A multi-robot 3D vision collaborative spatial posture perception calibration method and system

Abstract

The present invention provides a multi-robot 3D visual collaborative spatial posture perception calibration method and system, which relates to the technical field of multi-robot collaborative work. At least three robots are used to collect three-dimensional visual data in real time within the collaborative work area. A calibration accuracy factor is set to quantify the posture estimation error, and the collected data is preprocessed, including noise reduction and normalization. The weight of each data source is calculated to reflect its importance in the fusion process, and a fused posture estimation model is generated through dynamic Kalman filtering, introducing historical information. The motion estimation error is set and monitored in real time. When the error exceeds the threshold, the calibration re-correction mechanism is triggered, and the robots are reconfigured to collect data and update the weights to ensure that the calibration accuracy continues to meet the requirements. The present invention improves the posture perception accuracy of multiple robots in complex environments, reduces the complexity of data processing, enhances the real-time performance of the system, and lays the foundation for the widespread application of multi-robot collaborative work.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-robot collaborative work, and in particular to a multi-robot 3D visual collaborative spatial posture perception calibration method and system. Background Art

[0002] With the rapid development of robotics, multi-robot systems have been widely used in industrial automation, warehousing and logistics, intelligent manufacturing, and other fields. Multi-robot collaboration can effectively improve operational efficiency and flexibility. However, accurately sensing and calibrating the spatial posture of each robot remains a key challenge in achieving effective collaboration.

[0003] Traditional posture perception methods typically rely on a single robot using an inertial measurement unit (IMU) or visual sensor for positioning. However, in complex environments, a single sensor is susceptible to noise and occlusion, resulting in inaccurate posture estimation. Furthermore, as the number of robots increases, so does the complexity of data processing. Therefore, effectively fusing visual data from different robots to improve the accuracy and robustness of the overall system has become a pressing issue.

[0004] Recent advances in 3D vision technology have provided new solutions for multi-robot posture perception. Utilizing multiple robots to collect 3D visual data in real time within a collaborative workspace provides a more comprehensive understanding of the environment. However, processing and fusing this data to obtain reliable posture estimates still requires effective calibration methods and algorithms.

[0005] Existing research on data fusion and pose estimation in multi-robot systems focuses on theoretical model building, lacking effective algorithms and system designs for practical application scenarios. Therefore, we propose an innovative multi-robot 3D vision collaborative spatial pose perception calibration method that not only improves pose estimation accuracy but also significantly enhances the system's adaptability and reliability, possessing important practical application value and research significance. Summary of the Invention

[0006] In order to solve the technical problems in the prior art of insufficient posture perception accuracy of multiple robots in complex environments, high data processing complexity, poor real-time performance, incomplete data collection, and difficulty in monitoring and correcting calibration errors, the present invention provides a multi-robot 3D vision collaborative spatial posture perception calibration method and system.

[0007] The technical solutions provided by the present invention are as follows:

[0008] First aspect:

[0009] The present invention provides a multi-robot 3D vision collaborative spatial posture perception calibration method, comprising:

[0010] S1. Use at least three robots to collect 3D visual data sets in real time within a collaborative work area. , each data point Represents three-dimensional coordinate information;

[0011] S2. Set the calibration precision factor , used to quantify the error range of attitude estimation, the calibration precision factor The calculation formula is:

[0012] in, represents the number of collected 3D data points, Represents the dimension of each data point Indicates the The data point in The measurement value of the dimension, Indicates the The data point in The real coordinates of the dimension;

[0013] S3. Preprocess the collected 3D data, including noise reduction and data normalization, to obtain a unified data format ;

[0014] S4. Perform data fusion using the following formula: in, is the pose estimation result after fusion, Indicates the The weight of each data source reflects its importance in the fusion. Indicates the preprocessed data points;

[0015] S5. Generate a fused attitude estimation model and apply dynamic Kalman filtering to optimize the data fusion process;

[0016] S6. Setting motion estimation error , the motion estimation error The calculation method is: in, represents the estimated motion error, Indicates the The estimated motion value of the dimensions, Indicates the The true motion value of the dimension, Represents the weight factor, reflecting the impact of calibration accuracy on the error;

[0017] S7, when the motion estimation error When the set threshold is exceeded, the calibration recalibration mechanism is triggered.

