Land consolidation planning surveying and mapping collaborative operation method

Abstract

This invention provides a collaborative land consolidation planning and surveying method, belonging to the field of land consolidation and surveying engineering technology. The method includes: establishing a set of planning constraints; calculating the comprehensive weight of planning constraints using a multi-index cross-correction weight calculation method, allocating target sampling densities, and generating an initial collaborative surveying task sequence; executing real-time data acquisition and dynamically generating supplementary sampling task instructions; performing real-time detection of planning-surveying constraint conflicts and initiating a collaborative response mechanism; using a dynamic operation mode switching mechanism to control the collaborative operation of multi-source surveying equipment; and using sampling density deviation as a feedback signal to input an adaptive adjustment algorithm, dynamically adjusting the correction coefficient and feeding it back to the weight calculation step. This invention, employing the above-mentioned collaborative land consolidation planning and surveying method, achieves real-time guidance and dynamic adjustment of surveying operations by planning constraints, solves the problems of separation between planning and surveying and delayed conflict detection, and improves the targeting and efficiency of surveying operations.

Description

Technical Field

[0001] This invention relates to the field of land consolidation and surveying engineering technology, and in particular to a collaborative method for land consolidation planning and surveying. Background Technology

[0002] Land consolidation refers to the remediation of inefficiently used, irrationally used, unused, or damaged land due to production and construction activities and natural disasters, aiming to improve land use efficiency and the ecological environment. In land consolidation projects, the planning and design phase requires accurate surveying data as a foundation, while the surveying work itself needs to determine the key data collection points and accuracy indicators based on the planning and design requirements. Therefore, land consolidation planning and surveying are two highly coupled core components. Currently, land consolidation planning and surveying typically employ a sequential workflow: first, field surveying is conducted to collect basic geographic data; then, data is processed in the office to generate basic maps such as digital line maps and digital elevation models; and finally, planning and design are carried out based on these maps. If, during the planning and design phase, the surveying data is found to be unsuitable for the planning requirements, rework and supplementary surveying are necessary, leading to extended project cycles and increased costs.

[0003] In existing technologies, surveying operations typically collect data across the entire area according to uniform technical standards, lacking consideration for the differentiated needs of different plots within planning and design schemes. For example, areas slated for conversion into new arable land require high-precision topographic data to assess whether the slope meets cultivation requirements, while ecological reserves have even higher requirements for vegetation cover classification accuracy. Existing technologies struggle to translate planning constraints into differentiated collection parameters during the surveying task planning phase, leading to data redundancy in non-critical areas and insufficient accuracy in critical areas. Furthermore, in existing collaborative surveying solutions, data acquired through different methods such as UAVs, unmanned vehicles, and ground surveying are often processed and fused only after collection is complete, failing to dynamically adjust subsequent operational strategies based on the matching of collected data with planning constraints during the operation.

[0004] On the other hand, land consolidation planning and design schemes include various constraints such as basic farmland protection red lines, ecological protection red lines, and construction land control boundaries. These conditions are usually set based on historical surveying data during the planning and design phase. However, historical data may suffer from poor timeliness and low accuracy, leading to spatial conflicts between the planning boundaries and the actual topography and land types. For example, the planned new arable land area may actually be a steep slope or forest land. In existing technologies, such conflicts are often only discovered during the review phase after the planning and design are completed. At this point, the cost of redesigning is extremely high, and the surveying work needs to be reorganized, which seriously affects the project schedule. Therefore, there is an urgent need for a land consolidation planning and surveying collaborative operation method that can use planning constraints as the core driving logic and achieve two-way real-time collaboration between the planning and surveying ends to overcome the shortcomings of existing technologies, such as the separation of planning and surveying, delayed conflict detection, and high rework costs. Summary of the Invention

[0005] The purpose of this invention is to provide a collaborative land consolidation planning and surveying method to solve the technical problems in the prior art, such as the separation of land consolidation planning and surveying, the inability of planning constraints to guide surveying in real time, and the high rework costs caused by spatial conflicts between planning and terrain that can only be discovered afterward.

