Underwater leveling machine path optimization method based on foundation bed pore characteristics

Abstract

The invention discloses an underwater leveling machine path optimization method based on foundation bed pore characteristics, and relates to the technical field of ocean engineering. The method comprises the following steps: measuring three-dimensional topographic data and pore distribution characteristic data of a to-be-leveled water area foundation bed by using a multi-source sensing and control system and a multi-beam sounding method, establishing a digital topographic model of the to-be-leveled foundation bed, and generating candidate optimized leveling paths adapted to geological characteristics by constructing an initial construction path, the engineering feasibility of the leveling operation is evaluated, and the optimal position of each path point after iterative optimization is output as an optimized leveling path in combination with an optimization algorithm, so that the leveling of the underwater leveling machine is completed; according to the method, the pore distribution characteristics and three-dimensional topographic data of a seabed foundation bed are intelligently analyzed, reliable environment data are obtained, an initial construction blueprint is formed, a basis is provided for optimization, dynamic generation and online correction of a path are achieved by fusing pore characteristic analysis and an intelligent optimization algorithm, and efficient, uniform and high-precision autonomous leveling operation is achieved.

Description

Technical Field

[0001] The present invention relates to the field of marine engineering technology, and in particular to a method for optimizing the path of an underwater screed machine based on bed pore characteristics. Background Art

[0002] Currently, path planning for underwater leveling machines relies primarily on manual experience or fixed-pattern reciprocating operations. While simple and easy to implement, these methods suffer from the following limitations: 1. Lack of pore adaptability: Ignoring the spatial heterogeneity of the subgrade pore distribution leads to uneven density after leveling, impacting foundation stability. 2. Inadequate dynamic response: Static paths struggle to adapt to complex seabed topography (such as steep slopes and gullies) in real time, resulting in significant duplication or omissions in localized areas. 3. Significant efficiency bottlenecks: A high rate of redundant paths significantly increases operation time and energy consumption.

[0003] In existing technologies, leveling vessels rely solely on GPS positioning and simple tilt sensors, which are unable to perceive the impact of pore distribution on the leveling effect. This can easily lead to path deviations when porosity changes suddenly, resulting in poor geological adaptability, delayed dynamic response, and a lack of quantification of pore characteristics. Traditional methods fail to incorporate pore characteristics into path planning models, potentially causing local collapse. Summary of the Invention

[0004] The present invention aims to provide a method for optimizing the path of an underwater screed machine based on the porosity characteristics of the subgrade. This method establishes a multi-source sensing system integrating inclination sensors, pressure sensors, underwater cameras, and sonar devices to collect screed machine data in real time. This system optimizes the path and controls hydraulic outriggers, vector thrusters, and a ballast system to perform material laying and machine movement operations according to motion instructions. This method achieves efficient and precise screeding based on the porosity characteristics of the subgrade, and dynamically optimizes the screed machine's operating path.

[0005] The present invention provides a method for optimizing the path of an underwater screed machine based on the pore characteristics of the subgrade, comprising the following steps: Utilize a multi-source sensing and control system and multi-beam bathymetry to measure the three-dimensional topographic data and pore distribution characteristic data of the water bed to be leveled, establish a digital terrain model of the bed to be leveled, and adjust and update it in real time. Utilize a path planning algorithm to output the position of each leveling path point, and based on the position of each leveling path point, obtain the initial leveling path and construct the initial construction path. Based on the pore distribution characteristics of the subgrade and the initial leveling path, a geological sensitivity analysis algorithm is used to identify the operational weaknesses of the leveling machine in terms of porosity. A subgrade weakness feature set is constructed. Based on this subgrade weakness feature set, a geological parameter mapping method is used to make target-maintaining directional adjustments to the initial construction path. A candidate optimized leveling path adapted to the geological characteristics is generated using a fuzzy control method. The engineering feasibility of candidate optimized leveling paths is evaluated using terrain similarity metrics and a geological feature projection algorithm. The optimal position of each path point after iterative optimization is output as the optimized leveling path using the optimization algorithm. Based on the optimized leveling path, the pressure sensor and tilt sensor components in the multi-source perception and control system are used to complete the leveling of the underwater leveling machine. The parameters of the optimized leveling path are dynamically adjusted according to the operating effect of the leveling machine on the optimized leveling path, and the material laying operation and machine moving operation are completed by controlling the underwater leveling machine according to the action instructions.

