure that integrates amphibious technologies on land, water and air. Its design takes into account both structural strength and flexibility, allowing the robot to switch freely and operate stably in complex and changing environments. Its core rotor mechanism enables air flight, while rolling on land and diving underwater are achieved through special wheels and waterproof sealed wings respectively. This enables the robot to perform complex inspection tasks. The mechanical structure uses lightweight and corrosion-resistant materials, which not only ensures the durability of the amphibious inspection robot, but also reflects the design concept of environmental protection and sustainability.

[0064] The motion system of the amphibious inspection robot demonstrates excellent multi-mode adaptability. In the air, the precise control of the rotor mechanism enables stable flight and flexible navigation; on land, the efficient drive wheel or track system ensures smooth rolling and rapid movement; underwater, it relies on waterproof and sealed propellers to achieve stealth and precise positioning. The flexible switching of these three modes enables the amphibious inspection robot to efficiently perform inspection tasks in complex environments, demonstrating its unique advantages and application value.

[0065] Air-dwelling mode: The robot can take off by flipping the wheels and rotating the wings at high speed. The direction can be controlled by controlling the rotation speed of the wings. At the same time, air-dwelling is suitable for various complex terrains, and it can better scout the terrain and other mission requirements in the air. It adopts a fixed-wing structure, which is suitable for long-term flight. A gyroscope is added to sense the attitude information of the amphibious inspection robot in real time and maintain flight stability through control algorithms. The flight controller of the amphibious inspection robot receives sensor data, processes flight commands, and adjusts the movement of the flight mechanism in real time to respond to emergencies or environmental changes. At the same time, the amphibious inspection robot is also equipped with an advanced communication system to realize functions such as command transmission, data transmission, and remote control operation.

[0066] Road-dwelling mode: It combines the characteristics of multiple modes of movement. The wheeled structure is driven by the motor to rotate, and the direction is controlled by the steering gear linkage direction rod. The wheeled structure has the characteristics of fast speed. If you encounter smooth terrain during reconnaissance, you can quickly reconnaissance and save time. An obstacle avoidance module is designed for complex terrain structures. In the road-dwelling mode, the amphibious inspection robot relies on the motor as a power source to drive the wheels to move. The control system is responsible for receiving instructions and controlling the movement direction and speed of the amphibious inspection robot to ensure that it can move according to the preset route or task requirements.

[0067] Aquatic mode: The sealing structure of the amphibious inspection robot is the key to waterproofing. Consider all potential water inlet points, and use waterproof glue and sealing gaskets at these locations to ensure that the inside of the robot is completely isolated from the outside. Power is provided by the rotation of the propellers. The propellers facing downward provide upward buoyancy to prevent the amphibious inspection robot from sinking. The propellers facing backward provide power forward. Through coordination, the whole machine can collect and inspect more smoothly. Water samples are collected by using the servo to control the opening and closing of the valve, and a micro water tank is built in.

[0068] The amphibious inspection robot includes an ultrasonic obstacle avoidance module and uses the HC-SR04 ultrasonic ranging module to measure distance. The amphibious inspection robot is equipped with an ultrasonic obstacle avoidance system that can accurately detect and avoid surrounding obstacles. Whether flying in the air, driving on land or diving underwater, the system can work effectively to ensure the safety of the robot's operation. Ultrasonic obstacle avoidance technology provides the amphibious inspection robot with powerful environmental perception capabilities and enhances its autonomous operation capabilities.

[0069] The amphibious inspection robot includes a laser radar module, which determines the distance to surrounding objects by emitting laser beams and measuring the time it takes for them to reflect back, thereby building a three-dimensional model of the environment. For the amphibious inspection robot, laser radar can be used not only in land environments, but also in air and underwater environments.

[0070] The amphibious inspection robot also includes an image transmission perception module and uses a BH-205AV model wireless micro camera, which has a small mass, light size, and a wide and clear field of view.