[0006] To achieve the above objectives, the present invention provides a collaborative land consolidation planning and surveying method, comprising the following steps: Step S1: Establish a set of planning constraints; Step S2: Based on multiple evaluation indicators of each surveying and mapping unit and the cross-correction rules between indicators, calculate the comprehensive weight of planning constraints for each surveying and mapping unit, and allocate the target sampling density of each surveying and mapping unit according to the comprehensive weight of planning constraints to generate the initial collaborative surveying and mapping task sequence. Step S3: Perform real-time data acquisition guided by planning constraints, evaluate the quality of the acquired data in real time, dynamically generate supplementary sampling task instructions for areas where the actual sampling density is lower than the target sampling density, and adjust the operation parameters of the surveying equipment in real time. Step S4: Perform real-time detection of planning-surveying constraint conflicts, analyze spatial conflicts between real-time collected data and planning constraint set, determine conflict handling strategies based on conflict type and preset constraint priority map, and activate the real-time collaborative response mechanism between planning and surveying ends. Step S5: Use a dynamic operation mode switching mechanism to control the collaborative operation of multi-source surveying and mapping equipment; Step S6: During the execution of steps S3 to S5, the deviation between the target sampling density and the actual sampling density of each mapping unit is recorded in real time, and the deviation value is used as a feedback signal to be input into the adaptive adjustment algorithm to adjust the correction coefficient in the cross correction rule between indicators, and the adjusted correction coefficient is fed back to step S2.

[0007] Preferably, step S1 specifically includes: Step S11: Obtain the land consolidation planning and design scheme for the target area; Step S12: Extract spatial constraints and non-spatial constraints from the planning and design scheme; Step S13: Prioritize the extracted constraints and establish a set of planning constraints that includes constraint type, constraint boundary and priority level; Among them, spatial constraints include one or more of the following: basic farmland protection red line, ecological protection red line, urban development boundary, construction land control boundary, and land consolidation project scope boundary; non-spatial constraints include one or more of the following: surveying accuracy level requirements for various types of land parcels, key feature identification requirements, and change detection sensitivity requirements.

[0008] Preferably, the multi-index cross-correction weight calculation method driven by planning constraints in step S2 specifically includes: Step S21: Discretize the target area into multiple mapping units, denoted as the first unit. Each surveying unit is ,in This is the serial number of the surveying unit. , This represents the total number of surveying units; Step S22, for each surveying unit The following four evaluation indicators are set: remember For spatial constraint priority indicators, For surveying accuracy level indicators, For terrain complexity index, As an indicator of ecological sensitivity; Step S23: Define the cross-correction rules between indicators and calculate the comprehensive weight of planning constraints according to the following formula. : ; in, As the attenuation factor, when When it is higher than the first preset threshold, ,otherwise ; As the gain factor, when When it is higher than the second preset threshold, ,otherwise ; The fusion weighting coefficients for the terrain complexity index ; It is a second-order correction factor. , The preset sensitivity coefficient; Step S24: Based on the planning constraints of each surveying unit, integrate the weights. The target sampling density of each mapping unit is allocated according to the preset weight-sampling density mapping function, denoted as... For surveying unit The target sampling density is determined, and an initial collaborative mapping task sequence is generated accordingly.

[0009] Preferably, the weight-sampling density mapping function in step S24 is a piecewise linear function, defined as follows: To set a low threshold for the comprehensive weighting of planning constraints, To set a high threshold for the comprehensive weighting of planning constraints, Minimum sampling density, For the highest sampling density, then: ; in, =10 points / hectare =200 points / hectare and are positive real numbers and .

[0010] Preferably, step S3 specifically includes: Step S31: Start the multi-source mapping equipment to collect data according to the initial collaborative mapping task sequence generated in step S2, and record the data of each mapping unit. Actual sampling density ; Step S32: During the data acquisition process, each surveying unit... Real-time quality assessment of collected data and calculation of coverage metrics. Resolution indicators and geometric accuracy indicators ; Step S33, when At that time, determine the surveying unit This represents an area with insufficient sampling. Step S34: For areas with insufficient sampling, dynamically generate supplementary sampling task instructions, adjust the working trajectory and sampling frequency of the surveying equipment, and ensure that the actual sampling density after supplementary sampling meets the requirements. .

[0011] Preferably, step S4 specifically includes: Step S41: The actual land use type in the real-time collected data... Actual slope Actual ownership boundaries The planning land categories are respectively compared with the planning constraint set established in step S1. Planning slope threshold Planning Boundaries Perform spatial overlay analysis; Step S42, define the following deviation: remember Due to land category deviation, ; remember For slope deviation, ; remember For boundary deviation, ,in It is the minimum Euclidean distance function; set up , , These are tolerances for land type deviation, slope deviation, and boundary deviation, respectively. when or or At that time, it was determined to be a conflict between planning and surveying constraints; Step S43: Determine the conflict handling strategy based on the conflict type and the constraint priority map preset in step S13: If the conflict occurs within the basic farmland protection red line, the survey data will be used as the standard and the planning end will be corrected; if the conflict occurs in the general remediation area, the on-site verification measurement will be triggered before making a judgment. Step S44: Activate the real-time collaborative response mechanism between the planning end and the surveying end, and synchronize the conflict handling results to each surveying terminal in real time.

