A robot system overall calibration method and system based on multi-sensor cooperation

Abstract

The ground world coordinate system is calibrated by setting a reference point on the ground, and the robot base coordinate system is established based on the robot base features, a rigid transformation matrix between the two is calculated to realize spatial alignment. A multi-sensor network is constructed, the vision sensor and the laser radar are associated to a double-reference coordinate system, data is synchronously collected and unified to the same coordinate system. According to the spatial consistency error of the data, the sensor reliability weight is dynamically evaluated, the joint error equation is constructed and optimized based on the least square method, and the data is weighted and fused to output the calibration parameters. The application solves the problems of relying on the accuracy of a single sensor, lacking a system-level calibration scheme and a complex calibration process, and improves the calibration accuracy and efficiency of the robot system.

Description

Technical Field

[0001] This invention relates to the field of industrial automation, and specifically to a method and system for overall calibration of a robot system based on multi-sensor collaboration. Background Technology

[0002] In the field of industrial automation, robot systems are widely used in various production tasks, such as automated packing, assembly, and material handling. However, with the increasing complexity of robot application scenarios and the rising accuracy requirements of tasks, traditional robot calibration methods have gradually revealed many problems. First, traditional calibration methods typically rely on the measurement accuracy of a single sensor, such as using only a vision sensor or LiDAR for calibration. This method cannot provide an effective correction mechanism when the sensor malfunctions or fails, resulting in limited reliability of the calibration results. Second, most existing calibration schemes are designed for individual sensors or local systems, lacking schemes for overall calibration at the system level. This local calibration method cannot fully consider the collaborative relationships between multiple sensors, resulting in a lack of effective cross-validation mechanisms between sensors, which in turn affects the positioning accuracy and operational efficiency of the entire robot system. Furthermore, traditional calibration processes are usually complex, requiring multiple adjustments and calibrations, which not only increases the time cost of calibration but also raises the difficulty and cost of system maintenance. In actual industrial production, robot systems need to frequently switch tasks and adapt to environmental conditions, and complex calibration processes cannot meet the demands of rapid deployment and efficient operation. Therefore, how to solve the problems of relying on the accuracy of a single sensor, lack of a system-level overall calibration scheme, and complexity of the calibration process in existing calibration methods is a technical challenge that urgently needs to be addressed in the field of industrial robots. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for overall calibration of a robot system based on multi-sensor collaboration, so as to solve the problems of existing calibration methods such as reliance on the accuracy of a single sensor, lack of system-level overall calibration scheme, and complexity of calibration process.

[0004] To achieve the above objectives, the following technical solution is adopted.

[0005] A method for overall calibration of a robot system based on multi-sensor collaboration includes the following steps:

[0006] Step S1: Set up multiple reference points on the ground and calibrate the ground world coordinate system using a space measurement device; establish a robot base coordinate system based on the mechanical interface characteristics of the robot base; establish the spatial alignment relationship of the dual reference coordinate system by calculating the rigid transformation matrix between the ground world coordinate system and the robot base coordinate system.

[0007] Step S2: Construct a multi-sensor network by associating at least one visual sensor and at least one lidar to the ground world coordinate system or the robot base coordinate system in the dual reference coordinate system, thereby forming a topological dependency relationship between the sensors.

[0008] Step S3: Based on the dual reference coordinate system, synchronously collect measurement data of each sensor on the same target object, and unify the sensor data to the same coordinate system through the rigid transformation matrix;

[0009] Step S4: Dynamically evaluate the reliability weight of each sensor based on the spatial consistency error of the measurement data of each sensor in a unified coordinate system.

[0010] Step S5: Construct a joint error equation with the dual reference coordinate system as a constraint, iteratively optimize the error equation based on the least squares method, and combine the confidence weight to weighted fuse multi-sensor data to output the overall system calibration parameters.

[0011] Optionally, the calculation of the rigid transformation matrix in step S1 includes:

[0012] The robot's end effector is controlled to perform pose measurements in both the ground world coordinate system and the robot's base coordinate system.