[0071] The amphibious inspection robot also includes a PID (Proportional Integral Derivative) control module. In order to more accurately represent the posture of the amphibious inspection robot and increase the robustness of the system, the PID control module adopts a dual-loop cascade PID control, with an angle PID controller for the outer loop and an angular velocity PID controller for the inner loop. Compared with the traditional single-loop angle PID controller, this PID control module has a stronger ability to resist external interference and is more stable. Using the PID algorithm, it generates a control signal by calculating the proportion, integral, and differential of the error, thereby adjusting the posture and position of the amphibious inspection robot in the air. The angle value of the robot at the current moment is collected by various sensors, and then compared with the expected angle, and the pitch angle is continuously adjusted to stabilize the posture and position of the amphibious inspection robot.

[0072] The amphibious inspection robot uses the STM32 microcontroller, which has excellent stability and reliability and can maintain normal operation even in harsh environmental conditions. This is crucial for amphibious inspection robots that need to perform inspection tasks in complex environments. Connect various sensors and actuators to the STM32, including sensors such as gyroscopes, accelerometers, and barometers to obtain the posture and position information of the amphibious inspection robot; and actuators such as motor drivers to control the movements of the amphibious inspection robot. At the same time, the corresponding interfaces and pins of the STM32 need to be configured to ensure normal communication with sensors and actuators.

[0073] By reading sensor data, STM32 can obtain the real-time status of the amphibious inspection robot. These data need to be processed by the PID control algorithm to calculate the posture, speed and position of the amphibious inspection robot. The powerful computing power of STM32 enables it to run these algorithms in real time and issue control instructions based on the calculation results. After the STM32 and LoRa module are successfully connected and the parameters are configured, they can work together to achieve remote data transmission. STM32 also needs to receive signals from the remote control. These signals may include control instructions such as flight direction and speed. STM32 needs to parse these signals and adjust the actions of the amphibious inspection robot according to the instructions.

[0074] The control board used by the amphibious inspection robot is FPGA (Field-Programmable Gate Array). The robot control system is realized by using field programmable gate array (FPGA) devices, which are highly integrated, high-speed, and efficient. It is easy to reprogram on site and realize multiple configurations. Other control modules can be integrated in FPGA, which is small in size and highly reliable.

[0075] The linkage between the STM32 main control board and the FPGA board is mainly realized through specific communication interfaces and protocols. Through these interfaces and protocols, the STM32 main control board can send instructions to the FPGA board to tell it how to control the servo; at the same time, the FPGA board can also feed back the status information of the servo to the STM32 main control board so that it can monitor and adjust in real time. During the linkage process, the STM32 main control board will calculate the angle and speed that the servo should rotate according to the requirements of the inspection task, and send this information to the FPGA board through the communication interface. After receiving the instruction, the FPGA board will immediately execute the corresponding control algorithm to drive the servo to rotate as required. At the same time, the FPGA board will also monitor the status of the servo in real time, such as the rotation angle, speed, etc., and feed back this information to the STM32 main control board. The STM32 main control board makes real-time adjustments based on the feedback information to ensure that the motion trajectory and posture of the three-dimensional inspection robot meet expectations. The realization of this linkage enables the three-dimensional inspection robot to have higher motion accuracy and stability, so as to better complete the inspection task. At the same time, because FPGA has the ability of high-speed computing and parallel processing, it can greatly improve the response speed and processing efficiency.

[0076] The BUG2 algorithm is an obstacle avoidance algorithm. Its core idea is that if the amphibious inspection robot encounters an obstacle during its movement, it will move along the boundary of the obstacle until it can move directly to the target position. This algorithm can ensure that the amphibious inspection robot can effectively avoid obstacles when encountering them and eventually reach the target point.