[0006] The path planning algorithm is used to output the position of each leveling path point. The initial leveling path is obtained based on the position of each leveling path point. The construction of the initial construction path includes: The grid data of the digital terrain to be leveled is input, the key point set of the leveling path is calculated using the path planning algorithm, the coordinate sequence of the path points is output to obtain the initial leveling path, and the initial construction path is constructed by connecting the path points using a uniform B-spline curve.

[0007] Based on the pore distribution characteristic data of the subgrade and the initial leveling path, combined with the geological sensitivity analysis algorithm, the operational weaknesses of the leveling machine in terms of porosity are identified. The subgrade weakness feature set is constructed, including: Perform gradient analysis on the initial leveling path data to calculate the impact of porosity changes in different areas on the leveling quality, and use porosity as the main dimension of the base bed characteristics; Apply simulated disturbances to the bed area, record the leveling quality fluctuation data, quantify the leveling effect deviation under different geological disturbances, and identify areas with unstable quality; The disturbance sensitivity data were clustered and analyzed, and a bed weakness feature set containing geological parameters of weak areas was constructed according to the associated pore types.

[0008] Based on the weak feature set of the subgrade, the initial construction path is adjusted in a targeted manner by using the geological parameter mapping method. In combination with the fuzzy control method, candidate optimized leveling paths adapted to the geological characteristics are generated, including: Path control parameters are constructed based on the weak feature set, and a fuzzy control algorithm is established with leveling effect feedback as input and path parameter adjustment as output; For each leveling path, the parameter adjustment range is calculated through fuzzy reasoning according to the real-time detected porosity compliance rate; based on the parameter adjustment range, a candidate optimized leveling path is generated that adapts to the pore characteristics of the base bed and targets weak geology.

[0009] The engineering feasibility of candidate optimized leveling paths and initial construction paths is evaluated using terrain similarity measurement and geological feature projection algorithm, including: The candidate optimized leveling paths are mapped to the geological feature space, and the topographic differences in the areas covered by the paths of the water bed to be leveled are compared using a 3D point cloud similarity algorithm. The Euclidean distance between the two paths in the porosity space is calculated to determine the consistency of the geological conditions of the water bed to be leveled. If the Euclidean distance of the candidate optimized leveling path in the geological feature space is less than or equal to the threshold, the project is judged to be feasible; if the Euclidean distance is greater than the threshold, it is judged that the geological conditions deviate too much.

[0010] The optimal position of each path point after iterative optimization is output as the optimized leveling path by combining the optimization algorithm, including: A path optimization objective function covering pore characteristics is established, and the path optimization objective function is iteratively solved. The coordinate sequence of the path points is iteratively updated, and the optimized leveling path is obtained in combination with the engineering feasibility judgment.

[0011] The path optimization objective function F covering the pore characteristics is:

[0012] Among them, F is the path optimization objective function value; is the total length of the leveling path, where is the spatial coordinate of the leveling path point at time j, is the spatial coordinate of the leveling path point at time j+1, and r is the leveling path time; represents the standard deviation of the porosity of the ith block of the base bed area to be leveled, where is the average porosity index of the ith block base area in the area to be leveled, M is the number of blocks in the area to be leveled; E i is the block coverage efficiency, , where A c,i is the coverage area of ​​the ith block bed area, A z,i is the total area of ​​the ith block bedding area; 、 、 is the weight coefficient and satisfies .

[0013] The path optimization objective function is iteratively solved, and the coordinate sequence of the path points is iteratively updated including: Initialize the velocity and position of each particle at each path point in the leveling path, and calculate the fitness value of each particle at each path point. The sequence of path points is:

[0014] Where, is the spatial coordinate of the kth path point, and m is the number of path points.

[0015] The fabric operation is as follows: After the leveling machine completes its dive, use GPS positioning and inclination sensors to adjust the level of the equipment, move the material, and use a total station to position the equipment at high levels to meet the design elevation; adjust the walking chassis legs, and after the equipment is leveled, move the transverse and longitudinal trolleys to the starting point of the material laying, and lay the material according to the laying path of the predetermined block base bed area.

[0016] The machine transfer operation is as follows: After the leveling and spreading operation of the block base bed area is completed, start the equipment, move the transverse trolley and the longitudinal trolley to the center of the equipment, lift the legs of the walking chassis, retract the legs of the fabric chassis, lift the travel cylinder, and move the propeller to complete the forward or backward movement of the fabric machine; then, lift the legs of the fabric chassis, retract the legs, and retract the travel cylinder. This operation is repeated in a cycle until the leveling machine moves to the predetermined coordinate position.

