City skyline real scene visual touch interaction optimization method based on multi-scene simulation

Abstract

The invention provides an urban skyline live-action visual touch interaction optimization method based on multi-scene simulation. The method comprises the following steps: acquiring three-dimensional live-action data of an urban target area; uploading the three-dimensional live-action data to a digital twinborn visual platform, and reconstructing a skyline lightweight live-action model of the urban target district; recording and translating behavior and instruction information of a user for adjusting the skyline model one by one by using a wearable device, and summarizing the behavior and instruction information into preference training data; defining each building monomer in the skyline model as an agent, defining the action and reward function of the agent, and solving the skyline optimization problem of the skyline model by adopting multi-agent deep reinforcement learning; judging whether the loss function and the reward function meet requirements or not; and storing the output scheme, outputting the output scheme to a holographic sand table for operation and display, and performing secondary accurate adjustment. According to the invention, the behavior data of the skyline model can be adjusted based on the user, and the city skyline optimization scheme better meeting the user demand can be quickly generated.

Description

Technical Field

[0001] The present invention relates to the field of urban planning, and in particular to a method for optimizing real-scene visual touch interaction of a city skyline based on multi-scenario simulation. Background Art

[0002] On the one hand, the city skyline is an important window to show the city's style, and its planning and design is also a key task in urban design. In the current urban design work, the optimization design of the city skyline is mainly completed by planners through computer screen observation and manual adjustment of the three-dimensional model of urban buildings. In this process, there are problems such as single spatial cognition, large gap between the skyline display effect and the actual perception, and time-consuming and laborious adjustment process. When planners optimize different skyline schemes, there are problems such as different judgment standards. On the other hand, with the development of artificial intelligence technology, the working methods and presentation forms of urban design are also constantly changing. At present, reinforcement learning, as a representative artificial intelligence technology, has been applied in urban design work and has shown work capabilities that exceed those of human planners. However, reinforcement learning technology is currently used more in two-dimensional urban design work, lacks applications for three-dimensional space, and lacks integration with other artificial intelligence technologies such as multi-agents.

[0003] The method of real-scene visual touch interactive optimization of urban skyline based on multi-scenario simulation is of great significance. The method solves many problems existing in traditional urban skyline optimization work, such as single observation perspective, poor observation effect, low adjustment efficiency, and different judgment standards, and realizes automatic optimization of urban skyline based on multi-scenario and efficient and accurate adjustment of urban skyline plans by planners. Summary of the invention

[0004] In order to address the deficiencies mentioned in the above background technology, the purpose of the present invention is to provide a real-scene visual touch interactive optimization method for a city skyline based on multi-scenario simulation, which automatically generates and displays a city skyline optimization plan by recording and summarizing the user's behavior and instructions for adjusting the city skyline model and using a multi-agent deep reinforcement learning method.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] The method for optimizing the real-scene visual touch interaction of a city skyline based on multi-scenario simulation includes the following steps:

[0007] Step (1): collect oblique photography data of buildings in the urban target area through an unmanned aerial vehicle equipped with a multi-eye oblique camera and a GPS positioning module, collect three-dimensional morphological data of buildings in the urban target area through a lidar scanner, and obtain three-dimensional real scene data of the urban target area through data registration and fusion.

[0008] Step (2): Upload the real-life data to the digital twin visualization platform, reconstruct the three-dimensional model of the urban target area, perform monomer processing on the three-dimensional model of the urban target area based on the convolutional neural network, and use the digital Li Sheng fusion rendering engine to reconstruct a lightweight real-life model of the urban target area skyline.

[0009] Step (3): The lightweight real-life model of the urban target area skyline is further divided into near-ground buildings, landmark high-rise buildings, and matrix buildings according to the geometric position, density, normal characteristics, and position of the point cloud clusters from the skyline model viewpoint; the mixed reality glasses equipped with an eye movement detection module and the handheld touch controller equipped with an inertial sensor are connected to the digital twin visualization platform to record the user's touch behavior and command information for adjusting the skyline model; the user's behavior and command information are translated one by one and summarized as the monitoring data of the device during adjustment, the parameter changes of the model before and after adjustment, and the characteristic parameters of the model after adjustment.

