Spatial surveying and mapping calibration method and system based on synchronous positioning drawing

Abstract

The invention discloses a space surveying and mapping calibration method and system based on synchronous positioning drawing, and relates to the technical field of space surveying and mapping, the space surveying and mapping calibration method based on synchronous positioning drawing comprises the following steps: obtaining environment characteristic data; identifying a space environment and a key landmark; based on a stereoscopic vision technology, obtaining a space environment map according to the identified space environment and the key landmarks; adjusting the motion track of the detection equipment in the space environment by using an optimization algorithm; and calibrating the space environment map. The dynamic change in the space environment is monitored in real time, and the influence of the change on the map is evaluated, so that the environment change can be automatically judged and dealt with, the map is ensured to be always kept in the latest state, the space environment map can be automatically calibrated and updated, the need of manual intervention is reduced, and the working efficiency is improved. And the efficiency and timeliness of map maintenance are improved.

Description

Technical Field

[0001] The present invention relates to the field of space surveying and mapping technology, and in particular to a space surveying and mapping calibration method and system based on synchronous positioning mapping. Background Art

[0002] Simultaneous localization and mapping (SLAM) technology enables devices such as robots, drones, and self-driving cars to simultaneously locate and map in unknown environments. This technology collects data about the surrounding environment through sensors on the device (such as cameras, lidar, etc.), uses this data to estimate its own position and posture, and simultaneously builds or updates a map of its environment. The core challenge of SLAM technology is that the device must navigate and map without knowing the environment map in advance, relying only on data collected in real time. With the advancement of technology, SLAM not only focuses on mapping static environments, but is also increasingly able to cope with challenges in dynamic environments, such as identifying and tracking moving objects. In addition, combined with artificial intelligence technologies such as deep learning, SLAM is becoming more intelligent and adaptive, providing more accurate positioning and more detailed environmental maps, further promoting the development of automation and intelligent technology. In SLAM-based spatial mapping calibration, the algorithm analyzes the data collected by sensors on the device (such as cameras, lidar), and compares this data with pre-set or previously generated maps. If a deviation is found between the actual collected data and the existing map, the SLAM algorithm will automatically adjust and update specific areas or features in the map, thereby improving the accuracy and reliability of the map. This process involves not only the correction of geographic location, but also the identification and updating of dynamic and static objects in the environment.

[0003] In the prior art, it is not convenient to enhance the accuracy of spatial data through comprehensive utilization and processing of multi-source time-series spatial data, it is not convenient for the spatial environment map to more accurately reflect the actual environment, and it is not convenient to monitor the dynamic changes in the spatial environment in real time, and it is impossible to evaluate the impact of these changes on the map, thereby reducing the adaptability to complex dynamic environments. At the same time, it is not convenient to adjust the movement trajectory of the detection equipment in the spatial environment, which not only reduces the navigation efficiency of the detection equipment, but also increases the additional energy consumption and time consumption caused by poor path planning, and thus it is not convenient to automatically calibrate and update the spatial environment map, reducing the efficiency and timeliness of map maintenance.

[0004] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention proposes a spatial surveying and mapping calibration method and system based on synchronous positioning mapping, which solves the problems mentioned in the above background technology, such as the inconvenience of enhancing the accuracy of spatial data through comprehensive utilization and processing of multi-source time-series spatial data, the inconvenience of making the spatial environment map reflect the actual environment more accurately, the inconvenience of real-time monitoring of dynamic changes in the spatial environment, and the inability to evaluate the impact of these changes on the map, thereby reducing the adaptability to complex dynamic environments. At the same time, it is not convenient to adjust the motion trajectory of the detection equipment in the spatial environment, which not only reduces the navigation efficiency of the detection equipment, but also increases the extra energy consumption and time consumption caused by poor path planning, and thus it is not convenient to automatically calibrate and update the spatial environment map, reducing the efficiency and timeliness of map maintenance.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: According to one aspect of the present invention, a spatial surveying and mapping calibration method based on synchronous positioning mapping is provided, and the spatial surveying and mapping calibration method based on synchronous positioning mapping comprises the following steps: S1. Acquire multi-source temporal spatial data related to a preset spatial environment, and pre-process the multi-source temporal spatial data to obtain environmental feature data; S2. Based on the classification algorithm, the obtained environmental feature data is matched with the corresponding environmental feature data in the database to identify the spatial environment and key landmarks; S3, based on stereo vision technology, obtain the spatial environment map according to the identified spatial environment and key landmarks; S4. Based on the acquired spatial environment map, the real-time posture of the detection device is obtained through synchronous positioning mapping technology, and the motion trajectory of the detection device in the spatial environment is adjusted using an optimization algorithm; S5. Monitor the dynamic changes in the space environment in real time, use analysis algorithms to evaluate the impact of the dynamic changes on the space environment map, and determine whether the dynamic changes exceed a preset threshold. If so, calibrate the space environment map.

[0007] Furthermore, the step of acquiring multi-source temporal spatial data related to a preset spatial environment and preprocessing the multi-source temporal spatial data to obtain environmental feature data includes the following steps: S11, obtaining duplicate data, missing values ​​and outliers of multi-source time series spatial data related to a preset spatial environment, and performing denoising, filtering and smoothing on the duplicate data, missing values ​​and outliers; S12, connecting the unprocessed multi-source temporal spatial data in the multi-source temporal spatial data related to the preset spatial environment to generate a new data table, associating different data tables through external key values ​​to generate a complete data table, and obtaining an accurate data set; S13. Extract features from the accurate data set to obtain environmental feature data.

[0008] Furthermore, the method of matching the obtained environmental feature data with the corresponding environmental feature data in the database based on the classification algorithm to identify the spatial environment and key landmarks includes the following steps: S21, setting relevant parameters matching the environmental characteristic data, and generating preliminary environmental characteristic data and a corresponding characteristic trend matrix based on the relevant parameters; S22, calculating the fitness function value of preliminary environmental characteristic data; S23, calculate the group fitness variance of all environmental feature data, if it meets the preset conditions, then mutate and go to step S25, otherwise, continue to go to step S24 for optimization; S24, updating the inertia weight point according to the inertia weight calculation formula, and updating the change trend and position of the environmental feature data according to the change trend update formula and the position update formula respectively; S25, check whether the termination condition is met, if it is met, then end the optimization process, otherwise, return to step S22 to continue to optimize; S26, assigning the found optimal parameter vector to the classification model; S27, inputting the processed environmental feature data into the classification model, and training the classification model to obtain a trained classification model; S28. Use the trained classification model to classify new environmental feature data and identify spatial environments and key landmarks.

[0009] Furthermore, the method of using the trained classification model to classify the new environmental feature data and identify the spatial environment and key landmarks includes the following steps: S281, using the trained classification model to analyze the new environmental feature data, and predict the environmental category probability corresponding to each new environmental feature data; S282, setting a confidence threshold, screening the probability of each environmental category to which each environmental feature data belongs, eliminating the prediction results below the confidence threshold, and determining the category label of each environmental feature data from the environmental category probability output by the classification model according to the classification threshold; S282: The screened and determined category label is used as the final environment classification result, and the calculation process ends.