[0017] Compared with the prior art, the embodiments of the present invention achieve the following beneficial effects: The present invention utilizes a multi-source sensing and control system and a multi-beam bathymetry method to acquire data on the subgrade of the water area to be leveled. By acquiring reliable environmental data and forming an initial construction blueprint, this provides a basis for optimization and avoids blind construction. Based on the subgrade pore distribution characteristic data and the initial leveling path, the invention combines a geological sensitivity analysis algorithm to identify the operational weaknesses of the leveler in terms of porosity, thereby resolving issues such as instability, subsidence, or low efficiency of the leveler caused by subgrade unevenness during operation. The invention also uses a geological sensitivity analysis algorithm to identify these weaknesses, construct a weak feature set, quantify risk areas, improve the adaptability of the path to geological characteristics, and reduce construction risks. The present invention uses terrain similarity measurement and geological feature projection algorithms to evaluate the engineering feasibility of candidate optimized leveling paths and initial construction paths. By comparing the terrain consistency of candidate paths, geological constraints are evaluated to ensure path safety. An optimization algorithm is used to solve the problem of selecting the best option among multiple candidate paths. By integrating pore feature analysis with an intelligent optimization algorithm, dynamic path generation and online correction are achieved, achieving efficient, uniform, and high-precision autonomous leveling operations. The present invention fully considers the pore characteristics of the base bed. The objective function in the path optimization method covers the pore index characteristics as constraints, fully considers the spatial heterogeneity of the pore distribution in different block base bed areas, and the density after leveling is more uniform. It iterative optimization is performed for each path point to obtain the global optimal position of each path point in the leveling path, avoiding overlapping leveling areas in traditional construction operations, and performing dynamic path adjustment planning in real time based on terrain feedback data. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 1 is a flow chart of a method for optimizing a path of an underwater screed machine based on pore characteristics of a subgrade provided by an embodiment of the present invention; Figure 2 1 is a schematic structural diagram of an underwater leveling machine provided by an embodiment of the present invention; In the figure, the components of the underwater leveling machine include a walking chassis leg 1, a material distribution chassis truss 2, a material distribution chassis main beam 3, a transverse moving trolley 4, a longitudinal moving trolley 5, a material distribution pipe 6, a positioning instrument 7, a measuring tower 8, a walking chassis 9, a material distribution chassis leg 10, and a thruster 11. DETAILED DESCRIPTION

[0019] The present invention will be described in detail below with reference to the accompanying drawings.

[0020] Example 1

[0021] like Figure 1 The present invention provides a method for optimizing the path of an underwater screed machine based on the pore characteristics of the bed, comprising the following steps: Utilize a multi-source sensing and control system and multi-beam bathymetry to measure the three-dimensional topographic data and pore distribution characteristic data of the water bed to be leveled, establish a digital terrain model of the bed to be leveled, and adjust and update it in real time. Utilize a path planning algorithm to output the position of each leveling path point, and based on the position of each leveling path point, obtain the initial leveling path and construct the initial construction path. Based on the pore distribution characteristics of the subgrade and the initial leveling path, a geological sensitivity analysis algorithm is used to identify the operational weaknesses of the leveling machine in terms of porosity. A subgrade weakness feature set is constructed. Based on this subgrade weakness feature set, a geological parameter mapping method is used to make target-maintaining directional adjustments to the initial construction path. A candidate optimized leveling path adapted to the geological characteristics is generated using a fuzzy control method. The engineering feasibility of candidate optimized leveling paths is evaluated using terrain similarity metrics and a geological feature projection algorithm. The optimal position of each path point after iterative optimization is output as the optimized leveling path using the optimization algorithm. Based on the optimized leveling path, the pressure sensor and tilt sensor components in the multi-source perception and control system are used to complete the leveling of the underwater leveling machine. Dynamically adjust and optimize the leveling path parameters based on the leveling machine's operating results on the optimized leveling path, and control the underwater leveling machine to complete the material laying and machine moving operations according to the action instructions; The multi-source sensing and control system is a system that integrates multiple sensors and control technologies and is used to acquire and process multiple data. When measuring the water bed, data is collected through multiple sensors, including but not limited to pressure sensors, tilt sensors, sonar, and underwater cameras, and the control system can adjust the measurement strategy in real time. The multi-beam sounding system simultaneously measures the water depth in multiple directions by emitting multiple acoustic beams. The acquired three-dimensional terrain data and pore distribution characteristic data of the water bed to be leveled are pre-processed. The data space is aligned to a unified coordinate system, and a digital terrain model of the water bed to be leveled is generated through the Kriging interpolation algorithm. Feature vectors are annotated at grid nodes, and multi-source data collection is triggered at intervals to calculate the pore distribution characteristic index of the area to be leveled.