[0010] Step (4): Multi-agent deep reinforcement learning (RIAL) is used to solve the skyline optimization problem of the lightweight real-life model of the urban target area skyline. According to the monitoring data of the equipment during adjustment, each building unit in the near-ground building, landmark high-rise building and matrix building is defined as an agent, and the action and reward function of the agent are defined.

[0011] Step (5): During the training iteration process, record the changes in the loss function and judge whether the loss function converges to below 0.1 and whether the reward function converges to the optimal solution. When both conditions are met, stop iterating and output the city skyline optimization solution; otherwise, clear the experience replay buffer, restart the interaction between the agent and the environment, and repeat the training process until the termination condition is met.

[0012] Step (6): Save the output plan, output it to a holographic sandbox that stores the lightweight real-life model of the current urban target area skyline, run the display, compare the city skyline optimization plan with the current skyline, and make secondary interactive precise adjustments.

[0013] Furthermore, the step (1) comprises the following steps:

[0014] Oblique photography data of buildings in the urban target area with geographic coordinate information is obtained through drones equipped with multi-eye oblique cameras and GPS positioning modules, including top images and facade images of buildings; three-dimensional morphological data of buildings in the urban target area with geographic coordinate information is obtained through vehicle-mounted laser radar scanners and handheld laser radar scanners with GPS positioning modules, including point cloud data of building heights and building outlines. The oblique photography image data is aligned and fused with the laser scanning point cloud data to obtain three-dimensional real scene data of the urban target area.

[0015] Furthermore, the step (2) comprises the following steps:

[0016] The three-dimensional real scene data of the urban target area obtained in step (1) is input into the EasyV digital twin visualization platform to reconstruct the three-dimensional model of the urban target area. The three-dimensional model of the urban target area is processed singly through the image recognition algorithm based on the convolutional neural network to generate a lightweight real scene model of the current urban target area skyline that can adjust each building separately, and then rendered and displayed through the EasyTwi n digital twin fusion rendering engine.

[0017] Furthermore, the step (3) comprises the following steps:

[0018] S31. In the EasyV digital twin visualization platform, use the preset target detection algorithm to further divide the building units into near-ground buildings, landmark high-rise buildings, and matrix buildings, record their building heights, facade areas, and base area values, and link them to the corresponding attribute tables;

[0019] The preset target detection algorithm performs division based on the geometric position, density, normal features of the building point cloud clusters and their position from the skyline model viewpoint.

[0020] S32, connect the MR mixed reality glasses equipped with an eye movement detection module and the handheld touch controller equipped with an inertial sensor to the EasyV digital twin visualization platform, and record the user's touch adjustment behavior and command information of the skyline model. The optional adjustment settings of the command information include: "remove", "add", "increase height", "reduce height", and "indent the facade along the vertical line of the building" of near-ground buildings, landmark high-rise buildings, and matrix buildings;

[0021] S33. After the user completes the touch adjustment of the skyline model of the target urban area, the platform translates the user behavior and instruction information one by one, and summarizes it into the monitoring data of the device during the adjustment, the parameter changes of the model before and after the adjustment, and the characteristic parameters of the model after the adjustment, which specifically include:

[0022]

[0023]

[0024] Furthermore, the step (4) comprises the following steps:

[0025] S41. Multi-agent deep reinforcement learning (RIAL) is used to solve the optimization problem of the lightweight real-life model of the skyline. Each building unit in the near-ground building, landmark high-rise building and matrix building is defined as an agent. At each time t, the local state O of the observed environment system is the characteristic parameters of other agents within the 1km buffer zone around the geometric center of gravity of the agent, and an action in the action set A is sent to the environment. The actions include: "remove", "add", "increase height", "reduce height", and "indent the facade along the vertical line of the building" of the near-ground building, landmark high-rise building and matrix building;

[0026] S42, in a cloud computing platform with a computing power of 10 TFLOPS, import the user operation training data recorded in step (3), fit the reward function to evaluate the current state and the touch adjustment result of the skyline model;

[0027] The reward function is defined by integrating the parameter changes of the model before and after adjustment, and the similarity between the parameters of the adjusted model and the user training data. The specific formula is:

[0028] R=λ·MinR1+μ·MaxR2

[0029] Among them, MinR1 minimizes the parameter changes of the model before and after adjustment, and MaxR2 maximizes the similarity between the parameters of the adjusted model and the user training data; λ and μ are adjustment parameters used to balance the impact of changes in the two parameters, so that the reward function can better reflect the effect of model adjustment.