[0013] An overdetermined set of equations is constructed based on two sets of pose data, and the rotation matrix and translation vector between the coordinate systems are solved by the least squares method.

[0014] The rotation matrix and translation vector are combined into a rigid transformation matrix to complete the spatial alignment of the dual reference coordinate system.

[0015] Optionally, the construction of the sensor network in step S2 includes:

[0016] Based on the measurement type and installation location of the sensors, the vision sensors are associated with the ground world coordinate system, and the lidar is associated with the robot base coordinate system;

[0017] Define the hierarchical relationship between the master sensor and the slave sensor, and configure the data dependency path to achieve synchronous acquisition and transmission of multi-source data.

[0018] Optionally, the data synchronization in step S3 includes:

[0019] Based on a unified spatial reference with dual reference coordinate systems, spatiotemporal registration is performed on image data from visual sensors and point cloud data from lidar.

[0020] A timestamp alignment mechanism ensures temporal consistency of data from multiple sensors.

[0021] Optionally, the dynamic confidence assessment in step S4 includes:

[0022] Calculate the positional deviation of the same target object in the measurement data of each sensor, and generate the local error distribution of each sensor;

[0023] The reliability weight of the sensor is dynamically adjusted based on the variance of the error distribution; the larger the error variance, the lower the weight.

[0024] When the reliability of a certain sensor falls below a preset threshold, the redundant sensor switching mechanism is triggered.

[0025] Optionally, the construction of the joint error equation in step S5 includes:

[0026] Using the spatial alignment relationship of the dual reference coordinate systems as a constraint, a coupled model of robot body pose error and sensor measurement error is established.

[0027] The confidence weight is introduced as a weighting coefficient for the error term to generate the objective optimization function.

[0028] Optionally, the iterative optimization includes:

[0029] The objective optimization function is solved in multiple rounds using a nonlinear least squares algorithm, and the robot body parameters and sensor calibration parameters are gradually corrected.

[0030] In each iteration, the sensor confidence weights are updated based on the current optimization results, and the joint error equation is recalculated.

[0031] A robot system calibration system based on multi-sensor collaboration, comprising:

[0032] The reference coordinate system construction module is used to perform ground reference point calibration, robot base coordinate system establishment, and rigid transformation matrix calculation between two reference coordinate systems.

[0033] The sensor collaboration management module is used to construct multi-sensor network topologies, achieve synchronous data acquisition, and verify spatial consistency.

[0034] The accuracy optimization control module is used to construct and iteratively optimize the joint error equation for a strategy based on a dynamic adjustment of confidence weights.

[0035] Optionally, the reference coordinate system construction module includes:

[0036] The ground reference point identification unit identifies the coordinates of ground reference points using a laser rangefinder or visual markers.

[0037] The robot base calibration unit establishes the robot's base coordinate system based on the geometric features of the mechanical interface;

[0038] The transformation matrix calculation unit solves for rigid transformation parameters based on robot end-effector pose matching data.

[0039] A computer-readable storage medium storing a computer program, characterized in that, when the program is executed by a processor, it implements the aforementioned method for overall calibration of a robot system based on multi-sensor collaboration.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] This application proposes a multi-sensor collaborative robot system calibration method. By establishing a dual-reference coordinate system, it achieves unified fusion and overall optimization of multi-sensor data, effectively solving the problems mentioned in the background technology, such as reliance on the accuracy of a single sensor, lack of a system-level overall calibration scheme, and complexity of the calibration process. This method establishes a ground world coordinate system by setting reference points on the ground and establishes a robot base coordinate system by combining the mechanical interface characteristics of the robot base. Furthermore, it achieves spatial alignment of the two coordinate systems through a rigid transformation matrix. This dual-reference coordinate system provides a reliable reference framework for the fusion of multi-sensor data, ensuring the stability and accuracy of the calibration process. Simultaneously, by constructing a multi-sensor network and synchronously collecting data from each sensor, and combining the least squares method for error optimization, it achieves cross-validation and dynamic reliability adjustment of the multi-sensor data, further improving the robustness and overall accuracy of the system. This application further refines the specific steps and technical details of the calibration method, such as the calculation of the rigid transformation matrix, the construction of the sensor network, data synchronization, dynamic reliability assessment, and iterative optimization of the joint error equation. These technical features further enhance the accuracy and efficiency of the calibration method. This application applies the calibration method to a specific system device. Through modular design, including a reference coordinate system construction module, a sensor collaborative management module, and a precision optimization control module, the calibration system achieves efficient operation and convenient maintenance. Furthermore, the implementation through a computer program further enhances the automation and scalability of the calibration method. Attached Figure Description