[0010] Furthermore, the inertia weight calculation formula is: ; in, H Expressed as inertia weight; Hmax , H min Respectively represent the starting value and ending value of the inertia weight; W Represents the current iteration number; W max Expressed as the maximum number of iterations.

[0011] Furthermore, the method of obtaining the real-time posture of the detection device based on the acquired spatial environment map by synchronous positioning mapping technology and adjusting the motion trajectory of the detection device in the spatial environment by using an optimization algorithm includes the following steps: S41, identifying key parameters that affect the positioning and path planning of the detection device in the space environment, and setting the initial search range; S42, presetting an initial value of the evaluation period, and setting the number of iterations as the end condition of the detection task optimization process; S43, by comparing the real-time posture of the detection device with the preset path planning data, updating the key parameters according to the preset optimization method; S44, using statistical methods to evaluate the detection task execution efficiency after each key parameter update; S45, based on the evaluation result of the detection task execution efficiency, calculating the fitness score of the key parameter update in each iteration process; S46, comparing the fitness scores of two consecutive iterations, and selecting the key parameter update that has the greatest impact on the detection task execution efficiency as the current best solution; S47, increasing the initial value of the evaluation cycle, and if the preset upper limit of the number of iterations is reached, terminating the evaluation process; S48. After the optimization process is completed, the matching degree between the motion trajectory of the actual detection device and the preset path planning is analyzed to determine whether there is a significant impact. If there is a significant impact, adjustments are made.

[0012] Furthermore, by comparing the real-time posture of the detection device with the preset path planning data, updating the key parameters according to the preset optimization method includes the following steps: S431, setting the evaluation range and initial parameters of the gesture recognition algorithm; S432, setting an initial simulation value for detecting the deviation between the posture of the device and the preset path, and initializing the simulation position; S433, performing a functional evaluation on each simulated posture, identifying the parameters that have the most significant impact on the path deviation in the initial state, and recording the best parameter combination; S434, simulating the change of the posture of the detection device, and selecting the optimal calibration measure by comparing the effects of different parameter combinations; S435, comparing the effect of the new parameter combination with the best parameter combination in the record, if the new parameter combination is better than the best parameter combination in the record, updating the record, otherwise keeping the status quo; S436: If the maximum number of iterations is reached, the current best parameter combination is output as the final solution, and the process proceeds to step S437; S437: Use the obtained optimal parameter combination in the gesture recognition algorithm as a preset optimization method to update key parameters.

[0013] Furthermore, the use of statistical methods to evaluate the detection task execution efficiency after each key parameter update includes the following steps: S441, initializing the verification round number to the first round, and dividing the acquired real-time posture data set of the detection device into several subsets, in each round, selecting one subset as a test set, and the union of all the remaining subsets as a training set; S442, training the prediction model using the adjusted key parameters on the training set, and testing the prediction performance of the prediction model on the validation set; S443. Calculate the generalization error of the prediction model based on the actual performance of the key parameters on the test set; S444, looping step S441 to step S443 until each subset has been used as a test set; S445. Obtain an average value of the generalization errors calculated in all verification rounds as an overall evaluation result of the detection task execution efficiency after each key parameter update.

[0014] Furthermore, the real-time monitoring of dynamic changes in the space environment, using an analysis algorithm to evaluate the impact of the dynamic changes on the space environment map, and determining whether the dynamic changes exceed a preset threshold, if the threshold is exceeded, calibrating the space environment map includes the following steps: S51. Identify dynamic factors that affect the accuracy of spatial environment maps; S52. Determine a measure of the efficiency of the analysis of the impact of dynamic changes on the spatial environment map, and set efficiency thresholds and corresponding calibration measures; S53. If the dynamic change factor causes the analysis efficiency to be lower than a preset threshold, a preset calibration measure is adopted; S54, establishing an optimization model for updating the spatial environment map, and determining the objective function and constraint conditions of the optimization model; S55. Apply the analysis algorithm to the optimization model to find the best solution that meets the constraints as the best dynamic change, and evaluate the impact of the spatial change on the spatial environment map through the best dynamic change. If the best dynamic change exceeds a preset threshold, calibrate the spatial environment map.

[0015] Furthermore, the analysis algorithm is applied to the optimization model to find the best solution that meets the constraint conditions as the best dynamic change, and the impact of the spatial change on the spatial environment map is evaluated through the best dynamic change. If the best dynamic change exceeds a preset threshold, the spatial environment map is calibrated, including the following steps: S551, processing the dynamic change factors that affect the accuracy of the spatial environment map to obtain the processed dynamic change factors, and making the processed dynamic change factors meet the preset conditions; S552, initializing analysis algorithm parameters and calibration measures for dynamically changing factors that affect the accuracy of the spatial environment map; S553, evaluating the effect of each calibration measure on the spatial environment map through the objective function, and updating the state; S554, optimize, evaluate, and adjust each calibration measure according to preset conditions, and update the status; S555: If the preset number of iterations or effect evaluation threshold is reached, execute step S556; otherwise, return to step S553 to continue optimizing each calibration measure; S556. Determine the best calibration measure to adjust the spatial environment map to reflect dynamic changes.

[0016] According to another aspect of the present invention, a spatial surveying and mapping calibration system based on synchronous positioning mapping is also provided, and the spatial surveying and mapping calibration system based on synchronous positioning mapping includes: A data acquisition module is used to acquire multi-source temporal spatial data related to a preset spatial environment, and pre-process the multi-source temporal spatial data to obtain environmental feature data; A data matching module is used to match the obtained environmental feature data with the corresponding environmental feature data in the database based on the classification algorithm to identify the spatial environment and key landmarks; An environment map building module is used to obtain a spatial environment map according to the identified spatial environment and key landmarks based on stereo vision technology; The data optimization module is used to obtain the real-time posture of the detection device through synchronous positioning and mapping technology based on the acquired spatial environment map, and adjust the motion trajectory of the detection device in the spatial environment using an optimization algorithm; The data evaluation module is used to monitor the dynamic changes in the space environment in real time, use the analysis algorithm to evaluate the impact of the dynamic changes on the space environment map, and determine whether the dynamic changes exceed the preset threshold. If the threshold is exceeded, the space environment map is calibrated; Among them, the data acquisition module is connected with the environment map construction module through the data matching module, and the environment map construction module is connected with the data evaluation module through the data optimization module.

[0017] The beneficial effects of the present invention are: 1. The present invention enhances the accuracy and robustness of spatial data through comprehensive utilization and preprocessing of multi-source time-series spatial data, making environmental feature data richer and more reliable. By real-time monitoring of dynamic changes in the spatial environment and evaluating the impact of the changes on the map, it is possible to automatically judge and respond to environmental changes, ensuring that the map is always kept up to date. The motion trajectory of the detection device in the spatial environment is adjusted using an optimization algorithm, which not only improves the navigation efficiency of the detection device, but also reduces the additional energy consumption and time consumption caused by poor path planning. When the dynamic changes in the environment exceed a preset threshold, the spatial environment map can be automatically calibrated and updated, thereby reducing the need for manual intervention and improving the efficiency and timeliness of map maintenance.