[0012] Preferably, the dynamic job mode switching mechanism in step S5 specifically includes: Step S51: Define three operation modes: Census mode Detailed investigation mode Verification mode ; Step S52, record For sampling density deviation, ;remember The conflict confidence level is calculated by weighting the degree to which each deviation exceeds the tolerance in step S42, and its value ranges from [0,1]. The operating mode is dynamically selected according to the following rules: like Then switch to census mode. ; like and Then switch to detailed search mode. ; like Then switch to verification mode. ; Step S53: The generation mode switching command is sent to each surveying terminal in real time to adjust the flight altitude, acquisition spacing and scanning speed of the surveying equipment.

[0013] Preferably, the adaptive adjustment algorithm in step S6 is an improved reinforcement learning algorithm, specifically employing a deep Q-network framework, including: Step S61: Construct the state space, and record... For the first The state vector of the step, , of which Step corresponds to surveying unit ; It consists of the following three components: Comprehensive weight of planning constraints for current surveying unit ; recent Average deviation of each completed survey unit ,in For the first Sampling density deviation of the step, The preset sliding window size; Deviation change rate ,in The sampling density deviation of the current step. This represents the sampling density deviation from the previous step; Step S62: Construct the action space and record... For the first The step's action vector is incremented by a decay factor. and gain factor adjustment increment composition; Step S63: Construct the reward function, and record... For the first Reward value per step: ; in, Preset penalty coefficient; Step S64: Update the Q network using the experience replay mechanism.

[0014] Preferably, the improved reinforcement learning algorithm in step S6 also includes - Greedy exploration strategy, memorize For the first The probability of exploration in one step: ; when When <5000, use probability Randomly select actions to explore, with a probability of 1- Select the action with the highest current Q value; when When ≥5000, fixed =0.1.

[0015] Preferably, the correction coefficient dynamically adjusted in step S6 is fed back to step S23 according to the following formula: remember The updated decay factor, This is the updated gain factor; The decay factor before the update. The gain factor before the update; and These are the attenuation factor adjustment increment and gain factor adjustment increment output by the reinforcement learning algorithm in step S62, respectively. , , , The default boundary values ​​are used; the update rule is: ; .

[0016] Therefore, the present invention employs the above-mentioned collaborative land consolidation planning and surveying method, and the beneficial technical effects are as follows: (1) This invention establishes a set of planning constraints and employs a multi-index cross-correction weight calculation method to integrate multiple indicators such as spatial constraint priority, surveying accuracy level, terrain complexity, and ecological sensitivity, dynamically allocating the target sampling density of each surveying unit. This effectively solves the problem of data redundancy and insufficient accuracy caused by the disconnect between planning requirements and surveying objectives in existing technologies. This method enables surveying resources to be precisely directed to key planning areas, while avoiding redundant collection in non-key areas, thus realizing the precise guidance of planning constraints on surveying tasks.

[0017] (2) This invention performs real-time planning-surveying constraint conflict detection during the data collection process and adopts a three-level conflict resolution mechanism, which advances the traditionally discovered design conflicts to the surveying stage for real-time processing, significantly solving the technical problems of delayed conflict detection and high rework costs in the background technology. Through the progressive processing of automatic resolution, semi-automatic resolution and on-site verification resolution, the response time for planning adjustments is greatly reduced, effectively avoiding design rework caused by inconsistencies between planning and terrain, and reducing the implementation risk of land consolidation projects.