[0042] Figure 1 This is a schematic diagram illustrating the steps of an embodiment of a robot system overall calibration method based on multi-sensor collaboration according to the present invention.

[0043] Figure 2 This is a schematic diagram of a module of an embodiment of a robot system overall calibration system based on multi-sensor collaboration according to the present invention. Detailed Implementation

[0044] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0045] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used in this invention is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.

[0046] like Figure 1 and Figure 2 The present invention provides a method and system for overall calibration of a robot system based on multi-sensor collaboration, which aims to solve the problems of existing calibration methods, such as reliance on the accuracy of a single sensor, lack of a system-level overall calibration scheme, and complexity of the calibration process.

[0047] In this invention, a dual-reference coordinate system is first established, comprising a ground world coordinate system and a robot base coordinate system. The ground world coordinate system is established by setting multiple reference points on the ground and calibrating them using spatial measurement equipment (such as laser rangefinders, total stations, etc.). These reference points can be pre-set markers or coordinate points directly obtained through measurement equipment. By accurately measuring the positions of these reference points, the origin and orientation of the ground world coordinate system can be determined.

[0048] The robot's base coordinate system is established based on the mechanical interface features of the robot's base. Specifically, the origin and orientation of the robot's base coordinate system can be determined by measuring the geometric dimensions, installation position, and feature points of the mechanical interface of the robot's base. For example, a laser tracker or vision measurement equipment can be used to measure the feature points of the robot's base to establish the robot's base coordinate system.

[0049] To achieve spatial alignment between the ground world coordinate system and the robot's base coordinate system, a rigid transformation matrix between them needs to be calculated. The specific steps are as follows: First, the robot's end effector performs pose measurements in both the ground world coordinate system and the robot's base coordinate system, acquiring two sets of pose data. Then, based on these two sets of pose data, an overdetermined system of equations is constructed, and the rotation matrix and translation vector between the coordinate systems are solved using the least squares method. Finally, the rotation matrix and translation vector are combined into a rigid transformation matrix, thus completing the spatial alignment of the two reference coordinate systems.

[0050] In practical applications, pose measurement can be achieved in various ways. For example, a vision sensor can be used to acquire the position and orientation of the robot's end effector in the ground world coordinate system, while the robot's own sensors (such as an encoder) can be used to acquire its pose information in the robot's base coordinate system. By comparing the two sets of pose data, the rigid transformation matrix can be accurately calculated.

[0051] Furthermore, to improve the calculation accuracy of the rigid transformation matrix, the above measurement process can be repeated at multiple different positions and orientations to obtain multiple sets of pose data, which can then be optimized using the least squares method. This method can effectively reduce measurement errors and improve the accuracy of coordinate system alignment.

[0052] After establishing the dual-reference coordinate system, a multi-sensor network needs to be constructed, and each sensor needs to be associated with its corresponding coordinate system. In this invention, at least one vision sensor (such as a camera) and one lidar are included. The vision sensor can be associated with the ground world coordinate system, while the lidar can be associated with the robot's base coordinate system. This association method can be flexibly adjusted according to the sensor's measurement type and installation location.

