Rural space planning method based on mountain slope

Abstract

The invention belongs to the field of space planning, and particularly relates to a rural space planning method based on mountain slope, which comprises the steps of slope suitability division, signal infrastructure layout optimization and space planning. According to the scheme, the path loss from each point of the mountainous area to the nearest signal base station is calculated through the digital elevation model, the received signal power is evaluated, the signal blind area is identified, the relay signal nodes are deployed in the blind area, the signal capital construction space layout is optimized, and the planning accuracy and operability are remarkably improved; according to the scheme, historical activity tracks of villagers are extracted from position data of terminal equipment by using a convolutional neural network, a path density thermodynamic diagram is constructed, activity rules and space use conditions of the villagers are analyzed, space layout mismatch is identified, structured suggestions are generated, and a digital elevation model and the structured suggestions are used as multi-channel tensors to be input into a generative adversarial network. And generating a village space layout map conforming to activity rules of villagers and future planning requirements.

Description

Technical Field

[0001] The present invention relates to the field of spatial planning, and in particular to a rural spatial planning method based on mountain slopes. Background Art

[0002] Rural spatial planning is not only a technical issue, but also a multi-dimensional strategic issue involving economy and ecology. The rural spatial planning method based on mountain slope refers to a method of planning rural space using deep learning technology.

[0003] Among the existing approximate solutions, for example, CN114239250B, a system and method for national land space planning and design, addresses the technical problem that traditional spatial planning methods generally use remote sensing data observation to establish models, but are difficult to effectively predict changes based on real-time terrain. By using drones to capture cloud images, collecting terrain by dot matrix positioning, and combining sample materials, the correlation matrix method is used to dynamically predict land change trends, achieving a more accurate model construction effect. However, the existing solution focuses on natural land and lacks support for the planning of urban built-up areas such as buildings and roads. In addition, land change trends are a long-term non-dominant factor for spatial planning and design needs. There is a technical problem of over-emphasizing land change trends while not considering dominant factors such as existing man-made buildings and future planning needs.

[0004] In addition, for example, CN116070956B is a method for evaluating the benefits of national land space planning. This scheme addresses the urgent need to evaluate the supply and demand of ecosystem services in regional national land space planning and clarify the spatial characteristics of the supply and demand relationship of ecosystem services. First, based on statistical yearbooks, remote sensing data and ecological models, an ecosystem service spatial planning data set is established. The ecological service value equivalent method is used to quantify supply services (such as water resources and food production). Regional demand is estimated based on national per capita indicators (such as per capita carbon emissions and per capita water consumption). The correction coefficient is used to adjust the ecological service value equivalent. Through supply and demand coupling coordination analysis, based on the coordination degree difference, the technical means of clarifying the planning benefit level are achieved, achieving the technical effect of providing a scientific basis for the "ecological priority" national land space planning. However, there are still technical problems that regional differences are not fully reflected and the surface factors of mountainous areas are not fully considered. Summary of the Invention

[0005] In view of the above situation, in order to overcome the defects of the existing technology, the present invention provides a rural space planning method based on mountain slopes. The existing scheme focuses on natural land and provides insufficient support for the planning of urban built-up areas such as buildings and roads. In addition, the land change trend is a long-term non-dominant factor for spatial planning and design needs. There is a technical problem of excessive emphasis on land change trends without considering dominant factors such as existing artificial buildings and future planning needs. This scheme uses a digital elevation model to calculate the path loss from each point in the mountain area to the nearest signal base station, evaluate the received signal power, identify signal blind spots, deploy relay signal nodes in the blind spots, optimize the spatial layout of signal infrastructure, improve signal coverage, and significantly improve the accuracy and feasibility of planning. Operability: It solves the problems of difficulty in collecting data due to the natural terrain of rural areas, difficulty in collecting data required for training models, and poor applicability and effectiveness of traditional spatial planning methods; in response to the technical problems of insufficient reflection of regional differences and failure to fully consider surface factors in mountainous areas, this solution uses convolutional neural networks to extract villagers' historical activity trajectories from terminal device location data, construct path density heat maps, analyze villagers' activity patterns and space usage, identify spatial layout mismatches, generate structured suggestions, point out the shortcomings of spatial layout, avoid spatial layout mismatches, and use digital elevation models and structured suggestions as multi-channel tensor inputs to generate adversarial networks to generate rural spatial layout maps that conform to villagers' activity patterns and future planning needs.