[0018] (3) This invention introduces a dynamic operation mode switching mechanism and an improved reinforcement learning algorithm. It automatically switches between three modes—general survey, detailed survey, and verification—based on the real-time coverage rate, sampling density deviation, and conflict confidence of the surveying unit. It also adjusts the correction coefficient through a closed-loop reward signal, solving the problem of lacking real-time collaborative feedback driven by planning constraints in existing technologies. When multiple source devices work together, this method can automatically coordinate paths and acquisition parameters, avoid operational conflicts between devices, and improve overall operation efficiency. At the same time, through adaptive optimization of reinforcement learning, the correction coefficient continuously converges with the operation process, improving the method's generalization ability to different terrains and planning scenarios. Attached Figure Description

[0019] Figure 1 This is a flowchart of a collaborative land consolidation planning and surveying method according to the present invention; Figure 2 This is a schematic diagram illustrating the principle of multi-indicator cross-correction weight calculation. Figure 3 This is a logic diagram of the dynamic operation mode switching mechanism. Detailed Implementation

[0020] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0021] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0022] Example 1 Reference Figures 1-3 This embodiment uses simulation experiments to verify the collaborative land consolidation planning and mapping method provided by the present invention. A virtual test area is constructed in the simulation, with a total area of ​​25 square kilometers, including various land types such as cultivated land, forest land, water areas, and construction land. Spatial constraints such as basic farmland protection red lines, ecological protection red lines, urban development boundaries, and construction land control boundaries are simulated, as well as non-spatial constraints such as mapping accuracy requirements for different plots and key feature identification requirements. The simulation environment generates a digital elevation model based on real terrain data and simulates the data acquisition process of multi-source mapping equipment (UAV oblique photography, LiDAR, unmanned ground surveying vehicle, and handheld surveying terminal).

[0023] Step S1: Establish a set of planning constraints.

[0024] Step S11: Obtain the land consolidation planning and design scheme for the target area.

[0025] Step S12: Extract spatial and non-spatial constraints from the planning and design scheme. In this embodiment, spatial constraints include: 12 plots within the basic farmland protection red line area, with a total area of ​​approximately 8.5 square kilometers; 3 plots within the ecological protection red line area, with a total area of ​​approximately 2.3 square kilometers; 2 urban development boundaries; 5 construction land control boundaries; and 1 land consolidation project boundary. Non-spatial constraints include: the area to be converted into newly added arable land requires a surveying accuracy level of Level 1 (plane mean square error ≤ 0.1 meters, elevation mean square error ≤ 0.2 meters); key feature identification requires the identification of field ridges, ditches, and roads; and change detection sensitivity requires that any land type change exceeding 10 square meters must be marked.

[0026] Step S13: Prioritize the extracted constraints. In this embodiment, the basic farmland protection red line has the highest priority (level 1), followed by the ecological protection red line (level 2), then the construction land control boundary (level 3), and the rest are level 4. Establish a set of planning constraints that includes constraint type, constraint boundary, and priority level. Some planning constraints are shown in Table 1.

[0027] Table 1 Examples of Planning Constraints (Partial)

[0028] Step S2: Based on multiple evaluation indicators of each surveying unit and the cross-correction rules between indicators, calculate the comprehensive weight of planning constraints for each surveying unit, and allocate the target sampling density of each surveying unit according to the comprehensive weight of planning constraints to generate the initial collaborative surveying task sequence.

[0029] Step S21: Discretize the target area into multiple mapping units. In this embodiment, a 200m × 200m grid is used to divide the test area into 625 mapping units, denoted as the nth unit. Each surveying unit is ,in This is the serial number of the surveying unit. .

[0030] Step S22, for each surveying unit The following four evaluation indicators are set: remember As a spatial constraint priority indicator, according to Whether it falls within the red line and the priority level assignment (level 1 is 100, level 2 is 70, level 3 is 40, level 4 is 10). Assign values ​​to the surveying accuracy level indicators according to planning requirements (100 for Level 1, 60 for Level 2, and 30 for Level 3). As an indicator of terrain complexity, the standard deviation of slope is calculated based on the simulated DEM and normalized to 0~100; As an ecological sensitivity indicator, it is normalized to 0~100 based on the distance to the ecological protection red line and vegetation coverage.

[0031] Step S23: Define the cross-correction rules between indicators and calculate the comprehensive weight of planning constraints according to the following formula. : ; in, As the attenuation factor, when The value is higher than the first preset threshold (60 in this embodiment). ,otherwise ; As the gain factor, when When the value is higher than the second preset threshold (50 in this embodiment), ,otherwise ; The fusion weighting coefficients for the terrain complexity index In this embodiment, it is set to 0.3; It is a second-order correction factor. , The sensitivity coefficient is set to 0.05 in this embodiment.

[0032] Step S24: Based on the planning constraints of each surveying unit, integrate the weights. The target sampling density of each mapping unit is allocated according to the preset weight-sampling density mapping function, denoted as... For surveying unit The target sampling density is expressed in points per hectare; and an initial collaborative mapping task sequence is generated based on this density.