[0018] Second aspect:

[0019] The present invention provides a multi-robot 3D vision collaborative spatial posture perception and calibration system, comprising:

[0020] Multi-robot configuration, with at least three autonomous mobile robots equipped with high-precision 3D vision sensors capable of collecting 3D visual data in real time within the collaborative work area;

[0021] The data processing unit is equipped with a centralized processor that is responsible for receiving the three-dimensional data sets collected by each robot and performing data fusion and motion estimation;

[0022] Wireless communication module, configures wireless communication network to realize real-time data transmission between robots and between robots and data processing units to ensure the timeliness and accuracy of data sharing;

[0023] The calibration precision factor calculation module is used to dynamically calculate the calibration precision factor to quantify the error range of the attitude estimation and to optimize the data fusion process;

[0024] The weight calculation module calculates the weight factor based on the confidence of each data source to reflect the importance of each data source in the fusion process and improve the accuracy of the fusion results;

[0025] The calibration recalibration module monitors and analyzes motion estimation errors. When the error exceeds a set threshold, it automatically triggers a recalibration mechanism to update the robot calibration status.

[0026] The user interaction interface provides a visual operation interface, allowing users to monitor data acquisition, fusion and calibration status in real time and adjust related parameters.

[0027] The beneficial effects brought about by the technical solution provided by the present invention include at least:

[0028] (1) In the present invention, a multi-robot collaborative working mechanism is used to collect three-dimensional visual data in real time by at least three robots, effectively improving the accuracy of posture perception. This collaborative approach not only enhances the comprehensiveness of the data but also optimizes the diversity of data sources, making posture estimation more reliable in complex environments.

[0029] (2) In the present invention, the dynamic Kalman filter algorithm is used to dynamically adjust the state transfer matrix in combination with the relative position changes of each robot, which significantly reduces the complexity of data processing and enhances the real-time performance of the system. This technical means enables the system to respond to environmental changes in real time and quickly adjust the motion strategy, thereby improving the efficiency of data collection;

[0030] (3) In the present invention, through the calibration recalibration mechanism, the system can monitor the motion estimation error in real time and automatically trigger the recalibration process when the error exceeds the set threshold. This mechanism ensures that the robot can continue to maintain high-efficiency calibration accuracy during long-term operation, avoiding performance degradation due to environmental changes or equipment aging, thereby improving the stability and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0032] Figure 1 A schematic diagram of a flow chart of a multi-robot 3D vision collaborative spatial posture perception calibration method provided by an embodiment of the present invention;

[0033] Figure 2 A schematic diagram of the structure of a multi-robot 3D vision collaborative spatial posture perception and calibration system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0035] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0036] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0037] In the embodiment of the present invention, sometimes the subscript is as follows It may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meaning is the same.

[0038] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0039] Reference Manual Figure 1 , which shows a flow chart of a multi-robot 3D visual collaborative spatial posture perception calibration method provided by an embodiment of the present invention.

[0040] An embodiment of the present invention provides a multi-robot 3D vision collaborative spatial posture perception calibration method. The method can be implemented by a multi-robot 3D vision collaborative spatial posture perception calibration device. The multi-robot 3D vision collaborative spatial posture perception calibration device can be a terminal or a server. The processing flow of the multi-robot 3D vision collaborative spatial posture perception calibration method can include the following steps:

[0041] S1. Use at least three robots to collect 3D visual data sets in real time within a collaborative work area. , each data point Represents three-dimensional coordinate information.

[0042] It's important to note that by using at least three robots to collect 3D visual data sets in real time, we can fully cover the collaborative work area, reduce blind spots, and ensure rich and accurate environmental information. This multi-source data collection provides a more reliable foundation for subsequent pose estimation and improves the overall data quality of the system.

[0043] S2. Set the calibration precision factor Used to quantify the error range of attitude estimation and calibrate the precision factor The calculation formula is: in, represents the number of collected 3D data points, represents the dimension of each data point, Indicates the The data point in The real coordinates of the dimensions.

[0044] It should be noted that the calibration precision factor is set to quantify the error range in the pose estimation. Calculating this factor helps clarify the accuracy requirements in data processing, enhances confidence in the estimation results, and thus improves the accuracy of data fusion and motion estimation in subsequent steps.

[0045] S3. Preprocess the collected 3D data, including noise reduction and data normalization, to obtain a unified data format .