[0006] The technical solution adopted by the present invention is as follows: The present invention provides a rural space planning method based on mountain slope, which includes the following steps:

[0007] Step S1: Slope suitability classification, specifically, using drone aerial surveys to obtain the slope of rural mountainous areas and obtain a digital elevation model. The digital elevation model contains the current spatial layout of the village and the surface slope. Based on the digital elevation model, the rural mountainous areas are divided into four slope types: flat area, gentle slope area, medium slope area, and steep slope area;

[0008] Step S2: Optimize signal infrastructure layout;

[0009] Step S3: Spatial planning, using generative adversarial networks to generate rural spatial planning.

[0010] Furthermore, in step S2, the optimization of signal infrastructure layout specifically includes the following steps:

[0011] Step S21: Signal blind spot identification, specifically, calculating the terrain undulations between each point in the mountainous area and the nearest signal base station using a digital elevation model, evaluating the path loss, recording the difference between the signal base station's transmit power and the path loss at each point in the mountainous area as the received signal power, and obtaining a mountain signal quality map. Each pixel value in the mountain signal quality map corresponds to a corresponding received signal power. A threshold is set, and areas in the mountain signal quality map below the threshold are recorded as blind spots.

[0012] Step S22: loop iteration to obtain signal infrastructure space planning.

[0013] Furthermore, in step S22, the loop iteration specifically includes the following steps:

[0014] Step S221: Initialize the number of newly added relays to 0;

[0015] Step S222: Relay deployment, used to add relay signal nodes, specifically, extracting position coordinates from the digital elevation model, averaging the position coordinates of all terminal devices connected to the blind spot to obtain a three-dimensional average position, calculating the distance vectors from the three-dimensional average position to all terminal devices connected to the blind spot, and averaging the distance vectors to obtain an average distance vector;

[0016] Step S223: the number of newly added relays increases by 1;

[0017] Step S224: Decompose the three-dimensional average position in the coordinate system to obtain the axial component, and calculate the positions of all newly added relays using the following formula:

[0018]

[0019]

[0020] Wherein, i represents the index of the newly added relay, I represents the current number of newly added relays, m_ix represents the position of the i-th newly added relay in the x-axis direction, m_iy represents the position of the i-th newly added relay in the y-axis direction, dis represents the average distance vector, dis_xy represents the mapping of dis on the xy plane, dis_z represents the mapping of dis on the z plane, temp represents the intermediate variable, PP_x represents the x-axis component of the three-dimensional average position, PP_y represents the y-axis component of the three-dimensional average position, PP_z represents the z-axis component of the three-dimensional average position, and map_z_PP_xy represents the height at the plane coordinate (PP_x, PP_y) in the digital elevation model;

[0021] Allocate all connected terminal devices to the newly added relay closest to them;

[0022] Step S225: Execute step S21 again to update the blind area;

[0023] Step S226: Repeat steps S224 to S226 until the number of newly added relays is stable and no longer changes, and finally output the number of newly added relays and the locations of the newly added relays to obtain the optimized signal infrastructure space planning.

[0024] Furthermore, in step S3, the space planning specifically includes the following steps:

[0025] Step S31: Trajectory extraction, specifically, rasterizing the digital elevation model to obtain a grid matrix composed of grid cells. Based on the rural signal network optimized by signal infrastructure spatial planning, the location coordinates of the terminal devices are used to collect historical activities of villagers. A convolutional neural network is used to extract features of the villagers' historical activities to obtain activity trajectories. The activity trajectories are located on the grid matrix, and each activity trajectory is associated with a target.

[0026] Step S32: Heat map generation, specifically, accumulating all activity trajectories, counting the frequency of each grid cell being passed by an activity trajectory, and constructing a path density heat map;

[0027] Step S33: Marking the current spatial layout of the village, specifically, dividing the village into high-traffic frequency areas, high-stay frequency areas, and inaccessible areas based on the path density heat map;

[0028] Step S34: Identify spatial layout mismatches, specifically, mark grid cells with mismatches and generate structured suggestions. The mismatch refers to the unreasonable status quo in the rural spatial layout;

[0029] Step S35: Using a generative adversarial network to generate a plan for functional infrastructure in the village, the digital elevation model and the structured proposal are used as inputs to the generative adversarial network in the form of a multi-channel tensor. The generative adversarial network includes a generator and a discriminator. The generator is responsible for generating a rural spatial layout map, and the discriminator is responsible for judging whether the generated rural spatial layout map conforms to the structured proposal, so as to ensure the consistency of the generated structure and the structured proposal. The generator's behavior on the grid unit for determining the layout of the functional infrastructure is defined as the action space of the generative adversarial network. The slope difference between the grid unit where the functional infrastructure is to be laid out and the target associated with the activity trajectory of the grid unit is used as a penalty term. The proximal policy optimization algorithm is used to train the generative adversarial network.