[0033] The weight-sampling density mapping function is a piecewise linear function. In this embodiment, it is defined as follows: =200 is the low threshold for the comprehensive weight of planning constraints. =1000 is the high threshold for the comprehensive weight of planning constraints. =10 is the minimum sampling density. =200 is the highest sampling density, then: ; in, and are positive real numbers and .

[0034] Based on this, an initial collaborative mapping task sequence is generated: high-density areas ( >1000) is designated as a detailed investigation sub-region, allocated with UAV oblique photography (flight altitude 80 meters, lateral overlap 70%) and ground motion measurement; medium-density regions (200≤ ≤1000) are designated as census sub-areas, and UAV orthophotos (flying altitude 150 meters) are allocated; low-density areas ( Areas with a value of <200 are classified as general areas, and supplemented using existing data.

[0035] Step S3: Perform real-time data acquisition guided by planning constraints, conduct real-time quality evaluation of the acquired data, dynamically generate supplementary sampling task instructions for areas where the actual sampling density is lower than the target sampling density, and adjust the operation parameters of the surveying equipment in real time.

[0036] Step S31: Start the multi-source mapping equipment to collect data according to the initial collaborative mapping task sequence generated in step S2, and record the data of each mapping unit. Actual sampling density The unit is points per hectare. This embodiment simulates the simultaneous operation of 3 drones, 2 unmanned ground survey vehicles, and 6 groups of manual surveyors.

[0037] Step S32: During the data acquisition process, each surveying unit... Real-time quality assessment of collected data and calculation of coverage metrics. Resolution indicators and geometric accuracy indicators .

[0038] Step S33, when At that time, determine the surveying unit This represents an area with insufficient sampling.

[0039] Step S34: For areas with insufficient sampling, dynamically generate supplementary sampling task instructions, and adjust the operating trajectory and sampling frequency of the surveying equipment. In this embodiment, the system adjusts the UAV's flight altitude from 120 meters to 80 meters, increases the number of flight strips in the area, and doubles the sampling frequency until the actual sampling density after supplementary sampling meets the requirements. .

[0040] Step S4: Perform real-time detection of planning-surveying constraint conflicts, analyze spatial conflicts between real-time collected data and planning constraint set, determine conflict handling strategies based on conflict type and preset constraint priority map, and activate the real-time collaborative response mechanism between planning and surveying ends.

[0041] Step S41: The actual land use type in the real-time collected data... Actual slope Actual ownership boundaries The planning land categories are respectively compared with the planning constraint set established in step S1. Planning slope threshold Planning Boundaries Perform spatial overlay analysis; Step S42, define the following deviation: remember Due to land category deviation, ; remember For slope deviation, ; remember For boundary deviation, ,in It is the minimum Euclidean distance function; set up =1 (a difference of less than 1 in land category coding is acceptable). =2° (slope deviation within 2°) =1 meter represents the tolerance for land type deviation, slope deviation, and boundary deviation, respectively.

[0042] when or or At that time, it was determined to be a planning-surveying constraint conflict.

[0043] In this embodiment, in the simulation unit (The land was planned to be newly added to the farmland.) The actual slope detected was 18°, while the planned slope threshold was 15°. =3°>2°, triggering a conflict.

[0044] Step S43: Determine the conflict handling strategy based on the conflict type and the constraint priority map preset in step S13: If the conflict occurs within the basic farmland protection red line, the survey data will be used as the standard and the planning end will be corrected; if the conflict occurs in the general remediation area, the on-site verification measurement will be triggered before making a judgment.

[0045] In this embodiment, Located within a general remediation area (not within the basic farmland red line), the second-level semi-automatic resolution is adopted: the system pushes the conflict area image and slope analysis map to the planning simulation terminal. After online confirmation by the planners, the unit is removed from the newly added cultivated land area and included in the ecological restoration area. If the conflict occurs within the basic farmland protection red line (for example, a unit whose actual land type is forest land but is planned as cultivated land), the third-level on-site verification resolution is triggered, and a simulated ground survey vehicle is dispatched to conduct on-site verification.

[0046] Step S44: Activate the real-time collaborative response mechanism between the planning and surveying ends, and synchronize the conflict resolution results to each surveying terminal in real time. The adjusted planning boundary is sent to all UAVs and ground surveying vehicles within 30 seconds, and subsequent data collection tasks automatically avoid the adjusted area.