[0077] The installation of solar modules can also enhance the stability of the amphibious inspection robot. Since the solar modules can generate electricity autonomously, the amphibious inspection robot can rely on the solar modules to continue working even in the event of a power failure or power shortage. This energy backup mechanism is crucial for the amphibious inspection robot that performs critical tasks, as it ensures that the amphibious inspection robot can maintain a stable operating state in a variety of complex and unpredictable environments.

[0078] Preferably, the inspection scheduling module determines the slope profile inspection plan of the amphibious inspection robot according to the real-time monitoring data collected by multiple real-time monitoring devices, including:

[0079] Determine multiple slope profile inspection locations based on real-time monitoring data collected by multiple real-time monitoring devices;

[0080] Obtain real-time environmental information based on multiple slope profile inspection locations;

[0081] Through the scheme generation model, the slope profile inspection scheme of the amphibious inspection robot is determined according to multiple slope profile inspection positions and real-time environmental information, wherein the scheme generation model can be a convolutional neural network model.

[0082] Preferably, the inspection scheduling module determines multiple slope profile inspection positions according to the real-time monitoring data collected by multiple real-time monitoring devices, including:

[0083] Determine the positional correlation coefficient of any two contour measurement positions;

[0084] A plurality of slope profile inspection positions are determined based on the real-time monitoring data collected by a plurality of real-time monitoring devices and the position correlation coefficient of any two profile measurement positions.

[0085] Specifically, the correlation coefficient determination model can be used to determine the position correlation coefficient of any two contour measurement positions according to the risk characteristics of each risk area of ​​each target sample slope, the initial contour information of the current slope, and the deformation influence information. A clustering algorithm is used to cluster multiple contour measurement positions according to the position correlation coefficients of any two contour measurement positions and the distance between any two contour measurement positions to determine multiple position clustering clusters.

[0086] For each location cluster, the deformation risk value of the location cluster is determined by the risk prediction model based on the real-time monitoring data collected by the real-time monitoring devices set at the multiple contour measurement positions included in the location cluster, wherein the risk prediction model can be a long short-term memory network (LSTM) model. If the deformation risk value of the location cluster is greater than the deformation risk value threshold, the location cluster is regarded as a risk location cluster, and the contour measurement position corresponding to the cluster center of the risk location cluster is regarded as the slope contour inspection position.

[0087] Preferably, the inspection scheduling module determines the slope profile inspection scheme of the amphibious inspection robot through a scheme generation model according to multiple slope profile inspection positions and real-time environmental information, including:

[0088] S11. Establishing a scheme evaluation system, wherein the scheme evaluation system may include multiple evaluation indicators, for example, the scheme evaluation system may include at least an inspection time indicator, an inspection energy consumption indicator, an inspection terrain adaptability indicator, an inspection environment adaptability indicator, an obstacle avoidance risk evaluation indicator, and a mode switching number indicator, wherein the mode switching number indicator is used to evaluate the number of times the amphibious inspection robot switches operating modes in the slope contour inspection scheme, and the operating modes may include an air-dwelling mode, an aquatic-dwelling mode, and a road-dwelling mode;

[0089] S12, generating a plurality of candidate slope profile inspection schemes according to a plurality of slope profile inspection positions by a scheme generation unit of a scheme generation model;

[0090] S13. Calculate the scheme score of each candidate slope profile inspection scheme according to the scheme evaluation system and real-time environmental information;

[0091] S14, screening multiple candidate slope profile inspection schemes according to the scheme score of each candidate slope profile inspection scheme to obtain multiple target candidate slope profile inspection schemes;

[0092] S15, repeatedly executing S12 to S14 until the number of target candidate slope profile inspection schemes is greater than a preset number threshold. It can be understood that the multiple target candidate slope profile inspection schemes of the current iteration number are the union of the target candidate slope profile inspection schemes obtained in this iteration and the multiple target candidate slope profile inspection schemes obtained in the previous iteration;

[0093] S16, calculating the global similarity standard deviation of multiple target candidate slope profile inspection schemes;

[0094] S17, determining whether the global similarity standard deviation of multiple target candidate slope profile inspection schemes is greater than a preset global similarity standard deviation threshold, if so, executing S18, if not, executing S12;

[0095] S18. Generate a slope profile inspection plan for the amphibious inspection robot based on multiple target candidate slope profile inspection plans and the plan score of each target candidate slope profile inspection plan through the plan optimization unit of the plan generation model.