[0030] Step S36: Add a new layout and use a generative adversarial network to output a rural spatial layout map.

[0031] The present invention provides a rural space planning method based on mountain slope. The beneficial effects achieved by the above scheme are as follows:

[0032] (1) The existing scheme focuses on natural land and provides insufficient support for the planning of urban built-up areas such as buildings and roads. In addition, land change trends are a long-term non-dominant factor for spatial planning and design needs. There is a technical problem of over-emphasizing land change trends while not considering dominant factors such as existing artificial buildings and future planning needs. This scheme uses a digital elevation model to calculate the path loss from each point in the mountainous area to the nearest signal base station, evaluate the received signal power, identify signal blind spots, deploy relay signal nodes in the blind spots, optimize the spatial layout of signal infrastructure, improve signal coverage, and significantly improve the accuracy and operability of planning. It solves the problem that due to the natural terrain of rural areas, data is difficult to collect, the data required for training models is difficult to collect, and the traditional spatial planning methods are poorly applicable and ineffective.

[0033] (1) In order to address the technical issues of insufficient reflection of regional differences and insufficient consideration of surface factors in mountainous areas, this solution uses a convolutional neural network to extract villagers' historical activity trajectories from terminal device location data, construct a path density heat map, analyze villagers' activity patterns and space usage, identify spatial layout mismatches, generate structured suggestions, point out the shortcomings of spatial layout, avoid spatial layout mismatches, and use the digital elevation model and structured suggestions as multi-channel tensor inputs to generate an adversarial network to generate a rural spatial layout map that conforms to villagers' activity patterns and future planning needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic diagram of a flow chart of a rural space planning method based on mountain slope provided by the present invention;

[0035] Figure 2 is a schematic diagram of step S2;

[0036] Figure 3 is a schematic diagram of step S3.

[0037] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0039] In the description of the present invention, it should be understood that terms such as "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they should not be understood as limiting the present invention.

[0040] Example 1, see Figures 1 to 3 The present invention provides a rural space planning method based on mountain slope, which includes the following steps:

[0041] Step S1: Slope suitability classification, specifically, using drone aerial surveys to obtain the slope of rural mountainous areas and obtain a digital elevation model. The digital elevation model contains the current spatial layout of the village and the surface slope. Based on the digital elevation model, the rural mountainous areas are divided into four slope types: flat area, gentle slope area, medium slope area, and steep slope area;

[0042] Step S2: Optimize signal infrastructure layout;

[0043] Step S3: Spatial planning, using generative adversarial networks to generate rural spatial planning.

[0044] Example 2, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. In step S2, the optimization of signal infrastructure layout specifically includes the following steps:

[0045] Step S21: Signal blind spot identification, specifically, calculating the terrain undulations between each point in the mountainous area and the nearest signal base station using a digital elevation model, evaluating the path loss, recording the difference between the signal base station's transmit power and the path loss at each point in the mountainous area as the received signal power, and obtaining a mountain signal quality map. Each pixel value in the mountain signal quality map corresponds to a corresponding received signal power. A threshold is set, and areas in the mountain signal quality map below the threshold are recorded as blind spots.

[0046] Step S22: loop iteration to obtain signal infrastructure space planning.

[0047] Example 3, see Figures 1 to 2 This embodiment is based on the above embodiment. In step S22, the loop iteration specifically includes the following steps:

[0048] Step S221: Initialize the number of newly added relays to 0;

[0049] Step S222: Relay deployment, used to add relay signal nodes, specifically, extracting position coordinates from the digital elevation model, averaging the position coordinates of all terminal devices connected to the blind spot to obtain a three-dimensional average position, calculating the distance vectors from the three-dimensional average position to all terminal devices connected to the blind spot, and averaging the distance vectors to obtain an average distance vector;

[0050] Step S223: the number of newly added relays increases by 1;

[0051] Step S224: Decompose the three-dimensional average position in the coordinate system to obtain the axial component, and calculate the positions of all newly added relays using the following formula:

[0052]