[0046] It's important to note that preprocessing 3D data (noise reduction and data normalization) is intended to improve data quality and consistency. By eliminating noise and standardizing the data format, we ensure data comparability in subsequent analyses, thereby optimizing data fusion.

[0047] S4. Perform data fusion using the following formula: in, is the pose estimation result after fusion, Indicates the The weight of each data source reflects its importance in the fusion. Indicates the preprocessed data points.

[0048] It's important to note that data fusion aims to integrate multiple data sources by assigning weights to reflect their importance, resulting in a more accurate pose estimate. This step ensures the system can dynamically adapt to changes in data quality and enhance its responsiveness to environmental changes.

[0049] S5. Generate a fused attitude estimation model and apply dynamic Kalman filtering to optimize the data fusion process.

[0050] It should be noted that the purpose of generating a fused attitude estimation model and applying a dynamic Kalman filter is to update and optimize the data fusion process in real time. This step improves the accuracy of attitude estimation and the real-time performance of the system by reducing the impact of noise on the estimation results.

[0051] S6. Setting motion estimation error , motion estimation error The calculation method is:

[0052] in, represents the estimated motion error, Indicates the The estimated motion value of the dimensions, Indicates the The true motion value of the dimension, Represents the weight factor, reflecting the impact of calibration accuracy on the error.

[0053] It should be noted that the purpose of setting the motion estimation error and its calculation method is to monitor the robot's motion accuracy. By quantifying the error, we can evaluate the effectiveness of the model and provide a basis for optimization, ensuring the accuracy of the final output.

[0054] S7, when motion estimation error When the set threshold is exceeded, the calibration recalibration mechanism is triggered.

[0055] It's important to note that when the motion estimation error exceeds a set threshold, a recalibration mechanism is triggered to ensure the system remains efficient and accurate. This step ensures the system can self-optimize in real time, adjusting the calibration status in a timely manner to improve overall performance.

[0056] In one possible implementation, collecting a three-dimensional visual dataset specifically includes:

[0057] Each robot navigates in its coverage area in a staggered motion, ensuring coverage of non-overlapping areas to maximize comprehensiveness of data collection;

[0058] The robots collect point cloud data synchronously through their 3D vision sensors and use the following formula to calculate the effective data volume of each robot within its field of view:

[0059] in, For the The amount of effective data collected by each robot, in units of point clouds, For the The number of objects detected by a robot during data collection, The average number of points for each target in the point cloud data is set according to the resolution of the robot sensor. For the The area covered by a robot in a specific time, is the duration of data collection;

[0060] The robot transmits the collected data to the processor in real time via a wireless communication network.

[0061] It should be noted that when collecting 3D visual data sets, a staggered navigation method is used to ensure that the robot coverage areas do not overlap. This maximizes the comprehensiveness of data collection, avoids data blind spots, and improves the richness and diversity of the data. The 3D vision sensor simultaneously collects point cloud data and calculates the effective data volume, ensuring the accuracy and reliability of the obtained data. Furthermore, real-time data transmission to the processor accelerates data processing and improves system responsiveness, thereby more effectively supporting the subsequent pose estimation and fusion processes.

[0062] In one possible implementation, The weight of the data source The calculation method is:

[0063] Collection More than five measurements from a data source and calculate the mean and standard deviation ;

[0064] Defining data confidence for: in, To avoid constants with zero denominators, For the The confidence level of each data source;

[0065] Combined data confidence Calculating weights : in, is the total number of data sources, For the The confidence level of a data source.

[0066] It should be noted that the weight calculation method is to Multiple measurements are taken for each data source, and the mean and standard deviation are calculated to ensure the stability and accuracy of the collected data. Defining data confidence helps quantify the reliability of each data source and avoid errors introduced by data fluctuations. Combining confidence with calculated weights effectively reflects the importance of each data source in the data fusion process, ensuring the accuracy and credibility of the final pose estimation results, thereby improving the overall performance and robustness of the system.