[0018] 2. The present invention uses a class algorithm and optimization process to accurately match environmental feature data with corresponding data in the database, thereby efficiently identifying spatial environments and key landmarks, improving recognition accuracy and processing speed, and dynamically adjusting the optimization strategy using the group fitness variance and inertia weight calculation formula, so that the algorithm can self-adjust according to the distribution of actual data, enhancing the adaptability and flexibility of the model when facing different spatial environmental data. By continuously iterating and optimizing until the termination conditions are met, the accurate calibration and updating of the spatial environment map is ensured, the need for manual intervention is reduced, and the automation level of map updating is improved. By setting confidence thresholds and classification thresholds to screen classification results, the data quality based on which the decision-making process is based is ensured, and the reliability of spatial environment recognition and classification is enhanced.

[0019] 3. The present invention ensures dynamic adjustment and optimization of path planning by monitoring the difference between the posture of the detection device and the preset path planning in real time, and optimizes key parameters based on the difference, thereby improving the flexibility and adaptability of navigation. Statistical methods and fitness scores are used to evaluate the update effect of key parameters, so that the detection task can be performed under the optimal parameter configuration, thereby improving execution efficiency and reducing energy consumption and time costs. Through cyclic verification and calculation of generalization errors, the prediction model can continuously adjust and optimize itself according to actual performance, thereby enhancing the learning and adaptability of the detection device in the face of complex environments.

[0020] 4. The present invention monitors and analyzes the dynamic changes in the spatial environment in real time, promptly identifies factors that may affect the accuracy of the map, and performs necessary calibration to ensure that the map data reflects the latest environmental status. By dynamically adjusting the threshold of analysis efficiency and the calibration strategy, the system can flexibly respond to the actual situation of environmental changes, thereby enhancing the ability to adapt to complex and changing environments. Through iterative optimization and effect evaluation methods, it can continuously learn and adapt to environmental changes to more accurately reflect the actual situation and improve overall performance and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 is a flow chart of a spatial surveying and mapping calibration method based on synchronous positioning mapping according to an embodiment of the present invention; Figure 2 It is a principle block diagram of a spatial surveying and mapping calibration system based on synchronous positioning mapping according to an embodiment of the present invention.

[0023] In the figure: 1. Data acquisition module; 2. Data matching module; 3. Environmental map construction module; 4. Data optimization module; 5. Data evaluation module. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present invention will be described clearly and completely 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.

[0025] In the description of the present invention, unless otherwise specified, the meaning of "plurality" is two or more. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0026] According to an embodiment of the present invention, a spatial mapping calibration method and system based on synchronous positioning mapping are provided.

[0027] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the spatial surveying and mapping calibration method based on synchronous positioning drawing according to an embodiment of the present invention, the spatial surveying and mapping calibration method based on synchronous positioning drawing includes the following steps: S1. Acquire multi-source temporal spatial data related to a preset spatial environment, and pre-process the multi-source temporal spatial data to obtain environmental feature data; Specifically, the multi-source temporal spatial data related to the preset spatial environment includes: 1) Image data: Satellite images: provide high-resolution images of the earth's surface from outer space, which can show urban layout, natural features, etc.

[0028] Aerial imagery: Images taken by aircraft such as drones provide a more detailed view, such as the height and shape of buildings, traffic flow, etc.

[0029] Street View Imagery: Imagery at street level that provides details from a sidewalk perspective, including road conditions, building exteriors, etc.

[0030] 2) Radar data: Synthetic Aperture Radar (SAR): can penetrate clouds, rain and fog to provide information on surface structure and topography.

[0031] LiDAR data: It generates high-precision 3D terrain maps and object outlines by emitting laser pulses to measure the distance between objects and the sensor.

[0032] 3) Location data: Global Positioning System (GPS) data: Provides precise geolocation information and is used to determine the location of a specific object or device.

[0033] 4) Environmental sensor data: Meteorological data: Data such as temperature, humidity, and air pressure collected from weather stations or mobile devices that reflect weather conditions at a specific time and place.

[0034] Air Quality Index (AQI) data: Assessing air pollution levels is important for monitoring the health of the environment.

[0035] 5) Sound data: Urban soundscape: traffic noise, crowd sounds, etc., can be used to assess urban activity levels or specific events.

[0036] 6) Social Media Data: Geo-tagged social media posts: provide instant information about people’s movements and events in a specific location.

[0037] Specifically, multi-source temporal spatial data refers to data about the spatial environment collected from different sources and at different times, usually from different types of sensors and measurement devices, including: Imagery data: Visual images from cameras or satellites.

[0038] Radar data: For example, data acquired by synthetic aperture radar (SAR) or light detection and ranging (LiDAR) to measure the distance to terrain and objects.

[0039] Acoustic Data: Data collected by sonar or other acoustic sensors.

[0040] Location data: such as GPS data, which provides the precise location of a device or object.

[0041] Environmental sensor data: such as temperature, humidity, air pressure, etc. These data can be obtained from various environmental sensors.

[0042] Specifically, environmental feature data is specific information about the environment extracted from multi-source temporal spatial data. This information can represent or describe the key attributes of the spatial environment, including: Geographical features: such as topography, landforms, water bodies, etc.

[0043] Architectural features: including the location, size, shape, etc. of the building.

[0044] Roads and paths: This includes the road network, sidewalks and other paths for travel.

[0045] Dynamic changes: such as the trajectory of moving objects, changes in environmental conditions, etc.

[0046] Static objects: such as trees, road signs, buildings, etc.

[0047] S2. Based on the classification algorithm, the obtained environmental feature data is matched with the corresponding environmental feature data in the database to identify the spatial environment and key landmarks; Specifically, the spatial environment refers to the actual physical space that needs to be mapped, navigated or monitored, including indoor space (such as inside a building), urban environment, natural environment (such as forests, mountains) or any specific geographical area. The characteristics and conditions of the spatial environment are crucial for accurate spatial mapping and positioning.

[0048] Specifically, key landmarks refer to specific objects or locations that can be used as reference points or have important significance in a spatial environment, including: Natural landmarks: such as mountains, lakes, large trees, etc.

[0049] Man-made landmarks: such as tall buildings, bridges, sculptures, iconic buildings, etc.

[0050] Indoor landmarks: such as stairways, elevators, halls and other specific functional areas.

[0051] S3, based on stereo vision technology, obtain the spatial environment map according to the identified spatial environment and key landmarks; Specifically, stereoscopic vision technology is a technology that simulates the principle of human binocular vision and obtains three-dimensional spatial information by analyzing images taken from two or more different angles. This technology uses parallax (the difference in the position of the same object in different images) to calculate the depth information of the object, thereby constructing a three-dimensional model of the object.

[0052] Specifically, a spatial environment map refers to a three-dimensional model that describes the position, size, and shape of all objects in an area. Such a map can provide detailed information about the relative positions and spatial layout of objects.

[0053] Specifically, the process of obtaining a spatial environment map based on stereo vision technology includes: Image capture: Use two or more cameras to simultaneously capture images of the spatial environment from different angles. The placement of these cameras mimics the visual differences of the human eye, providing the basis for stereoscopic vision.

[0054] Feature matching: Analyze and identify common feature points (such as edges, corners, or other significant image features) in these images. This step is done by comparing similar areas in different images to determine which features correspond.