[0022] Where n i V is the porosity of the ith block base bed area in the area to be leveled; p,i V is the pore volume of the ith block of the base bed in the area to be leveled; z,i is the total volume of the ith block bed in the area to be leveled, and M is the number of blocks in the area to be leveled; The path planning algorithm is used to output the position of each leveling path point. The initial leveling path is obtained based on the position of each leveling path point. The construction of the initial construction path includes: Input the grid data of the digital terrain of the base bed to be leveled, use the path planning algorithm to calculate the key point set of the leveling path, output the path point coordinate sequence to obtain the initial leveling path, and use the uniform B-spline curve to connect the path points to construct the initial construction path; The path planning algorithm includes but is not limited to Dijkstra algorithm, heuristic search algorithm, and fast random tree algorithm; Based on the pore distribution characteristic data of the subgrade and the initial leveling path, combined with the geological sensitivity analysis algorithm, the operational weaknesses of the leveling machine in terms of porosity are identified. The subgrade weakness feature set is constructed, including: Perform gradient analysis on the initial leveling path data to calculate the impact of porosity changes in different areas on the leveling quality, and use porosity as the main dimension of the base bed characteristics; Apply simulated disturbances to the bed area, record the leveling quality fluctuation data, quantify the leveling effect deviation under different geological disturbances, and identify areas with unstable quality; Cluster analysis of disturbance sensitivity data was performed to construct a bed weakness feature set containing geological parameters of weak areas based on associated pore types. Specifically, the gradient analysis includes determining the direction in which the gradient needs to be calculated based on the initial leveling path data, calculating the gradient value of each path point coordinate using the difference method, generating a gradient data set, calculating the average value of the gradient to analyze the gradient data, evaluating the severity of the overall gradient change, and obtaining leveling quality data; obtaining porosity data in each block area to form a porosity data set, using linear regression to establish a relationship model between porosity and leveling quality, and quantifying the impact of porosity changes on leveling quality through regression coefficients; analyzing the relationship between porosity and leveling quality in different areas, and identifying areas where porosity changes have a greater impact on leveling quality; Design different types of disturbances, such as vibration, pressure change, temperature change, etc.; adjust disturbance parameters according to actual engineering requirements, determine the intensity and frequency of the disturbance, and apply the designed disturbance to the base bed area; use specialized equipment, such as vibration tables, pressure pumps, etc. to achieve disturbances; during the disturbance process, record the changes in leveling quality in real time; record the leveling quality data after each disturbance to form a fluctuation data set; compare the leveling quality data after disturbance with the data without disturbance, and calculate the deviation value of each area. The deviation value can be calculated using the following formula: Deviation value = leveling quality after disturbance - leveling quality without disturbance. Analyze the deviation data and identify areas with larger deviation values; determine the scope of the unstable area, and based on the size of the deviation value, identify the specific area with unstable leveling quality and determine the disturbance-sensitive area; The disturbance sensitivity data includes parameters such as porosity, disturbance type, and deviation value for each region. A clustering algorithm is selected, including K-means clustering or hierarchical clustering. Clustering parameters are determined, and cluster analysis is performed to cluster the disturbance sensitivity data. The data is divided into different categories, each representing a group of regions with similar sensitivity. Associated pore types are identified, which exhibit similar sensitivities under disturbance. The clustering results are used to understand the response patterns of different pore types to disturbance. Geological parameters related to weak areas, such as porosity, disturbance sensitivity, and deviation value, are extracted from the clustering results. The extracted features are combined into a feature set, and the feature set is used to identify potential weak areas in actual engineering projects.