[0067] In one possible implementation, the posture estimation model is constructed by the following formula:

[0068] in, is the generated pose estimation model, is the pose estimation result after fusion, is the last estimation result, used to introduce historical information, and is a regulating factor that controls the impact of historical information on current estimates;

[0069] In the dynamic Kalman filter, the following state update formula is introduced:

[0070] in, is the state vector at the current moment, is the state transfer matrix, which is dynamically updated in combination with the actual motion model. is the control input matrix, is the control vector, is the Kalman gain;

[0071] The Kalman gain is calculated as follows: in, is the estimation error covariance matrix at the previous moment, reflecting the reliability of state estimation, is the transpose of the observation matrix, is the observation noise covariance matrix, which describes the uncertainty in the observation process;

[0072] In the Kalman filtering process, the state transfer matrix is ​​dynamically adjusted based on the relative positions of the robots. , in order to optimize the accuracy of posture estimation, specifically by obtaining feedback data of each robot's position change in real time to form an adaptive state transition model.

[0073] It should be noted that the pose estimation model is constructed by fusing historical and current data and utilizing a modulation factor to control the influence of historical information to enhance the accuracy and robustness of the estimation. The introduction of a dynamic Kalman filter allows the system to update its state in real time based on the actual motion model, thereby improving its responsiveness to motion dynamics. The calculation of the Kalman gain combines estimation error and observation noise, ensuring that the system can still provide reliable state estimates in uncertain environments. Furthermore, by dynamically adjusting the state transition matrix based on the relative positions of the robots, pose estimation becomes more accurate, contributing to the collaborative operation and efficiency of multi-robot systems.

[0074] In one possible embodiment, the regulatory factors specifically include:

[0075] Regulatory Factors and It is used to control the weight of historical information and current data in the fused posture estimation. The calculation method is as follows:

[0076] in, It is the reliability measure of the current data set, which is defined as the ratio of the number of valid data points in the current data set to the noise level, indicating the signal-to-noise ratio of the current data. The reliability measure of the historical data set is defined as the ratio of the number of valid data points in the historical data set to the noise level, reflecting the validity of the historical data;

[0077] Through dynamic calculation and ,When the data quality is high, the influence of the current data is enhanced, and when the current data quality is low, the weight of the historical data is enhanced, thereby improving the robustness and accuracy of the pose estimation;

[0078] To ensure and The effectiveness of the calculation process, and All conditions must be met and .

[0079] It should be noted that the adjustment factor is designed to optimize the fused pose estimate by dynamically adjusting the influence of historical and current data. When the current data quality is high, its weight in the estimate is increased, allowing the system to quickly respond to the latest information. When the current data quality is low, the weight of historical data is increased to maintain the stability and reliability of the estimate. This flexible weight distribution mechanism not only improves the robustness and accuracy of the pose estimate, but also effectively copes with data fluctuations in dynamic environments, thereby enhancing the collaborative capabilities of the multi-robot system. At the same time, it ensures the validity of the weight calculation, avoids unreasonable fusion results, and provides guarantees for the overall performance of the system.

[0080] In one possible implementation, a method for determining whether a data point in a data set is valid is as follows:

[0081] Set the signal-to-noise ratio threshold to , ensuring that only data points with low noise levels are included in the valid dataset;

[0082] Set the distance range of valid data points to ,in Ensure that the data points are within the effective working distance;

[0083] Data Points The effective judgment formula is:

[0084] in, For data points Validity, 1 means valid, 0 means invalid, For data points The signal-to-noise ratio, For data points Distance to the robot;

[0085] Calculate the mean of all valid data points in the dataset and standard deviation , set the outlier threshold to , compare data points Value:

[0086] A data point is considered valid only if it meets the validity criteria, otherwise it is marked as an invalid data point.

[0087] It's important to note that the data point validity determination method aims to ensure that the selected data truly reflects the environmental state, thereby improving the accuracy of pose estimation. By setting a signal-to-noise ratio threshold, the system ensures that data is only included when the noise level is low, thereby filtering out noisy data that may affect the results. Furthermore, a distance range for valid data points is set to ensure that the collected data is within the robot's operational working distance, further improving data relevance. The validity determination formula combines the signal-to-noise ratio and distance to form a comprehensive criterion, ensuring that only qualified data points are included in the valid dataset. Finally, by calculating the mean and standard deviation and setting an outlier threshold, outliers in the dataset can be effectively identified, ensuring final data quality and providing a solid foundation for subsequent pose estimation and fusion. This approach not only improves data reliability but also lays the foundation for optimizing the overall performance of the system.