[0030] Furthermore, the step (5) comprises the following steps:

[0031] S51. During the training iteration process, monitor the change of the loss function during learning. When the loss function converges to below 0.1 and the reward function converges to the optimal solution, stop the iteration and proceed to the next step of judgment. The loss function adopts the policy gradient method to maximize the cumulative reward. The specific formula is:

[0032]

[0033] Among them, L(θ) is the loss function of the model, trajectory τ is the state sequence experienced by the agent when performing a series of actions in the environment, and R(τ) is the cumulative reward of the agent on trajectory τ, which represents the total reward obtained by the agent after performing a series of actions. represents the gradient of the parameter θ, and π(a|s;θ) is the action probability distribution output by the agent's policy network, which represents the probability of selecting each possible action a under a given state s. The optimal decision-making strategy is achieved by guiding the agent to make appropriate action choices under different states.

[0034] S52: If the termination condition is not met, the experience replay buffer is cleared, the interaction between the agent and the environment is restarted, new experience data is collected, and the policy network parameters of the agent are updated. The training process is executed repeatedly until the termination condition is met.

[0035] Furthermore, the step (6) comprises the following steps:

[0036] The city skyline scenario optimization scheme that meets the termination condition in step (5) is saved, and the three-dimensional building model of the optimization scheme is output to the holographic sand table storing the lightweight real-life model of the current city target area skyline, and the three-dimensional models of the current city skyline and the city skyline optimization scheme are displayed respectively, and the characteristic parameters of the models of the current city and the optimization scheme are displayed in a table. The user compares the current city skyline and the city skyline optimization scheme through MR mixed reality glasses, and makes secondary precise adjustments to the city skyline scenario optimization scheme model through a handheld touch device.

[0037] Beneficial effects of the present invention:

[0038] 1. The present invention uses a multi-agent deep reinforcement learning method to solve the skyline optimization problem based on the user operation data set, and can quickly and automatically generate an urban skyline optimization plan that meets user needs, thereby realizing the automation and intelligence of skyline plan optimization, reducing the workload of planners manually adjusting skyline plans, and improving the work and efficiency of urban design.

[0039] 2. The present invention records and translates the user's behavior and instruction data for adjusting the skyline model one by one, combines the parameter changes of the model before and after the adjustment and the characteristic parameters of the model after the adjustment, and converts the random user behavior into accurate characteristic parameters, thereby avoiding the problems of inconsistent judgment standards and high arbitrariness in traditional skyline optimization work, and improving the scientificity and accuracy of skyline optimization work.

[0040] 3. The present invention displays the real-life model of the city skyline to the user through MR mixed reality glasses, which can more accurately restore the visual effect of the skyline optimization plan. The user can adjust the optimization plan through a handheld touch device in all-round real-time perception, and view the comparison of characteristic parameters of the skyline model before and after the plan adjustment in real time, which improves the immediacy of the plan adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a flow chart of the method of the present invention;

[0042] Figure 2 It is a schematic diagram of touch interaction of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] A real-life visual touch interaction optimization method for city skylines based on multi-scenario simulation, such as Figure 1-2 As shown, the following steps are included:

[0045] 1. Obtain oblique photography data of buildings in the urban target area with geographic coordinate information through drones equipped with multi-eye oblique cameras and GPS positioning modules, and obtain three-dimensional morphological data of buildings in the urban target area with geographic coordinate information through vehicle-mounted laser radar scanners and handheld laser radar scanners with GPS positioning modules. Oblique photography image data is registered and fused with laser scanning point cloud data to obtain three-dimensional real scene data of the urban target area.

[0046] The building oblique photography data includes a top image and a facade image of the building.

[0047] The three-dimensional shape data of the building includes point cloud data of the building height and the building outline.

[0048] 2. Input the 3D real-life data of the urban target area obtained in step 1 into the EasyV digital twin visualization platform to reconstruct the 3D model of the urban target area. The 3D model of the urban target area is processed singly through the image recognition algorithm based on the convolutional neural network to generate a lightweight real-life model of the current urban target area skyline that can adjust each building separately. The model is then rendered and displayed through the EasyTwi n digital twin fusion rendering engine.