[0047] Step S5: Use a dynamic operation mode switching mechanism to control the collaborative operation of multi-source surveying and mapping equipment.

[0048] The dynamic job mode switching mechanism specifically includes: Step S51: Define three operation modes: Census mode (Data collection interval 50 meters, flight altitude 150 meters), detailed survey mode (Data acquisition interval 10 meters, flight altitude 80 meters), verification mode (Cross-sampling with repeated sampling, spaced 5 meters apart).

[0049] Step S52, record For sampling density deviation, ;remember The conflict confidence level is calculated by weighting the degree to which each deviation exceeds the tolerance in step S42, and its value ranges from [0,1]. The operating mode is dynamically selected according to the following rules: like Then switch to census mode. ; like and Then switch to detailed search mode. ; like Then switch to verification mode. .

[0050] In this embodiment, the simulation unit after conflict resolution , =0.9, the system automatically switches to verification mode, dispatching drones and ground survey vehicles to cross-check the area to ensure the reliability of the adjusted boundary data.

[0051] Step S53: The mode switching command is sent to each surveying terminal in real time to adjust the flight altitude, acquisition spacing, and scanning speed of the surveying equipment. For example, when a unit switches to the verification mode, the UAV's flight altitude is automatically adjusted to 50 meters, and the scanning speed of the ground surveying vehicle is reduced to 0.5 meters per second.

[0052] Step S6: During the execution of steps S3 to S5, the deviation between the target sampling density and the actual sampling density of each mapping unit is recorded in real time, and the deviation value is used as a feedback signal to be input into the adaptive adjustment algorithm to adjust the correction coefficient in the cross correction rule between indicators, and the adjusted correction coefficient is fed back to step S2.

[0053] The adaptive adjustment algorithm is an improved reinforcement learning algorithm, specifically employing a deep Q-network framework, including: Step S61: Construct the state space, and record... For the first The state vector of the step, , of which Step corresponds to surveying unit ; It consists of the following three components: Comprehensive weight of planning constraints for current surveying unit ; recent Average deviation of each completed survey unit ,in For the first Sampling density deviation of the step, The preset sliding window size; Deviation change rate ,in The sampling density deviation of the current step. This represents the sampling density deviation from the previous step.

[0054] Step S62: Construct the action space and record... For the first The step's action vector is incremented by a decay factor. and gain factor adjustment increment Composition. Initial =0.5, =1.5.

[0055] Step S63: Construct the reward function, and record... For the first Reward value per step: ; in, The preset penalty coefficient is set to 0.2 in this embodiment.

[0056] Step S64: Update the Q-network using an experience replay mechanism. Set the learning rate to 0.001 and the discount factor to 0.9. The improved reinforcement learning algorithm also includes... - Greedy exploration strategy, memorize For the first The probability of exploration in one step: ; when When <5000, use probability Randomly select actions to explore, with a probability of 1- Select the action with the highest current Q value; when When ≥5000, fixed =0.1.

[0057] The dynamically adjusted correction factor is fed back to step S23 according to the following formula: remember The updated decay factor, This is the updated gain factor; The decay factor before the update. The gain factor before the update; and These are the attenuation factor adjustment increment and gain factor adjustment increment output by the reinforcement learning algorithm in step S62, respectively. , , , The preset boundary values ​​are, where =0.3, =1.0, =1.0, =2.0; Update rule is: ; ; Adjusted correction factor and Used for calculating the comprehensive weight of planning constraints for subsequent mapping units. Specifically, it is used to replace the formula in step S23. and This forms a closed-loop adaptive optimization.

[0058] Following the steps outlined above, the method of this embodiment was run in the simulation environment. After complete acquisition and collaborative processing of all 625 mapping units, the statistical results are shown in Table 2.

[0059] Table 2 Comparison of the effects of the method of the present invention with those of the prior art

[0060] Simulation experiments verified the effectiveness and superiority of the method of this invention: compared with the traditional serial surveying operation mode (i.e., the process of first collecting data at a uniform density, then planning and designing, and then supplementing the survey), the total data collection time of this method is shortened from 12 days as set in the simulation to 7 days. The data accuracy of key areas (within the basic farmland red line and the proposed newly cultivated areas) is improved by about 40%, and the response time for planning adjustments is reduced from 3-5 days in the traditional mode to less than 30 minutes, effectively avoiding design rework caused by inconsistencies between planning and terrain. At the same time, through closed-loop adaptive optimization of reinforcement learning, the correction coefficients continuously converge as the operation progresses, improving the generalization ability of the method to different terrains and planning scenarios.