[0088] In one possible implementation, the weighting factors further include:

[0089] Weighting Factor It is used to quantify the contribution of each valid data source to the fusion result and reflect its importance in the calibration process. The calculation steps are as follows:

[0090] For each valid data source, calculate the measurement error , calculate the weight factor based on the measurement error :

[0091] in, For the The measurement error of a data source reflects the accuracy of the data source when collecting data. For the The weight factor of each data source, the smaller the measurement error, the larger the weight factor;

[0092] Normalize all weight factors:

[0093] in, is the normalized weight factor, It is the sum of the weight factors of all valid data sources. Ensure that the sum of all weight factors is 1 to facilitate relative comparison.

[0094] It's important to note that the weighting factor is introduced to quantify the contribution of each valid data source to the fusion result, thereby improving the accuracy and robustness of the system. First, by calculating the measurement error of each data source, we can directly assess its reliability during the data collection process. Smaller measurement errors indicate higher data source accuracy, and therefore, these data sources should receive a greater weight in the fusion process. This calculation method ensures that high-precision data sources are prioritized during fusion, thereby improving the accuracy of the overall pose estimation.

[0095] Secondly, all weight factors are normalized to make the contributions of different data sources comparable. The normalized weight factors sum to 1, ensuring that the weights of each data source are compared within a uniform range. This effectively avoids instability caused by large weight differences. This allows the system to dynamically adjust the importance of each data source in the fusion process based on its actual performance, thereby enhancing the system's adaptability to different environments and conditions and ultimately achieving more accurate pose estimation.

[0096] In one possible implementation, the calibration recalibration mechanism specifically includes:

[0097] Reconfigure the robot to collect data within the collaborative work area to ensure the acquisition of the latest 3D visual data set. During the collection process, the number of data points should be increased to improve the representativeness and accuracy of the data.

[0098] After new data is collected, the weight factors of each data source are recalculated and normalized to ensure that the updated weights reflect the validity of the current data;

[0099] Re-execute data fusion using the updated weight factors and the newly collected data to generate new fusion results. Based on the new fusion results, optimize the motion estimation model to ensure that it more accurately reflects the robot's posture;

[0100] The new estimated result is compared with the actual motion value to verify the effectiveness of the calibration recalibration. If the error still exceeds the threshold, the above steps can be repeated until the preset accuracy requirement is met.

[0101] It's important to note that the calibration and recalibration mechanism is designed to ensure high accuracy and reliability of the robot's pose estimation in complex environments. First, the robot is reconfigured for data collection to obtain the latest 3D vision dataset. Increasing the number of data points improves data representativeness and reduces estimation errors caused by insufficient samples. This step ensures that the system can promptly reflect environmental changes, thereby improving overall data accuracy.

[0102] Next, the weight factors for each data source are recalculated and normalized to ensure that the updated weights truly reflect the validity of the new data. This process helps dynamically adjust the importance of different data sources in the fusion process, further improving the reliability of the fusion results. These updated weight factors are then reintegrated with the newly collected data to generate a new pose estimate, optimizing the motion estimation model to more accurately reflect the current robot pose.

[0103] Finally, the effectiveness of the recalibration is verified by comparing the new estimated results with the actual motion values. This step not only confirms that the updated model meets the accuracy requirements but also provides a feedback mechanism. If the error exceeds a preset threshold, the above steps are repeated to ensure that the system is always performing at its optimal state. This dynamic recalibration mechanism greatly enhances the system's adaptability and accuracy.

[0104] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0105] (1) In the present invention, a multi-robot collaborative working mechanism is used to collect three-dimensional visual data in real time by at least three robots, effectively improving the accuracy of posture perception. This collaborative approach not only enhances the comprehensiveness of the data but also optimizes the diversity of data sources, making posture estimation more reliable in complex environments.

[0106] (2) In the present invention, the dynamic Kalman filter algorithm is used to dynamically adjust the state transfer matrix in combination with the relative position changes of each robot, which significantly reduces the complexity of data processing and enhances the real-time performance of the system. This technical means enables the system to respond to environmental changes in real time and quickly adjust the motion strategy, thereby improving the efficiency of data collection;

[0107] (3) In the present invention, through the calibration recalibration mechanism, the system can monitor the motion estimation error in real time and automatically trigger the recalibration process when the error exceeds the set threshold. This mechanism ensures that the robot can continue to maintain high-efficiency calibration accuracy during long-term operation, avoiding performance degradation due to environmental changes or equipment aging, thereby improving the stability and reliability of the system.