[0096] Specifically, for each candidate slope profile inspection scheme, the scores of the candidate slope profile inspection scheme in multiple evaluation indicators can be calculated, and the scores of the candidate slope profile inspection scheme in multiple evaluation indicators can be weighted summed to obtain the scheme score of the candidate slope profile inspection scheme, and the candidate slope profile inspection schemes with scheme scores less than the scheme score threshold can be screened out, and the candidate slope profile inspection schemes with scheme scores greater than or equal to the scheme score threshold can be taken as the target candidate slope profile inspection schemes.

[0097] For example, the global similarity standard deviation can be calculated according to the following formula:

[0098] ,

[0099] ,

[0100] in, is the global similarity standard deviation, For the The target candidate slope profile inspection plan and the The similarity between the target candidate slope profile inspection schemes, is the total number of target candidate slope profile inspection schemes, is the total number of evaluation indicators, For the The target candidate slope profile inspection plan is in the The scores of the evaluation indicators, For the The score of the mth evaluation index of the target candidate slope profile inspection scheme.

[0101] Preferably, step S17 may specifically include the following steps:

[0102] S171, determining whether the global similarity standard deviation of multiple target candidate slope profile inspection schemes is greater than a preset global similarity standard deviation threshold, if so, executing S18, if not, executing S172;

[0103] S172, determine whether EG is greater than the difference threshold, if so, execute S173, if not, execute S12, where G is a preset difference (for example, 3, 4, 5, etc.);

[0104] S173, clustering multiple target candidate slope profile inspection schemes according to the similarity between any two target candidate slope profile inspection schemes through a clustering algorithm, determining multiple scheme clusters, and sorting the multiple scheme clusters from large to small according to the number of target candidate slope profile inspection schemes included in each scheme cluster;

[0105] S173, determine whether E-1 is greater than a preset quantity threshold, if so, execute S174, if not, execute S12;

[0106] S174, according to the sorting result of the scheme clusters, screen out a target candidate slope profile inspection scheme from the top-ranked scheme cluster, calculate the global similarity standard deviation after the screening, and execute S175;

[0107] S175, determine whether the global similarity standard deviation after screening is greater than the preset global similarity standard deviation threshold, if so, execute S18, if not, update E=E-1, and execute S173.

[0108] Preferably, for the top-ranked scheme cluster, the mean similarities corresponding to the target candidate slope profile inspection schemes included in the scheme cluster can be calculated, and the target candidate slope profile inspection schemes included in the scheme cluster can be sorted from large to small according to the mean similarities, and the top-ranked target candidate slope profile inspection scheme can be screened out.

[0109] For example, the similarity mean corresponding to the target candidate slope profile inspection schemes included in the scheme cluster can be calculated according to the following formula:

[0110] ,

[0111] in, The solution cluster includes The mean similarity value corresponding to the target candidate slope profile inspection schemes, is the total number of target candidate slope profile inspection schemes included in the scheme cluster, The solution cluster includes The target candidate slope profile inspection plan and the The similarity of the target candidate slope profile inspection schemes.