[0061] It is worth noting that all contents not described in detail in this invention are existing technologies and are well known to those skilled in the art.

[0062] Therefore, this invention adopts the aforementioned collaborative land consolidation planning and surveying method. By establishing a set of planning constraints and using it as the core driving logic, a multi-index cross-correction weight calculation method is used to dynamically allocate the target sampling density of each surveying unit. During the data collection process, the constraint conflicts between planning and surveying are detected in real time and a multi-level resolution mechanism is triggered. At the same time, a reinforcement learning algorithm is introduced to adjust the correction coefficient in a closed loop, thereby achieving two-way real-time collaboration between the planning end and the surveying end. This significantly improves the efficiency of surveying data collection and the accuracy of data in key areas, reduces the response time for planning adjustments, and lowers the project implementation risk.

[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A collaborative surveying and mapping method for land consolidation planning, characterized in that, Includes the following steps: Step S1: Establish a set of planning constraints; Step S2: Based on multiple evaluation indicators of each surveying and mapping unit and the cross-correction rules between indicators, calculate the comprehensive weight of planning constraints for each surveying and mapping unit, and allocate the target sampling density of each surveying and mapping unit according to the comprehensive weight of planning constraints to generate the initial collaborative surveying and mapping task sequence. Step S3: Perform real-time data acquisition guided by planning constraints, evaluate the quality of the acquired data in real time, dynamically generate supplementary sampling task instructions for areas where the actual sampling density is lower than the target sampling density, and adjust the operation parameters of the surveying equipment in real time. Step S4: Perform real-time detection of planning-surveying constraint conflicts, analyze spatial conflicts between real-time collected data and planning constraint set, determine conflict handling strategies based on conflict type and preset constraint priority map, and activate the real-time collaborative response mechanism between planning and surveying ends. Step S5: Use a dynamic operation mode switching mechanism to control the collaborative operation of multi-source surveying and mapping equipment; Step S6: During the execution of steps S3 to S5, the deviation between the target sampling density and the actual sampling density of each mapping unit is recorded in real time, and the deviation value is used as a feedback signal to be input into the adaptive adjustment algorithm to adjust the correction coefficient in the cross correction rule between indicators, and the adjusted correction coefficient is fed back to step S2.

2. The collaborative land consolidation planning and surveying method according to claim 1, characterized in that, Step S1 specifically includes: Step S11: Obtain the land consolidation planning and design scheme for the target area; Step S12: Extract spatial constraints and non-spatial constraints from the planning and design scheme; Step S13: Prioritize the extracted constraints and establish a set of planning constraints that includes constraint type, constraint boundary and priority level; Among them, spatial constraints include one or more of the following: basic farmland protection red line, ecological protection red line, urban development boundary, construction land control boundary, and land consolidation project scope boundary; non-spatial constraints include one or more of the following: surveying accuracy level requirements for various types of land parcels, key feature identification requirements, and change detection sensitivity requirements.

3. The collaborative land consolidation planning and surveying method according to claim 1, characterized in that, The planning constraint-driven multi-index cross-correction weight calculation method in step S2 specifically includes: Step S21: Discretize the target area into multiple mapping units, denoted as the first unit. Each surveying unit is ,in This is the serial number of the surveying unit. , This represents the total number of surveying units; Step S22, for each surveying unit The following four evaluation indicators are set: remember For spatial constraint priority indicators, For surveying accuracy level indicators, For terrain complexity index, As an indicator of ecological sensitivity; Step S23: Define the cross-correction rules between indicators and calculate the comprehensive weight of planning constraints according to the following formula. : ； in, As the attenuation factor, when When it is higher than the first preset threshold, ,otherwise ; As the gain factor, when When it is higher than the second preset threshold, ,otherwise ; The fusion weighting coefficients for the terrain complexity index ; It is a second-order correction factor. , The preset sensitivity coefficient; Step S24: Based on the planning constraints of each surveying unit, integrate the weights. The target sampling density of each mapping unit is allocated according to the preset weight-sampling density mapping function, denoted as... For surveying unit The target sampling density is determined, and an initial collaborative mapping task sequence is generated accordingly.

4. The collaborative land consolidation planning and surveying method according to claim 3, characterized in that, The weight-sampling density mapping function in step S24 is a piecewise linear function, defined as follows: To set a low threshold for the comprehensive weighting of planning constraints, To set a high threshold for the comprehensive weighting of planning constraints, Minimum sampling density, For the highest sampling density, then: ； in, =10 points / hectare =200 points / hectare and are positive real numbers and .