[0023] Based on the weak feature set of the subgrade, the initial construction path is adjusted in a targeted manner by using the geological parameter mapping method. In combination with the fuzzy control method, candidate optimized leveling paths adapted to the geological characteristics are generated, including: Path control parameters are constructed based on the weak feature set, and a fuzzy control algorithm is established with leveling effect feedback as input and path parameter adjustment as output; For each leveling path, the parameter adjustment range is calculated through fuzzy reasoning based on the porosity compliance rate detected in real time; based on the parameter adjustment range, a candidate optimized leveling path that adapts to the pore characteristics of the subgrade and targets weak geology is generated; Specifically, key geological parameters, such as porosity, disturbance sensitivity, and deviation value, are extracted from the weak feature set; the degree of influence of these parameters on the leveling quality is analyzed to determine which parameters are the main influencing factors; and control variables related to the leveling path are selected based on the key parameters in the weak feature set. The flatness threshold is used to define the maximum allowable flatness deviation; the porosity threshold is used to define the maximum allowable porosity range; and the disturbance threshold is used to define the maximum allowable disturbance sensitivity; a reasonable range is set for each control variable, and all selected control variables are combined into a control parameter set; the control parameter set is expressed as {flatness threshold, porosity threshold, disturbance threshold}; a fuzzy set of leveling effect feedback is defined, such as "good", "general", and "poor"; and a fuzzy set of path parameter adjustment amounts is defined, such as "fine-tuning", "medium-tuning", and "major-tuning"; Fuzzy rules are formulated based on the leveling effect feedback and the weak feature set. As a specific embodiment, if the leveling effect is "good", the path parameter adjustment amount is "fine adjustment"; if the leveling effect is "average", the path parameter adjustment amount is "medium adjustment"; if the leveling effect is "poor", the path parameter adjustment amount is "major adjustment"; the leveling effect feedback detected in real time is fuzzified and mapped to a fuzzy set; the effect feedback includes flatness deviation and porosity compliance rate; according to the fuzzy rule table, the rule corresponding to the current input is matched; the path parameter adjustment amount is calculated through fuzzy reasoning, the fuzzy output is defuzzified, and a specific adjustment value is obtained.

[0024] The porosity of the base bed is detected in real time, and the porosity compliance rate is calculated based on the degree of match between the actual porosity and the target porosity. The porosity compliance rate is fuzzified and mapped to a fuzzy set. According to the fuzzy rule table, the rule corresponding to the current porosity compliance rate is matched; the path parameter adjustment amount is calculated through fuzzy reasoning, and the fuzzy output is defuzzified to obtain the specific adjustment value; the control parameters of the leveling path are adjusted according to the adjustment amount obtained by fuzzy reasoning; the leveling path is redesigned according to the adjusted parameters; and multiple candidate optimized leveling paths are generated, each of which adapts to the porosity characteristics of the base bed and is optimized for weak geology.

[0025] The engineering feasibility of candidate optimized leveling paths and initial construction paths is evaluated using terrain similarity measurement and geological feature projection algorithm, including: The candidate optimized leveling paths are mapped to the geological feature space, and the topographic differences in the areas covered by the paths of the water bed to be leveled are compared using a 3D point cloud similarity algorithm. The Euclidean distance between the two paths in the porosity space is calculated to determine the consistency of the geological conditions of the water bed to be leveled. If the Euclidean distance of the candidate optimized leveling path in the geological feature space is less than or equal to the threshold, the project is determined to be feasible; if the Euclidean distance is greater than the threshold, it is determined that the geological conditions deviate too much; Collect the elevation and coordinate data of the candidate optimized leveling path and the geological characteristic data of the base bed of the water area to be leveled, including porosity, terrain elevation, and disturbance sensitivity; standardize all geological characteristic data to the same dimension, and extract the coordinate and elevation data of key points from the candidate optimized leveling path.

[0026] For each path point, a feature vector is constructed, which contains the porosity, terrain elevation, and disturbance sensitivity characteristics of the point; the feature vector of each path point is mapped into the geological feature space to form a feature point cloud of the path.

[0027] Select a 3D point cloud similarity algorithm, which may include the ICP (Iterative Closest Point) algorithm or a feature matching-based algorithm, but is not specifically limited here. Use the selected algorithm to align the candidate path point cloud with the initial path point cloud and calculate the best match between the point clouds. Calculate the similarity between the aligned point clouds by calculating the root mean square error (RMS) or average distance between the point clouds. Based on the similarity calculation results, evaluate the topographic differences between the areas covered by the two paths. Calculate the Euclidean distance between the two paths in porosity space. Based on the Euclidean distance, evaluate the geological consistency of the two paths. Set a reasonable terrain difference threshold to determine whether the terrain difference is within an acceptable range. Set a reasonable porosity Euclidean distance threshold to determine the geological consistency. If the terrain difference is less than or equal to the terrain difference threshold, the terrain difference is within an acceptable range; if the porosity Euclidean distance is less than or equal to the porosity Euclidean distance threshold, the geological conditions are consistent; if both the terrain difference and the porosity distance are less than or equal to the corresponding thresholds, the candidate optimized leveling path is judged to be highly consistent with the initial construction path in the geological feature space, and the project is feasible; if the terrain difference or the porosity distance is greater than the corresponding threshold, the geological conditions are judged to deviate too much and the project is not feasible.