[0053] Wherein, i represents the index of the newly added relay, I represents the current number of newly added relays, m_ix represents the position of the i-th newly added relay in the x-axis direction, m_iy represents the position of the i-th newly added relay in the y-axis direction, dis represents the average distance vector, dis_xy represents the mapping of dis on the xy plane, dis_z represents the mapping of dis on the z plane, temp represents the intermediate variable, PP_x represents the x-axis component of the three-dimensional average position, PP_y represents the y-axis component of the three-dimensional average position, PP_z represents the z-axis component of the three-dimensional average position, and map_z_PP_xy represents the height at the plane coordinate (PP_x, PP_y) in the digital elevation model;

[0054] Allocate all connected terminal devices to the newly added relay closest to them;

[0055] Step S225: Execute step S21 again to update the blind area;

[0056] Step S226: Repeat steps S224 to S226 until the number of newly added relays is stable and no longer changes, and finally output the number of newly added relays and the locations of the newly added relays to obtain the optimized signal infrastructure space planning.

[0057] Example 4, see Figures 1 to 3 This embodiment is based on the above embodiment. In step S3, the space planning specifically includes the following steps:

[0058] Step S31: Trajectory extraction, specifically, rasterizing the digital elevation model to obtain a grid matrix composed of grid cells. Based on the rural signal network optimized by signal infrastructure spatial planning, the location coordinates of the terminal devices are used to collect historical activities of villagers. A convolutional neural network is used to extract features of the villagers' historical activities to obtain activity trajectories. The activity trajectories are located on the grid matrix, and each activity trajectory is associated with a target.

[0059] Step S32: Heat map generation, specifically, accumulating all activity trajectories, counting the frequency of each grid cell being passed by an activity trajectory, and constructing a path density heat map;

[0060] Step S33: Marking the current spatial layout of the village, specifically, dividing the village into high-traffic frequency areas, high-stay frequency areas, and inaccessible areas based on the path density heat map;

[0061] Step S34: Identify spatial layout mismatches, specifically, mark grid cells with mismatches and generate structured suggestions. The mismatch refers to the unreasonable status quo in the rural spatial layout;

[0062] Step S35: Using a generative adversarial network to generate a plan for functional infrastructure in the village, the digital elevation model and the structured proposal are used as inputs to the generative adversarial network in the form of a multi-channel tensor. The generative adversarial network includes a generator and a discriminator. The generator is responsible for generating a rural spatial layout map, and the discriminator is responsible for judging whether the generated rural spatial layout map conforms to the structured proposal, so as to ensure the consistency of the generated structure and the structured proposal. The generator's behavior on the grid unit for determining the layout of the functional infrastructure is defined as the action space of the generative adversarial network. The slope difference between the grid unit where the functional infrastructure is to be laid out and the target associated with the activity trajectory of the grid unit is used as a penalty term. The proximal policy optimization algorithm is used to train the generative adversarial network.

[0063] Step S36: Add a new layout and use a generative adversarial network to output a rural spatial layout map.

[0064] Example 5, see Figures 1 to 3 This embodiment is based on the above embodiment. In step S222, the terminal device specifically refers to a mobile communication device or positioning device used by villagers that can access a signal base station or relay node, including a mobile phone, a tablet computer, and a vehicle-mounted navigation terminal.

[0065] Example 6, see Figures 1 to 3This embodiment is based on the above embodiment. In step S1, according to the digital elevation model, the slope within 0-8° is defined as a flat area, the slope within 8-15° is defined as a gentle slope area, the slope within 15-25° is defined as a medium slope area, and the slope greater than 25° is defined as a steep slope area. Four slope types.

[0066] Example 7, see Figures 1 to 3 This embodiment is based on the above embodiment. In step S21, the path loss calculation formula is as follows:

[0067]

[0068] Where PL(d) represents the path loss, h obs represents the obstacle height, PL_0 represents the free space path loss at d_0, d_0 represents the reference distance, d represents the straight-line distance between the base station and the terminal device, n represents the path loss exponent, X_σ represents shadow fading, and L_terrain represents the terrain blocking additional loss.

[0069] Example 8, see Figures 1 to 3 This embodiment is based on the above embodiment. In step S31, the targets are squares and rest nodes.

[0070] Example 9, see Figures 1 to 3 This embodiment is based on the above embodiment. In step S34, the spatial layout mismatch is specifically:

[0071] If area A is frequently crossed by more than 50 people, but the area does not contain any roads in the existing road map, it is judged as a high-traffic area without roads, and the output is a structured suggestion of "adding roads to area A".

[0072] Example 10, see Figures 1 to 3 This embodiment is based on the above embodiment. In step S35, the functional infrastructure includes open space, roads, building land, public facilities, farmland, woodland, water bodies, squares and rest nodes.