[0049] 3. In the EasyV digital twin visualization platform, use the preset target detection algorithm to further divide the building units into near-ground buildings, landmark high-rise buildings and matrix buildings, record their building heights, facade areas, and base area values ​​and link them to the corresponding attribute tables. Connect the MR mixed reality glasses equipped with an eye movement detection module and the handheld touch controller equipped with an inertial sensor to the EasyV digital twin visualization platform to record the user's touch adjustment of the skyline model and command information. When the user completes the touch adjustment of the skyline model of the target area of ​​the city, the platform translates the user's behavior and command information one by one, and summarizes it into the monitoring data of the equipment during the adjustment, the parameter changes of the model before and after the adjustment, and the characteristic parameters of the adjusted model.

[0050] The optional adjustment settings of the command information include: "remove", "add", "increase height", "reduce height", "indent facade along the vertical line of the building" of the near-ground building, the landmark high-rise building and the matrix building.

[0051] The preset target detection algorithm performs division based on the geometric position, density, normal features of the building point cloud clusters and their position from the skyline model viewpoint.

[0052] Fourth, multi-agent deep reinforcement learning (RIAL) is used to solve the optimization problem of the skyline lightweight real-life model. Each building unit in the near-ground building, landmark high-rise building and matrix building is defined as an agent. At each time t, the local state O of the observed environment system is the characteristic parameters of other agents within the 1km buffer range around the geometric center of gravity of the agent, and an action in the action set A is sent to the environment. In a cloud computing platform with a computing power of 10TFLOPS, the user operation training data recorded in step 3 is imported, and the reward function is fitted to evaluate the current state and the touch adjustment results of the skyline model.

[0053] The actions include: "removal", "addition", "height increase", "height reduction", and "facade indentation along the vertical line of the building" of near-ground buildings, landmark high-rise buildings, and matrix buildings;

[0054] The reward function is defined by integrating the parameter changes of the model before and after adjustment, and the similarity between the parameters of the adjusted model and the user training data. The specific formula is:

[0055] R=λ·MinR1+μ·MaxR2

[0056] Among them, MinR1 minimizes the parameter changes of the model before and after adjustment, and MaxR2 maximizes the similarity between the parameters of the adjusted model and the user training data; λ and μ are adjustment parameters used to balance the impact of changes in the two parameters, so that the reward function can better reflect the effect of model adjustment.

[0057] 5. During the training iteration process, monitor the changes in the loss function during learning. When the loss function converges to below 0.1 and the reward function converges to the optimal solution, stop the iteration and proceed to the next step of judgment; if the termination condition is not met, clear the experience playback buffer, restart the interaction between the agent and the environment, collect new experience data, and update the policy network parameters of the agent. The training process is repeated until the termination condition is met.

[0058] The loss function uses a policy gradient method to maximize the cumulative reward. The specific formula is:

[0059]

[0060] Among them, L(θ) is the loss function of the model, trajectory τ is the state sequence experienced by the agent when performing a series of actions in the environment, and R(τ) is the cumulative reward of the agent on trajectory τ, which represents the total reward obtained by the agent after performing a series of actions. represents the gradient of the parameter θ, and π(a|s;θ) is the action probability distribution output by the agent's policy network, which represents the probability of selecting each possible action a under a given state s. The optimal decision-making strategy is achieved by guiding the agent to make appropriate action choices under different states.

[0061] 6. Save the city skyline scenario optimization plan that meets the termination conditions in step 5, and output the 3D building model of the optimization plan to the holographic sand table that stores the lightweight real-life model of the current city target area skyline, respectively display the 3D models of the current city skyline and the city skyline optimization plan, and display the characteristic parameters of the model of the current city and the optimization plan in a table. The user compares the current city skyline and the city skyline optimization plan through MR mixed reality glasses, and makes secondary precise adjustments to the city skyline scenario optimization plan model through a handheld touch device.

[0062] Example

[0063] The technical solution of the present invention will be described in detail below by taking a certain urban area as an example.

[0064] (1) Urban 3D real-scene oblique photography and scanning data collection, including:

[0065] (1.1) Obtain oblique photography data of buildings in the urban target area with geographic coordinate information through an unmanned aerial vehicle equipped with a multi-eye oblique camera and a GPS positioning module, including top images and facade images of the buildings.

[0066] (1.2) Obtain the three-dimensional shape data of buildings in the urban target area with geographic coordinate information through a vehicle-mounted laser radar scanner and a handheld laser radar scanner with a GPS positioning module, including point cloud data of building height and building outline.