[0028] The optimal position of each path point after iterative optimization is output as the optimized leveling path by combining the optimization algorithm, including: Establish a path optimization objective function that covers pore characteristics, iteratively solve the path optimization objective function, iteratively update the coordinate sequence of the path points, and combine it with the engineering feasibility judgment to obtain the optimized leveling path; The path optimization objective function F covering the pore characteristics is:

[0029] Among them, F is the path optimization objective function value; is the total length of the leveling path, where is the spatial coordinate of the leveling path point at time j, is the spatial coordinate of the leveling path point at time j+1, and r is the leveling path time; represents the standard deviation of the porosity of the ith block of the base bed area to be leveled, where is the average porosity index of the ith block base area in the area to be leveled, M is the number of blocks in the area to be leveled; E i is the block coverage efficiency, , where A c,i is the coverage area of ​​the ith block bed area, A z,i is the total area of ​​the ith block bedding area; 、 、 is the weight coefficient and satisfies ; The path optimization objective function is iteratively solved, and the coordinate sequence of the path points is iteratively updated including: Initialize the velocity and position of each particle at the leveling path, and calculate the fitness value of each particle at the path point; the sequence of the path points is:

[0030] Where, is the spatial coordinate of the kth path point, and m is the number of path points; Fitness F b Calculation formula:

[0031] in, is the objective function value corresponding to the kth path point at time t+1; the higher the fitness value, the better the particle position The corresponding path is better; Obtain the local optimal position and global optimal position of the path point particle, calculate the velocity of the path point particle at the next moment, and update the position of the path point particle; The speed update formula is:

[0032] in, represents the velocity of particle k at time t+1, represents the velocity of particle k at time t; represents the position of particle k at time t; is the inertia weight, which controls the inertia of the particle to maintain the current state of motion; c1 and c2 are the individual learning factor and the social learning factor, respectively, which are used to control the particle to move to the local optimal position p b and the global optimal position g b The speed of convergence; r1 and r2 are two random variables distributed in the range [0,1]; p b is the local optimal position that the particle has searched so far; g bis the global optimal position searched by the entire particle swarm so far.

[0033] To perform a location update:

[0034] in, represents the position of particle k at time t+1; Combined with the feasibility assessment of the project, the optimal path plan with consistent geological conditions is retained; Dynamically adjust and optimize the leveling path parameters based on the leveling machine's operating results on the optimized leveling path, and control the underwater leveling machine to complete the material laying and machine moving operations according to the action instructions, including: The porosity compliance index is obtained through the leveling machine operation and the parameters are iteratively adjusted; the leveling machine is controlled to perform material laying and machine moving operations according to the adjusted parameters; Specifically, during the operation of the leveling machine, porosity data is continuously collected and the porosity value of each operating point is recorded; a target range of porosity is set, and for each operating point, its porosity is judged whether it is within the target range, and the porosity compliance rate of the entire operating area is calculated: initial leveling path parameters are set, such as material distribution speed, material distribution amount, machine transfer speed, etc.; parameter adjustment rules are formulated according to the porosity compliance rate; as a specific embodiment, if the porosity compliance rate is lower than the target value, the material distribution amount is increased or the material distribution speed is adjusted; the leveling path parameters are dynamically adjusted according to the real-time monitored porosity compliance rate; and the leveling machine is controlled to perform material distribution operations and machine transfer operations according to the adjusted parameters.

[0035] Example 2

[0036] like Figure 2 As shown, the components of the underwater leveling machine of the present invention include a walking chassis leg 1, a material distribution chassis truss 2, a material distribution chassis main beam 3, a transverse trolley 4, a longitudinal trolley 5, a material distribution pipe 6, a positioning instrument 7, a measuring tower 8, a walking chassis 9, a material distribution chassis leg 10, and a propeller 11; the underwater leveling machine includes a power drive device, a leveling working device, a support and frame structure device, a positioning and guidance system, and a multi-source perception and control system. The leveling working device is located at the center of the leveling machine, the power drive device is located at the periphery of the leveling working device, the support and frame structure device connects the power drive device with the leveling working device, the multi-source perception and control system is connected to the power drive device, the leveling device, and the support and frame structure device, and the positioning and guidance system is connected to the multi-source perception and control system.

[0037] Furthermore, the power drive device includes a propeller, a walking chassis leg, and a cloth chassis leg. The walking chassis legs are located at the four corners of the leveling machine, and the cloth chassis legs are located on the inner side of the walking chassis legs. The propeller is connected to the cloth chassis legs and the walking chassis legs.

[0038] Furthermore, the leveling work device includes a material distribution hopper, a transverse moving trolley, and a longitudinal moving trolley. The material distribution hopper is connected to the transverse moving trolley, the transverse moving trolley is connected to the longitudinal moving trolley, and the longitudinal moving trolley is connected to the support and frame structure device.