[0055] Disparity calculation: For matched feature points, calculate the difference in their positions in two or more images. This difference (disparity) is inversely proportional to the distance from the object to the camera, so it can be used to infer the depth information of each point.

[0056] 3D reconstruction: Using the depth information of feature points, a 3D model of an object in space can be constructed. This step involves converting 2D image data into 3D spatial coordinates to form a 3D representation of the spatial environment.

[0057] Map construction and optimization: By integrating the identified spatial environment and key landmarks into the 3D model, the accuracy of the map can be further optimized and improved. In this process, various algorithms (such as point cloud fusion, mesh optimization, etc.) can be used to improve the accuracy and authenticity of the map.

[0058] S4. Based on the acquired spatial environment map, the real-time posture of the detection device is obtained through synchronous positioning mapping technology, and the motion trajectory of the detection device in the spatial environment is adjusted using an optimization algorithm; Specifically, synchronous positioning mapping technology refers to a technology that, without knowing the environment clearly, builds an environmental map in real time by detecting data collected by the device itself, while determining the position and posture of the device in the map.

[0059] Specifically, detection equipment refers to devices that can perform space detection and collect environmental information, such as self-driving cars, drones, or robots. These devices are usually equipped with cameras, LiDAR, sonar or other sensors to collect data about the surrounding environment.

[0060] Specifically, real-time posture refers to the position and orientation of the detection device at any given moment, including: Position: The three-dimensional coordinates of a device in space, usually expressed as x, y, and z coordinates.

[0061] Orientation: The direction the device is facing, usually described by three angles: pitch, roll, and yaw.

[0062] Specifically, the process of obtaining the real-time posture of the detection device through the synchronous positioning mapping technology includes: Data collection: The detection device uses its sensors (such as cameras, lidar) to collect data about the surrounding environment.

[0063] Feature extraction: Extract key feature points from the collected data. These feature points are objects, edges, or other significant landmarks in the environment.

[0064] Posture estimation: Use algorithms (such as Kalman filter, particle filter, etc.) to process these feature points and estimate the current posture of the device, including estimating the position and orientation of the device.

[0065] Environmental map construction: At the same time, a three-dimensional map of the environment is constructed based on feature points and their location information.

[0066] Iterative optimization: Continuously update the map and continuously adjust and optimize the device's posture estimation based on new data.

[0067] S5. Monitor the dynamic changes in the space environment in real time, use analysis algorithms to evaluate the impact of the dynamic changes on the space environment map, and determine whether the dynamic changes exceed a preset threshold. If so, calibrate the space environment map.

[0068] It should be explained that, taking unmanned vehicles as an example, in unmanned driving systems, vehicles collect data about the surrounding environment through a variety of sensors (such as radar, camera, GPS, etc.) carried on them. Preprocessing includes filtering out noise, data synchronization and format standardization to make the data suitable for subsequent processing. Classification algorithms in machine learning (such as support vector machines, decision trees, etc.) are used to compare the environmental feature data collected in real time with the features stored in the database to identify the specific spatial environment and key landmarks where the current vehicle is located. Stereoscopic vision technology is used to calculate the depth of field information through two images captured by the dual cameras on the vehicle to generate a three-dimensional map of the surrounding environment. The unmanned vehicle uses SLAM technology to update its position and posture in the three-dimensional map in real time. At the same time, based on the acquired map and current position, optimization algorithms (such as path planning algorithms, bird flock optimization algorithms, etc.) dynamically adjust the vehicle's driving trajectory to avoid obstacles and optimize the driving path. The system continuously monitors dynamic changes in the environment, such as changes in road conditions, new buildings, etc. Once these changes exceed the system's preset threshold, such as a road closure that is no longer applicable to the current path, the system will trigger the map update process and automatically perform calibration measures.

[0069] Specifically, dynamic changes in the spatial environment refer to any changes or movement events that may affect the accuracy of the established spatial environment map, including: Changes in physical structure: such as the demolition of a building or the construction of a new one.

[0070] The presence of moving objects: such as vehicles, people, or other objects that can move.

[0071] Changes in environmental conditions: such as changes in visibility due to weather or changes in road conditions.

[0072] Seasonal changes, such as trees losing their leaves or covering them, can change the appearance of the environment.

[0073] Specifically, the process of calibrating the spatial environment map includes: Change detection: First, changes that have occurred in the environment need to be identified and assessed. This is done by comparing the most recently collected environmental data with existing spatial environmental maps, using image processing or data analysis algorithms to identify differences.

[0074] Impact Assessment: Once changes are detected, analytical algorithms are used to assess the specific impact of these changes on map accuracy. This includes calculating the size and location of the changed area and the potential impact of the change on navigation or other applications.

[0075] Decide on a calibration strategy: Based on the impact assessment of the change, determine the calibration measures that need to be taken. This may involve updating information in specific areas of the map, remapping the affected areas, or adjusting certain parameters of the map.

[0076] Data integration and updating: Collect new environmental data, such as through remapping or using data from other sources, and then integrate this data into existing maps.

[0077] Validate and Iterate: Finally, validate the accuracy of the calibrated map and make further adjustments if necessary.

[0078] Preferably, the step of acquiring multi-source temporal spatial data related to a preset spatial environment and preprocessing the multi-source temporal spatial data to obtain environmental feature data comprises the following steps: S11, obtaining duplicate data, missing values ​​and outliers of multi-source time series spatial data related to a preset spatial environment, and performing denoising, filtering and smoothing on the duplicate data, missing values ​​and outliers; S12, connecting the unprocessed multi-source temporal spatial data in the multi-source temporal spatial data related to the preset spatial environment to generate a new data table, associating different data tables through external key values ​​to generate a complete data table, and obtaining an accurate data set; S13. Extract features from the accurate data set to obtain environmental feature data.

[0079] Preferably, the step of matching the obtained environmental feature data with the corresponding environmental feature data in the database based on the classification algorithm to identify the spatial environment and key landmarks comprises the following steps: S21, setting relevant parameters matching the environmental characteristic data, and generating preliminary environmental characteristic data and a corresponding characteristic trend matrix based on the relevant parameters; Specifically, the relevant parameters refer to a set of preset standards or variables used for environmental feature data matching and feature trend analysis. These parameters define how to identify, match and analyze features in the environment, and how to extract trends from time series data, including: Feature matching accuracy: Determines the acceptable error range in the matching process.

[0080] Time Window: The length of time used for trend analysis, for example, determining how far in the past data is used to analyze changes in environmental characteristics.

[0081] Data sampling frequency: determines how often environmental characteristic data are collected and analyzed.

[0082] Threshold: Sets the sensitivity of changes in a specific feature to identify significant changes or trends.

[0083] Weights: In multi-trait analysis, different traits may be assigned different weights to reflect their contribution to the overall characteristics of the environment.

[0084] Specifically, a characteristic trend refers to the pattern or direction of changes in environmental characteristic data over time.

[0085] Specifically, the feature trend matrix refers to a mathematical model or table that combines time series data with environmental features to show the trend of each feature over time, including: Rows: represent different environmental characteristics.

[0086] Columns: represent different time points or periods.

[0087] Value: represents the state or change of a specific environmental characteristic at a specific point in time or over a period of time.