[0067] (1.3) Align and fuse the oblique photography image data with the laser scanning point cloud data to obtain the three-dimensional real-life data of the urban target area.

[0068] (2) Generation of lightweight real-life model of city skyline, including:

[0069] (2.1) Input the three-dimensional real scene data of the urban target area obtained in step (1) into the EasyV digital twin visualization platform to reconstruct the three-dimensional model of the urban target area;

[0070] (2.2) The three-dimensional model of the urban target area is processed individually through the image recognition algorithm based on the convolutional neural network, and a lightweight real-life model of the current urban target area skyline is generated, in which each building can be adjusted separately. The model is then rendered and displayed through the EasyTwi n digital twin fusion rendering engine.

[0071] (3) User touch adjustment and preference training data output, including:

[0072] (3.1) In the EasyV digital twin visualization platform, a preset target detection algorithm is used to further divide building units into near-ground buildings, landmark high-rise buildings, and matrix buildings, and their building heights, facade areas, and base area values ​​are recorded and linked to the corresponding attribute table; the preset target detection algorithm is divided based on the geometric position, density, normal characteristics of the building unit point cloud clusters and their position from the skyline model viewpoint.

[0073] (3.2) Connect the MR mixed reality glasses equipped with an eye movement detection module and the handheld touch controller equipped with an inertial sensor to the EasyV digital twin visualization platform to record the user's touch-controlled behavior and command information for adjusting the skyline model. The optional adjustment settings of the command information include: "remove", "add", "increase height", "reduce height", and "indent the facade along the vertical line of the building" for near-ground buildings, landmark high-rise buildings, and matrix buildings;

[0074] (3.3) After the user completes the touch adjustment of the skyline model of the target urban area, the platform translates the user behavior and instruction information one by one, and summarizes it into the monitoring data of the device during the adjustment, the parameter changes of the model before and after the adjustment, and the characteristic parameters of the model after the adjustment, including:

[0075]

[0076]

[0077] (4) Multi-agent deep reinforcement learning simulates the skyline adjustment process, including:

[0078] (4.1) Multi-agent deep reinforcement learning (RIAL) is used to solve the optimization problem of the skyline lightweight real-life model. Each building unit in the near-ground building, landmark high-rise building and matrix building is defined as an agent. At each time t, the local state O of the observed environment system is the characteristic parameters of other agents within the 1km buffer zone around the geometric center of gravity of the agent, and an action in the action set A is sent to the environment, including: "remove", "add", "increase height", "reduce height" and "indent facade along the vertical line of the building" of the near-ground building, landmark high-rise building and matrix building;

[0079] (4.2) In a cloud computing platform with a computing power of 10 TFLOPS, import the user operation training data recorded in step 3, and fit the reward function to evaluate the current state and the touch adjustment results of the skyline model. It is defined by integrating the parameter changes of the model before and after adjustment, and the similarity between the parameters of the adjusted model and the user training data. The specific formula is:

[0080] R=λ·MinR1+μ·MaxR2

[0081] Among them, MinR1 minimizes the parameter changes of the model before and after adjustment, and MaxR2 maximizes the similarity between the parameters of the adjusted model and the user training data; λ and μ are adjustment parameters used to balance the impact of changes in the two parameters, so that the reward function can better reflect the effect of model adjustment.

[0082] (5) Iterative training of the city skyline optimization solution, including:

[0083] (5.1) During the training iteration process, monitor the changes in the loss function during learning. When the loss function converges to below 0.1 and the reward function converges to the optimal solution, stop the iteration and proceed to the next step of judgment. The loss function uses the policy gradient method to maximize the cumulative reward. The specific formula is:

[0084]

[0085] Among them, L(θ) is the loss function of the model, trajectory τ is the state sequence experienced by the agent when performing a series of actions in the environment, and R(τ) is the cumulative reward of the agent on trajectory τ, which represents the total reward obtained by the agent after performing a series of actions. represents the gradient of the parameter θ, and π(a|s;θ) is the action probability distribution output by the agent's policy network, which represents the probability of selecting each possible action a under a given state s. The optimal decision-making strategy is achieved by guiding the agent to make appropriate action choices under different states.