5. The collaborative land consolidation planning and surveying method according to claim 4, characterized in that, Step S3 specifically includes: Step S31: Start the multi-source mapping equipment to collect data according to the initial collaborative mapping task sequence generated in step S2, and record the data of each mapping unit. Actual sampling density ; Step S32: During the data acquisition process, each surveying unit... Real-time quality assessment of collected data and calculation of coverage metrics. Resolution indicators and geometric accuracy indicators ; Step S33, when At that time, determine the surveying unit This represents an area with insufficient sampling. Step S34: For areas with insufficient sampling, dynamically generate supplementary sampling task instructions, adjust the working trajectory and sampling frequency of the surveying equipment, and ensure that the actual sampling density after supplementary sampling meets the requirements. .

6. The collaborative land consolidation planning and surveying method according to claim 5, characterized in that, Step S4 specifically includes: Step S41: The actual land use type in the real-time collected data... Actual slope Actual ownership boundaries The planning land categories are respectively compared with the planning constraint set established in step S1. Planning slope threshold Planning Boundaries Perform spatial overlay analysis; Step S42, define the following deviation: remember Due to land category deviation, ; remember For slope deviation, ; remember For boundary deviation, ,in It is the minimum Euclidean distance function; set up , , These are tolerances for land type deviation, slope deviation, and boundary deviation, respectively. when or or At that time, it was determined to be a conflict between planning and surveying constraints; Step S43: Determine the conflict handling strategy based on the conflict type and the constraint priority map preset in step S13: If the conflict occurs within the basic farmland protection red line, the survey data will be used as the standard and the planning end will be corrected; if the conflict occurs in the general remediation area, the on-site verification measurement will be triggered before making a judgment. Step S44: Activate the real-time collaborative response mechanism between the planning end and the surveying end, and synchronize the conflict handling results to each surveying terminal in real time.

7. The collaborative land consolidation planning and surveying method according to claim 6, characterized in that, The dynamic job mode switching mechanism in step S5 specifically includes: Step S51: Define three operation modes: Census mode Detailed investigation mode Verification mode ; Step S52, record For sampling density deviation, ;remember The conflict confidence level is calculated by weighting the degree to which each deviation exceeds the tolerance in step S42, and its value ranges from [0,1]. The operating mode is dynamically selected according to the following rules: like Then switch to census mode. ; like and Then switch to detailed search mode. ; like Then switch to verification mode. ; Step S53: The generation mode switching command is sent to each surveying terminal in real time to adjust the flight altitude, acquisition spacing and scanning speed of the surveying equipment.

8. The collaborative land consolidation planning and surveying method according to claim 7, characterized in that, The adaptive adjustment algorithm in step S6 is an improved reinforcement learning algorithm, specifically employing a deep Q-network framework, including: Step S61: Construct the state space, and record... For the first The state vector of the step, , of which Step corresponds to surveying unit ; It consists of the following three components: Comprehensive weight of planning constraints for current surveying unit ; recent Average deviation of each completed survey unit ,in For the first Sampling density deviation of the step, The preset sliding window size; Deviation change rate ,in The sampling density deviation of the current step. This represents the sampling density deviation from the previous step; Step S62: Construct the action space and record... For the first The step's action vector is incremented by a decay factor. and gain factor adjustment increment composition; Step S63: Construct the reward function, and record... For the first Reward value per step: ； in, Preset penalty coefficient; Step S64: Update the Q network using the experience replay mechanism.

9. The collaborative land consolidation planning and surveying method according to claim 8, characterized in that, The improved reinforcement learning algorithm in step S6 also includes - Greedy exploration strategy, memorize For the first The probability of exploration in one step: ； when When <5000, use probability Randomly select actions to explore, with a probability of 1- Select the action with the highest current Q value; when When ≥5000, fixed =0.

1.

10. The collaborative land consolidation planning and surveying method according to claim 1, characterized in that, The dynamically adjusted correction coefficient in step S6 is fed back to step S23 according to the following formula: remember The updated decay factor, This is the updated gain factor; The decay factor before the update. The gain factor before the update; and These are the attenuation factor adjustment increment and gain factor adjustment increment output by the reinforcement learning algorithm in step S62, respectively. , , , The preset boundary values ​​are used; the update rule is: ； 。

Images