[0039] Furthermore, the support and frame structure device includes a material distribution chassis truss and a material distribution chassis main beam. The material distribution chassis truss is located on both sides of the periphery of the underwater leveling machine, and the material distribution chassis main beam is connected to the material distribution chassis truss.

[0040] Specifically, the positioning and guiding system includes a positioning instrument and a measuring tower. The positioning instrument is located at the top of the measuring tower. The measuring tower is located at the four corners of the underwater leveling machine and is connected to the material distribution chassis truss.

[0041] Specifically, the multi-source perception and control system includes a pressure sensor, an inclination sensor, a sonar, and an underwater camera. The pressure sensor and the inclination sensor are installed at the center position of the fabric chassis truss in the support and frame structure device and are rigidly connected. The sonar and the underwater camera are installed on the transverse trolley in the leveling work device. The sensor assembly is connected to the power drive device, the leveling work device, and the positioning and guidance system.

[0042] After the leveler has submerged, use GPS positioning and inclination sensors to adjust the level of the equipment, move the material, and use a total station to position the equipment at high levels to meet the designed elevation. Adjust the outriggers of the walking chassis. After the equipment is leveled, move the traverse trolley and the longitudinal trolley to the starting point of the material laying, and lay the material according to the predetermined laying path of the subgrade area. The specific operation of moving the machine is as follows: After the leveling and spreading operation of the block bed area is completed, start the equipment, move the traverse trolley and the longitudinal trolley to the center of the equipment, lift the legs of the walking chassis, retract the legs of the spreading chassis, lift the travel cylinder, and move the propeller to complete the forward or backward movement of the spreading machine; then lift the legs of the spreading chassis, retract the legs of the spreading chassis, and retract the travel cylinder. This cycle is repeated until the leveling machine moves to the predetermined coordinate position. In the description provided herein, numerous specific details are described. However, it is understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0043] Furthermore, those skilled in the art will appreciate that although some embodiments herein include certain features included in other embodiments but not other features, combinations of features from different embodiments are intended to be within the scope of the present invention and to form different embodiments. Any of the claimed embodiments may be used in any combination.

Claims

1. A method for optimizing the path of an underwater screed machine based on the pore characteristics of the bed, characterized in that: The following steps are involved: Utilize a multi-source sensing and control system and multi-beam bathymetry to measure the three-dimensional topographic data and pore distribution characteristic data of the water bed to be leveled, establish a digital terrain model of the bed to be leveled, and adjust and update it in real time. Utilize a path planning algorithm to output the position of each leveling path point, and based on the position of each leveling path point, obtain the initial leveling path and construct the initial construction path. Based on the pore distribution characteristics of the subgrade and the initial leveling path, a geological sensitivity analysis algorithm is used to identify the operational weaknesses of the leveling machine in terms of porosity. A subgrade weakness feature set is constructed. Based on this subgrade weakness feature set, a geological parameter mapping method is used to make target-maintaining directional adjustments to the initial construction path. A candidate optimized leveling path adapted to the geological characteristics is generated using a fuzzy control method. The engineering feasibility of candidate optimized leveling paths is evaluated using terrain similarity metrics and a geological feature projection algorithm. The optimal position of each path point after iterative optimization is output as the optimized leveling path using the optimization algorithm. Based on the optimized leveling path, the pressure sensor and tilt sensor components in the multi-source perception and control system are used to complete the leveling of the underwater leveling machine. The parameters of the optimized leveling path are dynamically adjusted according to the operating effect of the leveling machine on the optimized leveling path, and the laying operation and machine moving operation are completed by controlling the underwater leveling machine according to the action instructions.

2. The underwater screed machine path optimization method based on bed pore characteristics according to claim 1 is characterized in that: The path planning algorithm is used to output the position of each leveling path point. The initial leveling path is obtained based on the position of each leveling path point. The construction of the initial construction path includes: The grid data of the digital terrain to be leveled is input, the key point set of the leveling path is calculated using the path planning algorithm, the coordinate sequence of the path points is output to obtain the initial leveling path, and the initial construction path is constructed by connecting the path points using a uniform B-spline curve.