[0086] (5.2) If the termination condition is not met, clear the experience replay buffer, restart the interaction between the agent and the environment, collect new experience data, and update the policy network parameters of the agent. The training process is repeated until the termination condition is met.

[0087] (6) Output of the three-dimensional model of the city skyline optimization solution, including:

[0088] (6.1) Save the city skyline scenario optimization plan that meets the termination conditions in step 5, and output the three-dimensional building model of the optimization plan to a holographic sandbox that stores a lightweight real-life model of the current city target area skyline, and display the three-dimensional models of the current city skyline and the city skyline optimization plan respectively, and display the characteristic parameters of the city status quo and the optimization plan model in a table.

[0089] (6.2) The user compares the current city skyline and the city skyline optimization plan through MR mixed reality glasses, and uses a handheld touch device to make secondary precise adjustments to the city skyline scenario optimization plan model.

Claims

1. A method for optimizing real-life visual touch interaction of city skylines based on multi-scenario simulation, characterized in that: include: Step (1): using a drone equipped with a multi-eye oblique camera and a GPS positioning module to collect oblique photography data of buildings in the urban target area, using a laser radar scanner to collect three-dimensional morphological data of buildings in the urban target area, and obtaining three-dimensional real scene data of the urban target area through data registration and fusion; Step (2): Upload the real scene data to the digital twin visualization platform, reconstruct the three-dimensional model of the urban target area, perform monomer processing on the three-dimensional model of the urban target area based on the convolutional neural network, and use the digital Li Sheng fusion rendering engine to reconstruct a lightweight real scene model of the urban target area skyline; Step (3): The lightweight real-life model of the urban target area skyline is further divided into near-ground buildings, landmark high-rise buildings, and matrix buildings according to the geometric position, density, normal characteristics, and position of the point cloud clusters from the skyline model viewpoint; the mixed reality glasses equipped with an eye movement detection module and the handheld touch controller equipped with an inertial sensor are connected to the digital twin visualization platform to record the user's touch-adjustment behavior and command information of the skyline model; the user's behavior and command information are translated one by one, and summarized as the monitoring data of the device during adjustment, the parameter changes of the model before and after adjustment, and the characteristic parameters of the model after adjustment; Step (4): Use multi-agent deep reinforcement learning to solve the skyline optimization problem of the lightweight real-life model of the urban target area skyline. Define each building unit in the near-ground building, landmark high-rise building and matrix building as an agent based on the monitoring data of the equipment during adjustment, and define the action and reward function of the agent; Step (5): During the training iteration process, record the changes in the loss function and determine whether the loss function converges to below 0.1 and whether the reward function converges to the optimal solution. When both conditions are met, stop iterating and output the city skyline optimization solution. Otherwise, the experience replay buffer is cleared, the interaction between the agent and the environment is restarted, and the training process is repeated until the termination condition is met; Step (6): Save the output plan, output it to a holographic sandbox that stores the lightweight real-life model of the current urban target area skyline, run the display, compare the city skyline optimization plan with the current skyline, and make secondary interactive precise adjustments.

2. The method for optimizing real-scene visual touch interaction of city skyline based on multi-scenario simulation according to claim 1 is characterized in that: The step (1) comprises the following steps: Oblique photography data of buildings in the urban target area with geographic coordinate information is obtained through unmanned aerial vehicles equipped with multi-eye oblique cameras and GPS positioning modules, including top images and facade images of buildings; three-dimensional morphological data of buildings in the urban target area with geographic coordinate information is obtained through vehicle-mounted lidar scanners and handheld lidar scanners with GPS positioning modules, including point cloud data of building heights and building outlines; the oblique photography image data is aligned and fused with the laser scanning point cloud data to obtain three-dimensional real scene data of the urban target area.

3. The method for optimizing real-scene visual touch interaction of city skyline based on multi-scenario simulation according to claim 2 is characterized in that: The step (2) comprises the following steps: The three-dimensional real scene data of the urban target area obtained in step (1) is input into the EasyV digital twin visualization platform to reconstruct the three-dimensional model of the urban target area. The three-dimensional model of the urban target area is processed singly through the image recognition algorithm based on the convolutional neural network to generate a lightweight real scene model of the current urban target area skyline that can adjust each building separately, and then rendered and displayed through the EasyTwin digital twin fusion rendering engine.