[0088] S22, calculating the fitness function value of the preliminary environmental characteristic data (particles); S23, calculate the fitness variance of the group (particle group) of all environmental feature data, if it meets the preset conditions, then mutate and go to step S25, otherwise, continue to go to step S24 for optimization; S24, updating the inertia weight point according to the inertia weight calculation formula, and updating the change trend (particle speed) and position of the environmental feature data according to the change trend update formula (i.e., particle speed update formula) and the position update formula (i.e., particle position update formula); S25, check whether the termination condition is met, if it is met, then end the optimization process, otherwise, return to step S22 to continue to optimize; S26, assigning the found optimal parameter vector to the classification model (support vector machine model); S27, inputting the processed environmental feature data into the classification model, and training the classification model to obtain a trained classification model; S28. Use the trained classification model to classify new environmental feature data and identify spatial environments and key landmarks.

[0089] Specifically, the classification algorithm is an adaptive particle swarm optimization support vector machine algorithm, which is an algorithm that combines the particle swarm optimization algorithm and the support vector machine, and is used to optimize the parameters of the support vector machine model. The adaptive particle swarm optimization support vector machine algorithm continuously updates the position and speed of particles and uses the search capability of the particle swarm algorithm to find the optimal combination of support vector machine parameters, thereby improving the performance and generalization ability of the model.

[0090] Preferably, the method of using the trained classification model to classify the new environmental feature data and identify the spatial environment and key landmarks comprises the following steps: S281, using the trained classification model to analyze the new environmental feature data, and predict the environmental category probability corresponding to each new environmental feature data; S282, setting a confidence threshold, screening the probability of each environmental category to which each environmental feature data belongs, eliminating the prediction results below the confidence threshold, and determining the category label of each environmental feature data from the environmental category probability output by the classification model according to the classification threshold; S282: The screened and determined category label is used as the final environment classification result, and the calculation process ends.

[0091] Preferably, the inertia weight calculation formula is: ; in, H Expressed as inertia weight; H max , H min Respectively represent the starting value and ending value of the inertia weight; W Represents the current iteration number; W max Expressed as the maximum number of iterations.

[0092] Preferably, the method of acquiring the real-time posture of the detection device based on the acquired space environment map by synchronous positioning mapping technology, and adjusting the motion trajectory of the detection device in the space environment by using an optimization algorithm includes the following steps: S41, identifying key parameters that affect the positioning and path planning of the detection device in the space environment, and setting the initial search range; Specifically, the key parameters that affect the positioning and path planning of detection equipment in the space environment include: 1) Equipment performance parameters: Sensor accuracy: The accuracy of sensors on the detection device (such as GPS, lidar, and camera) directly affects the accuracy of positioning.

[0093] Sensor coverage: The range of the environment that the sensor can detect, which affects the device's ability to perceive the environment.

[0094] Moving speed: The maximum moving speed and acceleration of the device, which affects its dynamic response and path execution capabilities.

[0095] Turning radius: The ability of a device to turn in space, affecting its ability to navigate in narrow or complex environments.

[0096] 2) Environmental characteristic parameters: Terrain type: such as flat, hilly, mountainous, etc., affects the equipment's traffic capacity and path selection.

[0097] Obstacle distribution: The location and size of obstacles in the environment affect the device's obstacle avoidance capabilities and path planning.

[0098] Space size: The overall size of the environment, such as height limit, width limit, etc., affects the device's path selection and movement strategy.

[0099] 3) Task requirements parameters: Target location: The target location that the detection device needs to reach or approach is the main goal of path planning.

[0100] Time limit: The time limit required to complete the task affects the length and speed of the path.

[0101] Safety requirements: such as the requirement to avoid dangerous areas or maintain a certain safe distance, which affects the safety of the path.

[0102] S42, preset an initial value of the evaluation period (i.e., preset an initial value of the time counter), and set the number of iterations as the end condition of the detection task optimization process; Specifically, a time counter is a counting variable used to track the iteration process in an algorithm or program. In optimization and computational tasks, a time counter is used to monitor the number of iterations that an algorithm has executed and decide when to stop the iteration process.

[0103] Specifically, the termination conditions include that the time recorded by the time counter reaches a preset maximum value and the number of iterations of the optimization algorithm reaches a preset maximum number.

[0104] S43, by comparing the real-time posture of the detection device with the preset path planning data, updating the key parameters according to the preset optimization method; S44, use statistical methods (i.e., k-fold cross validation) to evaluate the detection task execution efficiency after each key parameter update; Specifically, k-fold cross validation is a machine learning model evaluation method, which refers to dividing the key parameter data set that affects the positioning and path planning of the detection device in the spatial environment into k subsets, one of which is used as a validation set and the remaining k-1 subsets are used as training sets.

[0105] S45, based on the evaluation result of the detection task execution efficiency, calculating the fitness score of the key parameter update in each iteration process; S46, comparing the fitness scores of two consecutive iterations, and selecting the key parameter update that has the greatest impact on the detection task execution efficiency as the current best solution; S47, increasing the initial value of the evaluation cycle, and if the preset upper limit of the number of iterations is reached, terminating the evaluation process; S48. After the optimization process is completed, the matching degree between the motion trajectory of the actual detection device and the preset path planning is analyzed to determine whether there is a significant impact. If there is a significant impact, adjustments are made.

[0106] Specifically, the optimization algorithm is an SVM parameter selection method based on an improved bird flock optimization algorithm, which is a strategy for applying a nature-inspired algorithm to machine learning parameter optimization. In this method, an improved bird flock optimization algorithm is used to select or optimize the parameters of the SVM to improve the performance of the model.

[0107] Specifically, SVM is a supervised learning algorithm based on the principle of structural risk minimization. It uses expected risk minimization as the evaluation criterion and aims to obtain a decision function with high generalization ability through finite sample training, that is, to find an optimal hyperplane to completely separate the samples under the condition of linear separability.

[0108] Specifically, the improved bird flock optimization algorithm refers to the improvement and variant of the traditional particle swarm optimization algorithm (Particle Swarm Optimization, PSO). The particle swarm optimization algorithm is an optimization technology based on swarm intelligence, which simulates the social behavior of bird flocks or fish schools to solve optimization problems.

[0109] Specifically, the improved bird flock optimization algorithm includes the following improvements: Improvements in speed and position update rules: In standard PSO, the movement of particles is based on their best historical position and the best position of the entire group. Improved algorithms may use different rules to adjust the speed and position of particles to improve convergence speed and avoid falling into local optimality.

[0110] Parameter adjustment and control: Improved algorithms may include dynamic adjustment mechanisms for PSO parameters such as inertia weight, individual best and group best learning factors to balance exploration (global search) and exploitation (local search) during the search process.

[0111] Customization to adapt to specific problems: For specific types of problems (such as nonlinear problems, multi-objective problems, etc.), the improved bird flock optimization algorithm may be specially designed to solve these problems more effectively.