3. The underwater screed machine path optimization method based on bed pore characteristics according to claim 1 is characterized in that: Based on the pore distribution characteristic data of the subgrade and the initial leveling path, combined with the geological sensitivity analysis algorithm, the operational weaknesses of the leveling machine in terms of porosity are identified. The subgrade weakness feature set is constructed, including: Perform gradient analysis on the initial leveling path data to calculate the impact of porosity changes in different areas on the leveling quality, and use porosity as the main dimension of the base bed characteristics; Apply simulated disturbances to the bed area, record the leveling quality fluctuation data, quantify the leveling effect deviation under different geological disturbances, and identify areas with unstable quality; The disturbance sensitivity data were clustered and analyzed, and a bed weakness feature set containing geological parameters of weak areas was constructed according to the associated pore types.

4. The underwater screed machine path optimization method based on bed pore characteristics according to claim 1 is characterized in that: Based on the weak feature set of the subgrade, the initial construction path is adjusted in a targeted manner by using the geological parameter mapping method. In combination with the fuzzy control method, candidate optimized leveling paths adapted to the geological characteristics are generated, including: Path control parameters are constructed based on the weak feature set, and a fuzzy control algorithm is established with leveling effect feedback as input and path parameter adjustment as output; For each leveling path, the parameter adjustment range is calculated through fuzzy reasoning according to the real-time detected porosity compliance rate; based on the parameter adjustment range, a candidate optimized leveling path is generated that adapts to the pore characteristics of the base bed and targets weak geology.

5. The underwater screed machine path optimization method based on bed pore characteristics according to claim 1 is characterized in that: The engineering feasibility of candidate optimized leveling paths and initial construction paths is evaluated using terrain similarity measurement and geological feature projection algorithm, including: The candidate optimized leveling paths are mapped to the geological feature space, and the topographic differences in the areas covered by the paths of the water bed to be leveled are compared using a 3D point cloud similarity algorithm. The Euclidean distance between the two paths in the porosity space is calculated to determine the consistency of the geological conditions of the water bed to be leveled. If the Euclidean distance of the candidate optimized leveling path in the geological feature space is less than or equal to the threshold, the project is judged to be feasible; if the Euclidean distance is greater than the threshold, it is judged that the geological conditions deviate too much.

6. The underwater screed machine path optimization method based on bed pore characteristics according to claim 1, characterized in that: The optimal position of each path point after iterative optimization is outputted by the optimization algorithm as the optimized leveling path, including: A path optimization objective function covering pore characteristics is established, and the path optimization objective function is iteratively solved. The coordinate sequence of the path points is iteratively updated, and the optimized leveling path is obtained in combination with the engineering feasibility judgment.

7. The underwater screed machine path optimization method based on bed pore characteristics according to claim 6 is characterized in that: The path optimization objective function F covering the pore characteristics is: Among them, F is the path optimization objective function value; is the total length of the leveling path, where is the spatial coordinate of the leveling path point at time j, is the spatial coordinate of the leveling path point at time j+1, and r is the leveling path time; represents the standard deviation of the porosity of the ith block of the base bed area to be leveled, where is the average porosity index of the ith block base area in the area to be leveled, M is the number of blocks in the area to be leveled; E i is the block coverage efficiency, , where A c,i is the coverage area of ​​the ith block bed area, A z,i is the total area of ​​the ith block bedding area; 、 、 is the weight coefficient and satisfies .

8. The underwater screed machine path optimization method based on bed pore characteristics according to claim 6 is characterized in that: The path optimization objective function is iteratively solved, and the coordinate sequence of the path points is iteratively updated including: Initialize the velocity and position of each particle at each path point in the leveling path, and calculate the fitness value of each particle at each path point. The sequence of path points is: Where, is the spatial coordinate of the kth path point, and m is the number of path points.

9. The underwater screed machine path optimization method based on bed pore characteristics according to claim 1, characterized in that: The fabric operation is as follows: After the leveling machine completes its dive, use GPS positioning and inclination sensors to adjust the level of the equipment, move the material, and use a total station to position the equipment at high levels to meet the design elevation; adjust the walking chassis legs, and after the equipment is leveled, move the transverse and longitudinal trolleys to the starting point of the material laying, and lay the material according to the laying path of the predetermined block base bed area.

10. The underwater screed machine path optimization method based on bed pore characteristics according to claim 1, characterized in that: The machine transfer operation is as follows: After the leveling and spreading operation of the block base bed area is completed, start the equipment, move the transverse trolley and the longitudinal trolley to the center of the equipment, lift the legs of the walking chassis, retract the legs of the fabric chassis, lift the travel cylinder, and move the propeller to complete the forward or backward movement of the fabric machine; then, lift the legs of the fabric chassis, retract the legs, and retract the travel cylinder. This operation is repeated in a cycle until the leveling machine moves to the predetermined coordinate position.

Images