4. The method for optimizing real-scene visual touch interaction of city skyline based on multi-scenario simulation according to claim 3 is characterized in that: The step (3) comprises the following steps: S31. In the EasyV digital twin visualization platform, use the preset target detection algorithm to further divide the building units into near-ground buildings, landmark high-rise buildings and matrix buildings, and record their building heights, facade areas and base area values; The preset target detection algorithm is divided based on the geometric position, density, normal characteristics of the building point cloud clusters and their position from the skyline model viewpoint; S32, connecting the MR mixed reality glasses equipped with an eye movement detection module and the handheld touch controller equipped with an inertial sensor to the EasyV digital twin visualization platform, and recording the user's touch-control behavior and command information for adjusting the skyline model; the optional adjustment settings of the command information include: "remove", "add", "increase height", "reduce height", and "indent facade along the vertical line of the building" of near-ground buildings, landmark high-rise buildings, and matrix buildings; S33. After the user completes the touch adjustment of the skyline model of the target urban area, the platform translates the user behavior and instruction information one by one, and summarizes it into the monitoring data of the device during the adjustment, the parameter changes of the model before and after the adjustment, and the characteristic parameters of the model after the adjustment.

5. The method for optimizing real-scene visual touch interaction of city skyline based on multi-scenario simulation according to claim 4 is characterized in that: The step (4) comprises the following steps: S41. Multi-agent deep reinforcement learning is used to solve the optimization problem of the lightweight real-life model of the skyline. Each building unit among the near-ground buildings, landmark high-rise buildings and matrix buildings is defined as an agent. At each time t, the local state O of the observed environment system is the characteristic parameters of other agents within a 1km buffer zone around the geometric center of gravity of the agent, and an action in the action set A is sent to the environment; the actions include: "remove", "add", "increase height", "reduce height", and "indent facade along the vertical line of the building" of the near-ground buildings, landmark high-rise buildings and matrix buildings; S42, in a cloud computing platform with a computing power of 10 TFLOPS, import the user operation training data recorded in step (3), fit the reward function to evaluate the current state and the touch adjustment result of the skyline model; The reward function is defined by integrating the parameter changes of the model before and after adjustment, and the similarity between the parameters of the adjusted model and the user training data. The specific formula is: R=λ·MinR1+μ·MaxR2 Among them, MinR1 minimizes the parameter changes of the model before and after adjustment, and MarR2 maximizes the similarity between the parameters of the adjusted model and the user training data; λ and μ are adjustment parameters used to balance the impact of changes in the two parameters so that the reward function can better reflect the effect of model adjustment.

6. The method for optimizing real-scene visual touch interaction of city skyline based on multi-scenario simulation according to claim 5 is characterized in that: The step (5) comprises the following steps: S51. During the training iteration process, monitor the change of the loss function during learning. When the loss function converges to below 0.1 and the reward function converges to the optimal solution, stop the iteration and proceed to the next step of judgment. The loss function adopts the policy gradient method to maximize the cumulative reward. The specific formula is: Among them, L(θ) is the loss function of the model, trajectory τ is the state sequence experienced by the agent when performing a series of actions in the environment, and R(τ) is the cumulative reward of the agent on trajectory τ, which represents the total reward obtained by the agent after performing a series of actions. represents the gradient of the parameter θ, π(a|s;θ) is the action probability distribution output by the agent's policy network, which represents the probability of selecting each possible action a under a given state s; the optimal decision-making strategy is achieved by guiding the agent to make appropriate action choices under different states; S52: If the termination condition is not met, clear the experience replay buffer, restart the interaction between the agent and the environment, collect new experience data, and update the policy network parameters of the agent; the training process is executed repeatedly until the termination condition is met.

7. The method for optimizing real-scene visual touch interaction of city skyline based on multi-scenario simulation according to claim 6 is characterized in that: The step (6) comprises the following steps: The city skyline scenario optimization scheme that meets the termination condition in step (5) is saved, and the three-dimensional building model of the optimization scheme is output to a holographic sandbox storing a lightweight real-life model of the current city target area skyline, and the three-dimensional models of the current city skyline and the city skyline optimization scheme are displayed respectively, and the characteristic parameters of the models of the current city and the optimization scheme are displayed in a table; the user compares the current city skyline and the city skyline optimization scheme through MR mixed reality glasses, and performs secondary precise adjustments to the city skyline scenario optimization scheme model through a handheld touch device.

Images