[0112] It can be understood that by adding the judgment steps of S172 and S173, the decision-making process is made more detailed and accurate. This helps to further analyze the diversity and differences of the target candidate slope profile inspection schemes when the global similarity standard deviation does not meet the requirements, so as to make more reasonable decisions. The clustering algorithm is introduced in step S173 to aggregate the target candidate slope profile inspection schemes with higher similarity into clusters, which helps to quickly identify schemes with similar features. This not only improves the analysis efficiency, but also helps to find potential optimization directions. Through the screening and recalculation of the global similarity standard deviation in steps S174 and S175, the dynamic adjustment of the target candidate slope profile inspection scheme is realized. This iterative optimization method helps to gradually approach the optimal solution and improve the diversity and quality of the inspection scheme. The refined step S17 can flexibly respond to different situations. When the global similarity standard deviation does not meet the requirements, it can be adjusted by clustering analysis, screening schemes, etc.; when the adjusted global similarity standard deviation still does not meet the requirements, new candidate schemes are continued to be generated for iterative optimization. This flexibility ensures the stability and reliability of the entire inspection plan generation process. By reducing redundant and repeated plans, resource utilization efficiency is improved. Cluster analysis and screening steps help to eliminate plans with high similarity, thereby reducing unnecessary calculations and resource consumption.

[0113] The profile measurement module can be used to determine the current profile measurement information of the current slope based on the slope profile data collected by the amphibious inspection robot based on the slope profile inspection plan.

[0114] Preferably, the profile measurement module determines the current profile measurement information of the current slope according to the slope profile data collected by the amphibious inspection robot based on the slope profile inspection scheme, including:

[0115] According to the slope image collected by the amphibious inspection robot based on the slope contour inspection scheme and the position correlation coefficient of any two contour measurement positions, the real-time relative position parameters between multiple contour markers are determined.

[0116] Specifically, for each position cluster, the real-time relative position parameters between the contour markers set at the multiple contour measurement positions included in the position cluster can be determined based on the slope image collected by the amphibious inspection robot based on the slope contour inspection scheme. The real-time relative position parameters may include real-time relative orientation parameters, real-time relative tilt angle parameters, real-time relative distance parameters, etc.

[0117] The deformation monitoring module can be used to determine the deformation information of the current slope based on the initial contour information of the current slope and the current contour measurement information.

[0118] Preferably, the deformation monitoring module determines the deformation information of the current slope according to the initial contour information of the current slope and the current contour measurement information, including:

[0119] Determine the initial relative position parameters between multiple contour markers according to the initial contour information of the current slope and the position correlation coefficient of any two contour measurement positions;

[0120] The deformation information of the current slope is determined according to the initial relative position parameters between the multiple contour markers and the real-time relative position parameters between the multiple contour markers.

[0121] Specifically, for each position cluster, the initial relative position parameters between the contour markers set at the multiple contour measurement positions included in the position cluster can be determined according to the initial contour information of the current slope. The initial relative position parameters may include initial relative orientation parameters, initial relative inclination angle parameters, initial relative distance parameters, etc.

[0122] The relative position parameter difference between the initial relative position parameters of multiple contour markers and the real-time relative position parameters of multiple contour markers can be calculated, and the deformation information of the current slope, including deformation position, deformation direction, deformation speed, etc., can be determined based on the relative position parameter difference.

[0123] Figure 3 is a flow chart of a slope profile measurement method based on an amphibious inspection robot according to some embodiments of this specification, such as Figure 3 As shown, the slope profile measurement method based on the amphibious inspection robot may include the following process.

[0124] Step 310, acquiring and processing deformation influence information, initial contour information and contour measurement information of a plurality of sample slopes;

[0125] Step 320, based on the deformation influence information and initial contour information of the current slope, retrieve contour measurement information of multiple target sample slopes;

[0126] Step 330, determining multiple contour measurement positions of the current slope based on the contour measurement information of the multiple target sample slopes;

[0127] Step 340, setting a plurality of real-time monitoring devices at the contour measurement position;

[0128] Step 350, determining a slope profile inspection plan of the amphibious inspection robot according to the real-time monitoring data collected by the multiple real-time monitoring devices;

[0129] Step 360, determining current profile measurement information of the current slope according to the slope profile data collected by the amphibious inspection robot based on the slope profile inspection scheme;

[0130] Step 370, determining deformation information of the current slope according to the initial profile information of the current slope and the current profile measurement information.