[0112] Preferably, by comparing the real-time posture of the detection device with the preset path planning data, updating the key parameters according to the preset optimization method includes the following steps: S431, setting the evaluation range and initial parameters of the gesture recognition algorithm; S432, setting an initial simulation value for detecting the deviation between the posture of the device and the preset path, and initializing the simulation position; S433, performing a functional evaluation on each simulated posture, identifying the parameters that have the most significant impact on the path deviation in the initial state, and recording the best parameter combination; S434, simulating the change of the posture of the detection device, and selecting the optimal calibration measure by comparing the effects of different parameter combinations; S435, comparing the effect of the new parameter combination with the best parameter combination in the record, if the new parameter combination is better than the best parameter combination in the record, updating the record, otherwise keeping the status quo; S436: If the maximum number of iterations is reached, the current best parameter combination is output as the final solution, and the process proceeds to step S437; S437: Use the obtained optimal parameter combination in the gesture recognition algorithm as a preset optimization method to update key parameters.

[0113] Specifically, the posture recognition algorithm is a hybrid optimization algorithm of artificial fish school and particle swarm based on chaotic mapping. It utilizes the characteristics of chaotic search that does not repeat traversal search according to a certain rule, transfers the optimization variable space to the chaotic space, and efficiently obtains the optimal satisfactory solution domain, which greatly improves the local optimization ability of the entire algorithm.

[0114] Preferably, the use of statistical methods to evaluate the detection task execution efficiency after each key parameter update includes the following steps: S441, initializing the verification round number to the first round, and dividing the acquired real-time posture data set of the detection device into several subsets, in each round, selecting one subset as a test set, and the union of all the remaining subsets as a training set; S442, training a prediction model (i.e., an SVM model) using the adjusted key parameters on the training set, and testing the prediction performance of the prediction model on the validation set; S443. Calculate the generalization error of the prediction model based on the actual performance of the key parameters on the test set; Specifically, generalization error is an important indicator for evaluating the generalization ability of machine learning models. It reflects the prediction error of the model on new data that has not been seen.

[0115] Specifically, calculating the generalization error of the prediction model (ie, the SVM model) refers to calculating the prediction error of the prediction model (ie, the SVM model) on the test set.

[0116] S444, looping step S441 to step S443 until each subset has been used as a test set; S445. Obtain an average value of the generalization errors calculated in all verification rounds as an overall evaluation result of the detection task execution efficiency after each key parameter update.

[0117] Preferably, the real-time monitoring of dynamic changes in the space environment, using an analysis algorithm to evaluate the impact of the dynamic changes on the space environment map, and determining whether the dynamic changes exceed a preset threshold, if the threshold is exceeded, calibrating the space environment map includes the following steps: S51. Identify dynamic factors that affect the accuracy of spatial environment maps; Specifically, the dynamic factors that affect the accuracy of spatial environment maps include: 1) Environmental changes: Natural disasters: Natural disasters such as earthquakes, floods, and volcanic eruptions can change the landform, making the original spatial environment map no longer accurate.

[0118] Climate change: Long-term climate change, such as melting glaciers and expanding deserts, will also affect the layout and characteristics of the spatial environment.

[0119] 2) Human activities: Urban Development: Building construction, demolition, and urban expansion can change the accuracy of the map.

[0120] Agricultural activities: Activities such as agricultural reclamation and deforestation can also lead to changes in the landscape.

[0121] 3) Technical factors: Measurement errors: In the process of spatial environment mapping, the technologies and tools used (such as satellites and drones) may have measurement errors, which will directly affect the accuracy of the map.

[0122] Data update frequency: If the update frequency of the spatial environment map is not high enough and cannot reflect the latest status of the environment in a timely manner, its accuracy will also be affected.

[0123] 4) Ecological changes: Changes in vegetation cover: Changes in vegetation due to seasonal changes, drought, disease, etc. can also affect the accuracy of the map.

[0124] Animal activity: The movement of certain large animals, such as the migration of elephants, may also affect the terrain and thus affect map accuracy.

[0125] 5) Geographical dynamics: River diversion: The direction and drainage of a river may change due to sedimentation or human intervention.

[0126] Land erosion and sedimentation: Natural processes such as wind erosion and water erosion can also lead to changes in topography.

[0127] S52. Determine a measure of the efficiency of the analysis of the impact of dynamic changes on the spatial environment map, and set efficiency thresholds and corresponding calibration measures; Specifically, the measurement method refers to the specific indicators or methods used to quantify the efficiency of the analysis. When evaluating the dynamic changes of the spatial environment map, it includes: Time efficiency: Evaluates the length of time it takes from collecting data to updating the map. This can be measured by average processing time or the total time required to complete an update cycle.

[0128] Accuracy: The accuracy of the analysis results, that is, the precision of dynamic change identification. It can be measured by comparing the difference between the map before and after the update.

[0129] Data processing capability: Evaluates the system's ability to handle large amounts of data, such as the amount of data processed per second or the number of data streams processed simultaneously.

[0130] Resource consumption: The resources required during the analysis process, such as computing resources, storage resources, and human resources.

[0131] Specifically, the corresponding calibration measures refer to a series of technologies and methods used to adjust and optimize the spatial environment map to reflect the dynamic changes in the real world. These calibration measures aim to ensure the high accuracy of the map and reflect the changes in the environment in real time. They include: 1) Parameter adjustment The calibration process may require adjusting the algorithm's parameters to make it better adapt to changes in environmental data, including adjusting filter settings, updating model weights, or changing threshold settings in the data processing flow.

[0132] 2) Data fusion technology Multi-source data fusion technology is used to integrate data from different sensors (such as cameras, radars, LiDAR, etc.) to improve map coverage and accuracy. Data fusion helps eliminate the bias and uncertainty of a single data source.

[0133] 3) Machine Learning and Artificial Intelligence Apply machine learning and AI techniques to automatically detect and correct errors in maps. For example, use deep learning models to automatically identify new buildings or changed road conditions in images and update the map with this information.

[0134] 4) Change Detection Algorithm Use change detection algorithms to identify differences between map data and the actual environment. These algorithms can automatically mark areas that need to be updated based on methods such as image analysis and time series data comparison.

[0135] 5) Dynamic Optimization Continuously optimize calibration measures based on real-time data and feedback, including adjusting the execution frequency of the algorithm, improving the data processing flow, or retraining the machine learning model to adapt to new environmental characteristics.

[0136] 6) Iterative testing and verification The effectiveness of calibration measures is tested through an iterative approach, including simulation and field testing, to ensure that each updated map accurately reflects the state of the real world. After each calibration, the effectiveness of the calibration needs to be evaluated through an objective function, and the calibration strategy is adjusted based on the evaluation results.

[0137] 7. Final decision and implementation Determine the most effective combination of calibration measures and apply them regularly or continuously during the map update process to ensure that the map remains up to date and effectively reflects the latest changes in the environment.

[0138] S53. If the dynamic change factor causes the analysis efficiency to be lower than a preset threshold, a preset calibration measure is adopted; S54, establishing an optimization model for updating the spatial environment map, and determining the objective function and constraint conditions of the optimization model; Specifically, the objective function is a mathematical expression that quantifies the goal that the optimization model attempts to achieve. In the spatial environment map update, the objective function includes: Accuracy: Ensures that the updated map is as close to reality as possible. Measured by comparing the difference between the updated map and the actual measurement data.

[0139] Timeliness: Quickly respond to environmental changes and ensure the timeliness of maps. This involves minimizing the time from data collection to map updates.

[0140] Cost-effectiveness: Consider cost factors when updating maps, such as reducing the cost of data collection, processing and updating.