[0108] Reference Manual Figure 2 , which shows a structural diagram of a multi-robot 3D vision collaborative spatial posture perception and calibration system provided by an embodiment of the present invention.

[0109] The present invention also provides a multi-robot 3D vision collaborative space posture perception calibration system, which is applied to the multi-robot 3D vision collaborative space posture perception calibration method, comprising:

[0110] Multi-robot configuration, with at least three autonomous mobile robots equipped with high-precision 3D vision sensors capable of collecting 3D visual data in real time within the collaborative work area;

[0111] The data processing unit is equipped with a centralized processor that is responsible for receiving the three-dimensional data sets collected by each robot and performing data fusion and motion estimation;

[0112] Wireless communication module, configures wireless communication network to realize real-time data transmission between robots and between robots and data processing units to ensure the timeliness and accuracy of data sharing;

[0113] The calibration precision factor calculation module is used to dynamically calculate the calibration precision factor to quantify the error range of the attitude estimation and to optimize the data fusion process;

[0114] The weight calculation module calculates the weight factor based on the confidence of each data source to reflect the importance of each data source in the fusion process and improve the accuracy of the fusion results;

[0115] The calibration recalibration module monitors and analyzes motion estimation errors. When the error exceeds a set threshold, it automatically triggers a recalibration mechanism to update the robot calibration status.

[0116] The user interaction interface provides a visual operation interface, allowing users to monitor data acquisition, fusion and calibration status in real time and adjust related parameters.

[0117] The system's working mechanism specifically includes:

[0118] Dynamic data collection: Each robot navigates in a staggered motion within its coverage area to ensure comprehensive and non-overlapping data collection. Each robot synchronously collects point cloud data through its onboard 3D vision sensor and transmits it to the data processing unit in real time via a wireless network.

[0119] Real-time data fusion: Through the dynamic Kalman filter algorithm, the data collected from each robot is integrated to generate a comprehensive posture estimation model, and the motion strategy of each robot is adjusted in real time to optimize the data collection quality;

[0120] Adaptive adjustment of weight factors: During the fusion process, the weight factors of each data source are dynamically calculated and adjusted according to the quality and confidence of the data collected in real time to improve the robustness and accuracy of pose estimation;

[0121] Calibration and recalibration process,In the calibration and recalibration mechanism, the system will re-evaluate the weight of each data source based on the latest collected 3D visual data, and re-execute data fusion to ensure the accuracy of the robot's posture.

[0122] The multi-robot 3D vision collaborative spatial posture perception calibration system provided by the present invention can execute the multi-robot 3D vision collaborative spatial posture perception calibration method described above and achieve the same or similar technical effects. To avoid repetition, the present invention will not go into details.

[0123] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0124] (1) In the present invention, a multi-robot collaborative working mechanism is used to collect three-dimensional visual data in real time by at least three robots, effectively improving the accuracy of posture perception. This collaborative approach not only enhances the comprehensiveness of the data but also optimizes the diversity of data sources, making posture estimation more reliable in complex environments.

[0125] (2) In the present invention, the dynamic Kalman filter algorithm is used to dynamically adjust the state transfer matrix in combination with the relative position changes of each robot, which significantly reduces the complexity of data processing and enhances the real-time performance of the system. This technical means enables the system to respond to environmental changes in real time and quickly adjust the motion strategy, thereby improving the efficiency of data collection;

[0126] (3) In the present invention, through the calibration recalibration mechanism, the system can monitor the motion estimation error in real time and automatically trigger the recalibration process when the error exceeds the set threshold. This mechanism ensures that the robot can continue to maintain high-efficiency calibration accuracy during long-term operation, avoiding performance degradation due to environmental changes or equipment aging, thereby improving the stability and reliability of the system.

[0127] The above content is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

[0128] There are a few points to note:

[0129] (1) The drawings of the embodiments of the present invention only relate to the structures related to the embodiments of the present invention. Other structures may refer to conventional designs.

[0130] (2) For the sake of clarity, the thickness of layers or regions in the drawings used to describe the embodiments of the present invention are exaggerated or reduced, that is, these drawings are not drawn to scale. It is understood that when an element such as a layer, film, region, or substrate is referred to as being "on" or "under" another element, the element may be "directly on" or "under" the other element or intervening elements may be present.

[0131] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to form new embodiments.