[0131] The slope profile measurement method based on the amphibious inspection robot can be applied to the slope profile measurement system based on the amphibious inspection robot. For more descriptions of the slope profile measurement method based on the amphibious inspection robot, please refer to the relevant descriptions of the slope profile measurement system based on the amphibious inspection robot, which will not be repeated here.

[0132] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. The slope profile measurement system based on the amphibious inspection robot is characterized by: include: A data management module, used to obtain and process deformation influence information, initial profile information and profile measurement information of multiple sample slopes, wherein the profile measurement information includes profile deformation monitoring data of multiple locations; A data retrieval module, used to obtain deformation influence information and initial contour information of the current slope, and based on the deformation influence information and initial contour information of the current slope, retrieve contour measurement information of multiple target sample slopes; A real-time monitoring module, for determining a plurality of contour measurement positions of the current slope based on the contour measurement information of the plurality of target sample slopes, and setting a plurality of real-time monitoring devices at the contour measurement positions; An inspection scheduling module, used to determine a slope profile inspection plan of the amphibious inspection robot according to the real-time monitoring data collected by the multiple real-time monitoring devices; A profile measurement module, used to determine the current profile measurement information of the current slope according to the slope profile data collected by the amphibious inspection robot based on the slope profile inspection scheme; A deformation monitoring module, used to determine deformation information of the current slope according to initial profile information and current profile measurement information of the current slope; The data management module processes deformation influence information, initial contour information and contour measurement information of multiple sample slopes, including: Calculate the first information similarity of the deformation influence information of any two sample slopes; Calculate the second information similarity of the initial contour information of any two sample slopes; Calculate the third information similarity of the contour measurement information of any two sample slopes; Clustering multiple sample slopes according to the first information similarity of deformation influence information of any two sample slopes through a clustering algorithm to determine multiple first clusters; For each of the first clusters, clustering the multiple sample slopes included in the first cluster according to the second information similarity of the initial contour information of any two sample slopes included in the first cluster, to determine multiple second clusters included in the first cluster; For each of the second clusters, clustering the plurality of sample slopes included in the second cluster according to the third information similarity of the contour measurement information of any two sample slopes included in the second cluster, to determine a plurality of third clusters included in the second cluster; The data retrieval module retrieves contour measurement information of multiple target sample slopes based on the deformation influence information and initial contour information of the current slope, including: Calculating fourth information similarity between the deformation influence information of the current slope and the deformation influence information of the cluster center of each of the first clusters; determining a target first cluster from the plurality of first clusters based on fourth information similarity between the deformation influence information of the current slope and the deformation influence information of the cluster center of each of the first clusters; Calculating a fifth information similarity between the initial contour information of the current slope and the initial contour information of the cluster center of each second cluster included in the target first cluster; Based on the fifth information similarity between the initial contour information of the current slope and the initial contour information of the cluster center of each second cluster included in the target first cluster, determining a target second cluster from the plurality of second clusters, and retrieving the contour measurement information of the plurality of sample slopes included in the target second cluster as the contour measurement information of the plurality of target sample slopes; The real-time monitoring module determines a plurality of contour measurement positions of the current slope based on the contour measurement information of the plurality of target sample slopes, including: For each of the target sample slopes, determining a contour deformation risk position of the target sample slope based on contour measurement information of the target sample slope; A plurality of contour measurement positions of the current slope are determined by a position determination model according to the contour deformation risk position of each target sample slope and the initial contour information of the current slope.

2. The slope profile measurement system based on the amphibious inspection robot according to claim 1 is characterized in that: The inspection scheduling module determines the slope profile inspection plan of the amphibious inspection robot according to the real-time monitoring data collected by the multiple real-time monitoring devices, including: Determining a plurality of slope profile inspection positions according to the real-time monitoring data collected by the plurality of real-time monitoring devices; Based on the multiple slope profile inspection positions, real-time environmental information is obtained; Through the scheme generation model, the slope profile inspection scheme of the amphibious inspection robot is determined according to the multiple slope profile inspection positions and the real-time environmental information.