[0073] Example 11, see Figures 1 to 3 This embodiment is based on the fourth embodiment. In step S35, the action space is shown in the following table:

[0074] Action category number Space usage 0 Lay out open space 1 Laying out roads 2 Laying out building land 3 Deployment of public facilities 4 Laying out farmland 5 Laying out woodland 6 Layout of water bodies 7 Layout of squares and rest nodes

[0075] The action space of the GAN consists of terms that determine the deployment behavior of the functional infrastructure.

[0076] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0077] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

[0078] The present invention and its embodiments are described above. This description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without inventiveness, they shall fall within the scope of protection of the present invention.

Claims

1. A rural spatial planning method based on mountain slope, characterized by: The method comprises the following steps: Step S1: Slope suitability classification, specifically, using drone aerial survey to obtain the slope of the rural mountainous area and obtain a digital elevation model, which includes the current spatial layout of the village and the surface slope; Step S2: Optimize signal infrastructure layout; Step S3: Spatial planning, using generative adversarial networks to generate rural spatial planning.

2. The rural space planning method based on mountain slope according to claim 1, characterized in that: In step S2, the optimization of signal infrastructure layout specifically includes the following steps: Step S21: Signal blind spot identification, specifically, calculating the terrain undulations between each point in the mountainous area and the nearest signal base station using a digital elevation model, evaluating the path loss, recording the difference between the signal base station's transmit power and the path loss at each point in the mountainous area as the received signal power, obtaining a mountainous signal quality map, setting a threshold, and recording areas in the mountainous signal quality map below the threshold as blind spots; Step S22: loop iteration to obtain signal infrastructure space planning.

3. The rural space planning method based on mountain slope according to claim 2, characterized in that: In step S22, the loop iteration specifically includes the following steps: Step S221: Initialize the number of newly added relays to 0; Step S222: Relay deployment, used to add relay signal nodes, specifically, extracting position coordinates from the digital elevation model, averaging the position coordinates of all terminal devices connected to the blind spot to obtain a three-dimensional average position, calculating the distance vectors from the three-dimensional average position to all terminal devices connected to the blind spot, and averaging the distance vectors to obtain an average distance vector; Step S223: the number of newly added relays increases by 1; Step S224: Decompose the three-dimensional average position in the coordinate system to obtain axial components, and calculate the positions of all newly added relays; Allocate all connected terminal devices to the newly added relay closest to them; Step S225: Execute step S21 again to update the blind area; Step S226: Repeat steps S224 to S226 until the number of newly added relays is stable and no longer changes, and finally output the number of newly added relays and the locations of the newly added relays to obtain the optimized signal infrastructure space planning.

4. The rural space planning method based on mountain slope according to claim 3 is characterized by: In step S3, the space planning specifically includes the following steps: Step S31: Trajectory extraction, specifically, rasterizing the digital elevation model to obtain a grid matrix composed of grid cells. Based on the rural signal network optimized by signal infrastructure spatial planning, the location coordinates of the terminal devices are used to collect historical activities of villagers. A convolutional neural network is used to extract features of the villagers' historical activities to obtain activity trajectories. The activity trajectories are located on the grid matrix, and each activity trajectory is associated with a target. Step S32: Heat map generation, specifically, accumulating all activity trajectories, counting the frequency of each grid cell being passed by an activity trajectory, and constructing a path density heat map; Step S33: Marking the current spatial layout of the village, specifically, dividing the village into high-traffic frequency areas, high-stay frequency areas, and inaccessible areas based on the path density heat map; Step S34: Identify spatial layout mismatches, specifically, mark grid cells with mismatches and generate structured suggestions; Step S35: Using a generative adversarial network to generate a plan for functional infrastructure in the village, the digital elevation model and the structured proposal are used as inputs to the generative adversarial network in the form of a multi-channel tensor. The generative adversarial network includes a generator and a discriminator. The generator is responsible for generating a rural spatial layout map, and the discriminator is responsible for determining whether the generated rural spatial layout map conforms to the structured proposal. The generator's behavior on the grid unit for determining the layout of the functional infrastructure is defined as the action space of the generative adversarial network. The slope difference between the grid unit where the functional infrastructure is to be laid out and the target associated with the activity trajectory of the grid unit is used as a penalty term. The proximal policy optimization algorithm is used to train the generative adversarial network. Step S36: Add a new layout and use a generative adversarial network to output a rural spatial layout map.

Images