[0141] User experience: Optimize user satisfaction with map updates, involving aspects such as update frequency, map usability and interactivity.

[0142] Specifically, constraints are restrictive rules followed during the optimization process, which ensure the feasibility and practical applicability of the optimization model. In the spatial environment map update, constraints include: Data quality: Ensure that the data used for updates meets certain quality standards, such as data resolution, accuracy, and completeness.

[0143] Resource constraints: Consider the limitations of available resources, such as computing power, storage space, and human resources.

[0144] Legal and Policy: Comply with legal and policy requirements related to map updates.

[0145] Technical limitations: Consider the capabilities of existing technologies and tools, such as sensor accuracy and coverage.

[0146] Security: Ensure that the update process does not adversely affect users or the environment.

[0147] Real-time: In a dynamically changing environment, map updates are required to reflect the latest environmental status in a timely manner.

[0148] S55. Apply the analysis algorithm to the optimization model to find the best solution that meets the constraints as the best dynamic change, and evaluate the impact of the spatial change on the spatial environment map through the best dynamic change. If the best dynamic change exceeds a preset threshold, calibrate the spatial environment map.

[0149] Specifically, the analysis algorithm is the fish school algorithm, which is an optimization algorithm based on swarm intelligence. It is inspired by the foraging, clustering and following behaviors of fish in nature. The algorithm imitates the collective behavior of fish in finding food and avoiding danger.

[0150] Preferably, the analysis algorithm is applied to the optimization model to find the best solution that meets the constraint conditions as the best dynamic change, and the impact of the spatial change on the spatial environment map is evaluated through the best dynamic change. If the best dynamic change exceeds a preset threshold, the spatial environment map is calibrated, including the following steps: S551, processing the dynamic change factors that affect the accuracy of the spatial environment map (i.e., whitening processing) to obtain the processed dynamic change factors, and making the processed dynamic change factors meet the preset conditions; Specifically, whitening is a signal processing technique used to eliminate redundant information in data and reduce the correlation between features. In the present invention, the purpose of whitening is to prepare data to meet the input requirements of the analysis algorithm.

[0151] S552, initializing analysis algorithm parameters (i.e., initializing fish school algorithm parameters) and calibration of dynamic change factors affecting the accuracy of the spatial environment map (i.e., artificial fish); S553, evaluating the evaluation value (i.e., negative entropy) of the effect of each calibration measure (i.e., each artificial fish) on the spatial environment map through the objective function, and updating the state; S554, optimizing (i.e., foraging behavior), evaluating (i.e., flocking behavior), and adjusting (i.e., tail-chasing behavior) each calibration measure according to preset conditions, and updating the status; S555: If the preset number of iterations or effect evaluation threshold is reached, execute step S556; otherwise, return to step S553 to continue optimizing each calibration measure; S556. Determine the best calibration measure to adjust the spatial environment map to reflect dynamic changes.

[0152] According to another embodiment of the present invention, Figure 2 As shown, a spatial surveying and mapping calibration system based on synchronous positioning drawing is also provided, and the spatial surveying and mapping calibration system based on synchronous positioning drawing includes: The data acquisition module 1 is used to acquire multi-source temporal spatial data related to a preset spatial environment, and pre-process the multi-source temporal spatial data to obtain environmental feature data; Data matching module 2, used to match the obtained environmental feature data with the corresponding environmental feature data in the database based on the classification algorithm, and identify the spatial environment and key landmarks; An environment map construction module 3 is used to obtain a space environment map according to the identified space environment and key landmarks based on stereo vision technology; The data optimization module 4 is used to obtain the real-time posture of the detection device through the synchronous positioning mapping technology based on the acquired space environment map, and adjust the motion trajectory of the detection device in the space environment by using the optimization algorithm; The data evaluation module 5 is used to monitor the dynamic changes in the space environment in real time, evaluate the impact of the dynamic changes on the space environment map by using an analysis algorithm, and determine whether the dynamic changes exceed a preset threshold. If the threshold is exceeded, the space environment map is calibrated; Among them, the data acquisition module 1 is connected to the environment map construction module 3 through the data matching module 2, and the environment map construction module 3 is connected to the data evaluation module 5 through the data optimization module 4.

[0153] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A spatial surveying and mapping calibration method based on synchronous positioning mapping, characterized in that: The spatial surveying and mapping calibration method based on synchronous positioning mapping comprises the following steps: S1. Acquire multi-source temporal spatial data related to a preset spatial environment, and pre-process the multi-source temporal spatial data to obtain environmental feature data; S2. Based on the classification algorithm, the obtained environmental feature data is matched with the corresponding environmental feature data in the database to identify the spatial environment and key landmarks; S3, based on stereo vision technology, obtain the spatial environment map according to the identified spatial environment and key landmarks; S4. Based on the acquired spatial environment map, the real-time posture of the detection device is obtained through synchronous positioning mapping technology, and the motion trajectory of the detection device in the spatial environment is adjusted using an optimization algorithm; S5. Real-time monitoring of dynamic changes in the space environment, using analysis algorithms to evaluate the impact of dynamic changes on the space environment map, and determining whether the dynamic changes exceed a preset threshold. If so, calibrate the space environment map; The real-time monitoring of dynamic changes in the space environment, using an analysis algorithm to evaluate the impact of the dynamic changes on the space environment map, and determining whether the dynamic changes exceed a preset threshold, if the threshold is exceeded, calibrating the space environment map includes the following steps: S51. Identify dynamic factors that affect the accuracy of spatial environment maps; S52. Determine a measure of the efficiency of the analysis of the impact of dynamic changes on the spatial environment map, and set efficiency thresholds and corresponding calibration measures; S53. If the dynamic change factor causes the analysis efficiency to be lower than a preset threshold, a preset calibration measure is adopted; S54, establishing an optimization model for updating the spatial environment map, and determining the objective function and constraint conditions of the optimization model; S55. Apply the analysis algorithm to the optimization model to find the best solution that meets the constraints as the best dynamic change, and evaluate the impact of the spatial change on the spatial environment map through the best dynamic change. If the best dynamic change exceeds a preset threshold, calibrate the spatial environment map.

2. The method for spatial surveying and calibration based on synchronous positioning mapping according to claim 1, characterized in that: The step of acquiring multi-source temporal spatial data related to a preset spatial environment and preprocessing the multi-source temporal spatial data to obtain environmental feature data comprises the following steps: S11, obtaining duplicate data, missing values ​​and outliers of multi-source time series spatial data related to a preset spatial environment, and performing denoising, filtering and smoothing on the duplicate data, missing values ​​and outliers; S12, connecting the unprocessed multi-source temporal spatial data in the multi-source temporal spatial data related to the preset spatial environment to generate a new data table, associating different data tables through external key values ​​to generate a complete data table, and obtaining an accurate data set; S13. Extract features from the accurate data set to obtain environmental feature data.