[0132] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A multi-robot 3D vision collaborative spatial posture perception calibration method, characterized by: include: S1. Use at least three robots to collect 3D visual data sets in real time within a collaborative work area. , each data point Represents three-dimensional coordinate information; S2. Set the calibration precision factor , used to quantify the error range of attitude estimation, the calibration precision factor The calculation formula is: in, represents the number of collected 3D data points, represents the dimension of each data point, Indicates the The data point in The measurement value of the dimension, Indicates the The data point in The real coordinates of the dimensions; S3. Preprocess the collected 3D data, including noise reduction and data normalization, to obtain a unified data format ; S4. Perform data fusion using the following formula: in, is the pose estimation result after fusion, Indicates the The weight of each data source reflects its importance in the fusion. Indicates the preprocessed data points; S5. Generate a fused attitude estimation model and apply dynamic Kalman filtering to optimize the data fusion process; S6. Setting motion estimation error , the motion estimation error The calculation method is: in, represents the estimated motion error, Indicates the The estimated motion value of the dimensions, Indicates the The true motion value of the dimension, Represents the weight factor, reflecting the impact of calibration accuracy on the error; S7, when the motion estimation error When the set threshold is exceeded, the calibration recalibration mechanism is triggered.

2. A multi-robot 3D vision collaborative spatial posture perception calibration method according to claim 1, characterized in that: The collecting of the three-dimensional visual data set specifically includes: Each robot navigates in its coverage area in a staggered motion, ensuring coverage of non-overlapping areas to maximize comprehensiveness of data collection; The robots collect point cloud data synchronously through their 3D vision sensors and use the following formula to calculate the effective data volume of each robot within its field of view: in, For the The amount of effective data collected by each robot, in units of point clouds, For the The number of objects detected by a robot during data collection, The average number of points for each target in the point cloud data is set according to the resolution of the robot sensor. For the The area covered by a robot in a specific time, is the duration of data collection; The robot transmits the collected data to the processor in real time via a wireless communication network.

3. The multi-robot 3D vision collaborative spatial posture perception calibration method according to claim 2 is characterized in that: Further including: The said The weight of the data source The calculation method is: Collection More than five measurements from a data source and calculate the mean and standard deviation : Defining data confidence for: in, To avoid constants with zero denominators, For the The confidence level of each data source; Combined with the data confidence Calculating weights : in, is the total number of data sources, For the The confidence level of a data source.

4. The multi-robot 3D vision collaborative spatial posture perception calibration method according to claim 1, characterized in that: include: The posture estimation model is constructed by the following formula: in, is the generated pose estimation model, is the pose estimation result after fusion, is the last estimation result, used to introduce historical information, and is a regulating factor that controls the impact of historical information on current estimates; In the dynamic Kalman filter, the following state update formula is introduced: in, is the state vector at the current moment, is the state transfer matrix, which is dynamically updated in combination with the actual motion model. is the control input matrix, is the control vector, is the Kalman gain; The Kalman gain is calculated by the following formula: in, is the estimation error covariance matrix at the previous moment, reflecting the reliability of state estimation, is the transpose of the observation matrix, is the observation noise covariance matrix, which describes the uncertainty in the observation process; In the Kalman filtering process, the state transfer matrix is ​​dynamically adjusted based on the relative positions of the robots. , in order to optimize the accuracy of posture estimation, specifically by obtaining feedback data of each robot's position change in real time to form an adaptive state transition model.

5. The multi-robot 3D vision collaborative spatial posture perception calibration method according to claim 4 is characterized in that: The regulating factors specifically include: The regulatory factor and It is used to control the weight of historical information and current data in the fused posture estimation. The calculation method is as follows: in, It is the reliability measure of the current data set, which is defined as the ratio of the number of valid data points in the current data set to the noise level, indicating the signal-to-noise ratio of the current data. It is a reliability measure of the historical data set, defined as the ratio of the number of valid data points in the historical data set to the noise level, reflecting the validity of the historical data; Through dynamic calculation and ,When the data quality is high, the influence of the current data is enhanced, and when the current data quality is low, the weight of the historical data is enhanced, thereby improving the robustness and accuracy of the pose estimation; To ensure and The effectiveness of the calculation process, and All conditions must be met and .