3. The slope profile measurement system based on the amphibious inspection robot according to claim 2 is characterized in that: The inspection scheduling module determines a plurality of slope profile inspection positions according to the real-time monitoring data collected by the plurality of real-time monitoring devices, including: determining a position correlation coefficient between any two of the profile measurement positions; The plurality of slope profile inspection positions are determined based on the real-time monitoring data collected by the plurality of real-time monitoring devices and the position correlation coefficients of any two of the profile measurement positions.

4. The slope profile measurement system based on the amphibious inspection robot according to claim 3 is characterized in that: The inspection scheduling module determines the slope profile inspection scheme of the amphibious inspection robot according to the multiple slope profile inspection positions and the real-time environmental information through a scheme generation model, including: S11. Establish a program evaluation system; S12, generating a plurality of candidate slope profile inspection schemes according to the plurality of slope profile inspection positions by a scheme generation unit of a scheme generation model; S13, calculating a scheme score of each of the candidate slope profile inspection schemes according to the scheme evaluation system and the real-time environmental information; S14, screening the plurality of candidate slope profile inspection schemes according to the scheme score of each candidate slope profile inspection scheme to obtain a plurality of target candidate slope profile inspection schemes; S15, repeatedly executing S12 to S14 until the number of the target candidate slope profile inspection schemes is greater than a preset number threshold; S16, calculating the global similarity standard deviation of multiple target candidate slope profile inspection schemes; S17, determining whether the global similarity standard deviation of multiple target candidate slope profile inspection schemes is greater than a preset global similarity standard deviation threshold, if so, executing S18, if not, executing S12; S18. Generate a slope profile inspection plan for the amphibious inspection robot based on multiple target candidate slope profile inspection plans and the plan score of each target candidate slope profile inspection plan through the plan optimization unit of the plan generation model.

5. The slope profile measurement system based on the amphibious inspection robot according to claim 3 is characterized in that: The real-time monitoring device includes at least a contour marker; The contour measurement module determines the current contour measurement information of the current slope according to the slope contour data collected by the amphibious inspection robot based on the slope contour inspection scheme, including: According to the slope image collected by the amphibious inspection robot based on the slope contour inspection scheme and the position correlation coefficient of any two of the contour measurement positions, the real-time relative position parameters between the multiple contour markers are determined.

6. The slope profile measurement system based on the amphibious inspection robot according to claim 5 is characterized in that: The deformation monitoring module determines the deformation information of the current slope according to the initial contour information and the current contour measurement information of the current slope, including: Determining initial relative position parameters between a plurality of contour markers according to the initial contour information of the current slope and the position correlation coefficients of any two contour measurement positions; The deformation information of the current slope is determined according to the initial relative position parameters between the plurality of contour markers and the real-time relative position parameters between the plurality of contour markers.

7. The slope profile measurement method based on the amphibious inspection robot is characterized by: The slope profile measurement system based on the amphibious inspection robot applied to any one of claims 1 to 6 comprises: Acquire and process deformation influence information, initial profile information and profile measurement information of multiple sample slopes, wherein the profile measurement information includes profile deformation monitoring data at multiple locations; Based on the deformation influence information and initial contour information of the current slope, retrieving contour measurement information of multiple target sample slopes; Determining a plurality of contour measurement positions of the current slope based on the contour measurement information of the plurality of target sample slopes; A plurality of real-time monitoring devices are arranged at the contour measurement position; Determine a slope profile inspection plan of the amphibious inspection robot according to the real-time monitoring data collected by the multiple real-time monitoring devices; Determine the current profile measurement information of the current slope according to the slope profile data collected by the amphibious inspection robot based on the slope profile inspection scheme; Deformation information of the current slope is determined according to the initial contour information and the current contour measurement information of the current slope.

Images