3. The method for spatial surveying and calibration based on synchronous positioning mapping according to claim 1, characterized in that: The method of matching the obtained environmental feature data with the corresponding environmental feature data in the database based on the classification algorithm to identify the spatial environment and key landmarks includes the following steps: S21, setting relevant parameters matching the environmental characteristic data, and generating preliminary environmental characteristic data and a corresponding characteristic trend matrix based on the relevant parameters; S22, calculating the fitness function value of preliminary environmental characteristic data; S23, calculate the group fitness variance of all environmental feature data, if it meets the preset conditions, then mutate and go to step S25, otherwise, continue to go to step S24 for optimization; S24, updating the inertia weight point according to the inertia weight calculation formula, and updating the change trend and position of the environmental feature data according to the change trend update formula and the position update formula respectively; S25, check whether the termination condition is met, if it is met, then end the optimization process, otherwise, return to step S22 to continue to optimize; S26, assigning the found optimal parameter vector to the classification model; S27, inputting the processed environmental feature data into the classification model, and training the classification model to obtain a trained classification model; S28. Use the trained classification model to classify new environmental feature data and identify spatial environments and key landmarks.

4. The method for spatial surveying and calibration based on synchronous positioning mapping according to claim 3, characterized in that: The method of using the trained classification model to classify new environmental feature data and identify spatial environments and key landmarks includes the following steps: S281, using the trained classification model to analyze the new environmental feature data, and predict the environmental category probability corresponding to each new environmental feature data; S282, setting a confidence threshold, screening the probability of each environmental category to which each environmental feature data belongs, eliminating the prediction results below the confidence threshold, and determining the category label of each environmental feature data from the environmental category probability output by the classification model according to the classification threshold; S282: The screened and determined category label is used as the final environment classification result, and the calculation process ends.

5. The method for spatial surveying and calibration based on synchronous positioning mapping according to claim 3, characterized in that: The inertia weight calculation formula is: ； in, H Expressed as inertia weight; H max , H min Respectively represent the starting and ending values ​​of the inertia weight; W Represents the current iteration number; W max Expressed as the maximum number of iterations.

6. The method for spatial surveying and calibration based on synchronous positioning mapping according to claim 1, characterized in that: The method of obtaining the real-time posture of the detection device based on the acquired space environment map by synchronous positioning mapping technology and adjusting the motion trajectory of the detection device in the space environment by using an optimization algorithm includes the following steps: S41, identifying key parameters that affect the positioning and path planning of the detection device in the space environment, and setting the initial search range; S42, presetting an initial value of the evaluation period, and setting the number of iterations as the end condition of the detection task optimization process; S43, by comparing the real-time posture of the detection device with the preset path planning data, updating the key parameters according to the preset optimization method; S44, using statistical methods to evaluate the detection task execution efficiency after each key parameter update; S45, based on the evaluation result of the detection task execution efficiency, calculating the fitness score of the key parameter update in each iteration process; S46, comparing the fitness scores of two consecutive iterations, and selecting the key parameter update that has the greatest impact on the detection task execution efficiency as the current best solution; S47, increasing the initial value of the evaluation cycle, and if the preset upper limit of the number of iterations is reached, terminating the evaluation process; S48. After the optimization process is completed, the matching degree between the motion trajectory of the actual detection device and the preset path planning is analyzed to determine whether there is a significant impact. If there is a significant impact, adjustments are made.

7. The method for spatial surveying and calibration based on synchronous positioning mapping according to claim 6, characterized in that: The method of comparing the real-time posture of the detection device with the preset path planning data and updating the key parameters according to the preset optimization method includes the following steps: S431, setting the evaluation range and initial parameters of the gesture recognition algorithm; S432, setting an initial simulation value for detecting the deviation between the posture of the device and the preset path, and initializing the simulation position; S433, performing a functional evaluation on each simulated posture, identifying the parameters that have the most significant impact on the path deviation in the initial state, and recording the best parameter combination; S434, simulating the change of the posture of the detection device, and selecting the optimal calibration measure by comparing the effects of different parameter combinations; S435, comparing the effect of the new parameter combination with the best parameter combination in the record, if the new parameter combination is better than the best parameter combination in the record, then updating the record, otherwise keeping the status quo; S436: If the maximum number of iterations is reached, the current best parameter combination is output as the final solution, and the process proceeds to step S437; S437: Use the obtained optimal parameter combination in the gesture recognition algorithm as a preset optimization method to update key parameters.

8. The method for spatial surveying and calibration based on synchronous positioning mapping according to claim 6, characterized in that: The use of statistical methods to evaluate the detection task execution efficiency after each key parameter update includes the following steps: S441, initializing the verification round number to the first round, and dividing the acquired real-time posture data set of the detection device into several subsets, in each round, selecting one subset as a test set, and the union of all the remaining subsets as a training set; S442, training the prediction model using the adjusted key parameters on the training set, and testing the prediction performance of the prediction model on the validation set; S443. Calculate the generalization error of the prediction model based on the actual performance of the key parameters on the test set; S444, looping step S441 to step S443 until each subset has been used as a test set; S445. Obtain an average value of the generalization errors calculated in all verification rounds as an overall evaluation result of the detection task execution efficiency after each key parameter update.

9. The method for spatial surveying and calibration based on synchronous positioning mapping according to claim 1, characterized in that: The analysis algorithm is applied to the optimization model to find the best solution that meets the constraint conditions as the best dynamic change, and the impact of the spatial change on the spatial environment map is evaluated through the best dynamic change. If the best dynamic change exceeds a preset threshold, the spatial environment map is calibrated, including the following steps: S551, processing the dynamic change factors that affect the accuracy of the spatial environment map to obtain the processed dynamic change factors, and making the processed dynamic change factors meet the preset conditions; S552, initializing analysis algorithm parameters and calibration measures for dynamically changing factors that affect the accuracy of the spatial environment map; S553, evaluating the effect of each calibration measure on the spatial environment map through the objective function, and updating the state; S554, optimize, evaluate, and adjust each calibration measure according to preset conditions, and update the status; S555, if the preset number of iterations or effect evaluation threshold is reached, execute step S556; otherwise, return to step S553 to continue optimizing each calibration; S556. Determine the best calibration measure to adjust the spatial environment map to reflect dynamic changes.

10. A spatial surveying and mapping calibration system based on synchronous positioning mapping, used to implement the spatial surveying and mapping calibration method based on synchronous positioning mapping according to any one of claims 1 to 9, characterized in that: The spatial surveying and calibration system based on synchronous positioning mapping includes: A data acquisition module is used to acquire multi-source temporal spatial data related to a preset spatial environment, and pre-process the multi-source temporal spatial data to obtain environmental feature data; A data matching module is used to match the obtained environmental feature data with the corresponding environmental feature data in the database based on the classification algorithm to identify the spatial environment and key landmarks; An environment map building module is used to obtain a spatial environment map according to the identified spatial environment and key landmarks based on stereo vision technology; The data optimization module is used to obtain the real-time posture of the detection device through synchronous positioning mapping technology based on the acquired spatial environment map, and adjust the motion trajectory of the detection device in the spatial environment using an optimization algorithm; The data evaluation module is used to monitor the dynamic changes in the space environment in real time, use the analysis algorithm to evaluate the impact of the dynamic changes on the space environment map, and determine whether the dynamic changes exceed the preset threshold. If the threshold is exceeded, the space environment map is calibrated; Among them, the data acquisition module is connected with the environment map construction module through the data matching module, and the environment map construction module is connected with the data evaluation module through the data optimization module.

Images