6. The multi-robot 3D vision collaborative spatial posture perception calibration method according to claim 5, characterized in that: Further including: The method to determine whether a data point in a dataset is valid is as follows: Set the signal-to-noise ratio threshold to , ensuring that only data points with low noise levels are included in the valid data set; setting the distance range of valid data points to ,in , , ensure that the data points are within the effective working distance; Data Points The effective judgment formula is: in, For data points Validity, 1 means valid, 0 means invalid, For data points The signal-to-noise ratio, For data points Distance to the robot; Calculate the mean of all valid data points in the dataset and standard deviation , set the outlier threshold to , compare data points Value: A data point is considered valid only if it meets the validity criteria, otherwise it is marked as an invalid data point.

7. The multi-robot 3D vision collaborative spatial posture perception calibration method according to claim 1, characterized in that: The weighting factors further include: Weighting Factor It is used to quantify the contribution of each valid data source to the fusion result and reflect its importance in the calibration process. The calculation steps are as follows: For each valid data source, calculate the measurement error , calculate the weight factor based on the measurement error : in, For the The measurement error of a data source reflects the accuracy of the data source when collecting data. For the The weight factor of each data source, the smaller the measurement error, the larger the weight factor; Normalize all weight factors: in, is the normalized weight factor, It is the sum of the weight factors of all valid data sources. Ensure that the sum of all weight factors is 1 to facilitate relative comparison.

8. The multi-robot 3D vision collaborative spatial posture perception calibration method according to claim 1, characterized in that: The calibration recalibration mechanism specifically includes: Reconfigure the robot to collect data within the collaborative work area to ensure the acquisition of the latest 3D visual data set. During the collection process, the number of data points should be increased to improve the representativeness and accuracy of the data. After new data is collected, the weight factors of each data source are recalculated and normalized to ensure that the updated weights reflect the validity of the current data; Re-execute data fusion using the updated weight factors and the newly collected data to generate new fusion results. Based on the new fusion results, optimize the motion estimation model to ensure that it more accurately reflects the robot's posture; The new estimated result is compared with the actual motion value to verify the effectiveness of the calibration recalibration. If the error still exceeds the threshold, the above steps can be repeated until the preset accuracy requirement is met.

9. A multi-robot 3D vision collaborative spatial posture perception calibration method according to any one of claims 1 to 8, characterized in that: It also includes a multi-robot 3D vision collaborative spatial posture perception and calibration system: Multi-robot configuration, with at least three autonomous mobile robots equipped with high-precision 3D vision sensors capable of collecting 3D visual data in real time within the collaborative work area; The data processing unit is equipped with a centralized processor that is responsible for receiving the three-dimensional data sets collected by each robot and performing data fusion and motion estimation; Wireless communication module, configures wireless communication network to realize real-time data transmission between robots and between robots and data processing units to ensure the timeliness and accuracy of data sharing; The calibration precision factor calculation module is used to dynamically calculate the calibration precision factor to quantify the error range of the attitude estimation and to optimize the data fusion process; The weight calculation module calculates the weight factor based on the confidence of each data source to reflect the importance of each data source in the fusion process and improve the accuracy of the fusion results; The calibration recalibration module monitors and analyzes motion estimation errors. When the error exceeds a set threshold, it automatically triggers a recalibration mechanism to update the robot calibration status. The user interaction interface provides a visual operation interface, allowing users to monitor data acquisition, fusion and calibration status in real time and adjust related parameters.

10. The multi-robot 3D vision collaborative spatial posture perception calibration method according to claim 9, characterized in that: The working mechanism of the multi-robot 3D vision collaborative spatial posture perception and calibration system specifically includes: Dynamic data collection: Each robot navigates in a staggered motion within its coverage area to ensure comprehensive and non-overlapping data collection. Each robot synchronously collects point cloud data through its onboard 3D vision sensor and transmits it to the data processing unit in real time via a wireless network. Real-time data fusion: Through the dynamic Kalman filter algorithm, the data collected from each robot is integrated to generate a comprehensive posture estimation model, and the motion strategy of each robot is adjusted in real time to optimize the data collection quality; Adaptive adjustment of weight factors: During the fusion process, the weight factors of each data source are dynamically calculated and adjusted according to the quality and confidence of the data collected in real time to improve the robustness and accuracy of pose estimation; Calibration and recalibration process,In the calibration and recalibration mechanism, the system will re-evaluate the weight of each data source based on the latest collected 3D visual data, and re-execute data fusion to ensure the accuracy of the robot's posture.

Images