[0053] To achieve synchronous acquisition and transmission of multi-sensor data, it is necessary to define the hierarchical relationship between master and slave sensors and configure data dependency paths. For example, a vision sensor can be used as the master sensor, and a LiDAR sensor as the slave sensor. The master sensor is responsible for triggering the data acquisition process, and the slave sensors synchronously acquire data based on the trigger signal from the master sensor. In this way, the consistency of multi-source data in time and space can be ensured.

[0054] During data synchronization, spatiotemporal registration of image data from visual sensors and point cloud data from LiDAR is required. Specifically, a timestamp alignment mechanism can be used to ensure temporal consistency of multi-sensor data. For example, timestamp information can be added to the data from each sensor, and a timestamp alignment algorithm can be used to align the data from different sensors to the same point in time. Simultaneously, based on a unified spatial reference using dual reference coordinate systems, spatial registration of image data and point cloud data can be performed, transforming them to the same coordinate system.

[0055] In practical applications, spatiotemporal registration can be achieved using various algorithms. For example, interpolation algorithms can be used to interpolate the time of image data and point cloud data to compensate for the sampling time differences between sensors. For spatial registration, a rigid transformation matrix can be used to transform point cloud data to the coordinate system of the image data, or vice versa, thereby achieving spatial alignment between the two.

[0056] Furthermore, to improve the accuracy of data synchronization, a data verification mechanism can be established between sensors. For example, by comparing the measurement results of different sensors on the same target object, anomalies or errors can be detected. If anomalies are detected, data resampling or sensor calibration processes can be triggered to ensure data accuracy and consistency.

[0057] After simultaneous acquisition of data from multiple sensors, the reliability of each sensor needs to be dynamically evaluated. Specifically, this can be achieved by calculating the positional deviation of the same target object in the measurement data from each sensor, thereby generating the local error distribution of each sensor. For example, the difference between the measurement positions of the same target object by a visual sensor and a LiDAR can be calculated, and the distribution of these differences can be statistically analyzed.

[0058] Based on the variance of the error distribution, the reliability weights of sensors can be dynamically adjusted. Specifically, a larger error variance indicates that the sensor's measurement results are more unstable, and its reliability weight should be reduced accordingly. Conversely, a smaller error variance indicates that the sensor's measurement results are more reliable, and its reliability weight should be increased accordingly. In this way, the weights of each sensor in the data fusion process can be adjusted in real time, thereby improving the overall accuracy and robustness of the system.

[0059] Furthermore, when the reliability of a certain sensor falls below a preset threshold, a redundant sensor switching mechanism can be triggered. For example, if the reliability of a certain vision sensor remains below a set threshold, it can be switched to a backup vision sensor, or the weight of the LiDAR can be increased to ensure the normal operation of the system.

[0060] In terms of error optimization, a joint error equation constrained by a dual-reference coordinate system needs to be constructed. Specifically, a coupled model of robot pose error and sensor measurement error can be used as the basis for the error equation, and the reliability weight of the sensors can be introduced as a weighting coefficient for the error term. In this way, an objective optimization function can be generated for iterative optimization of the error equation.

[0061] During the iterative optimization process, a nonlinear least squares algorithm can be used to solve the objective optimization function in multiple rounds. Specifically, in each iteration, the confidence weights of the sensors are updated based on the current optimization results, and the joint error equation is recalculated. By progressively correcting the robot's body parameters and sensor calibration parameters, the overall system calibration parameters can be finally output, thereby achieving overall optimization of the robot system.

[0062] In practical applications, error optimization can be achieved through various algorithms. For example, nonlinear least squares algorithms such as the Gauss-Newton method and the Levonberg-Marquardt method can be used for optimization. These algorithms can select appropriate optimization strategies based on the characteristics of the error equation, thereby improving optimization efficiency and accuracy.

[0063] Furthermore, to improve the robustness of error optimization, constraints can be introduced during the optimization process. For example, the spatial alignment of the two reference coordinate systems can be used as a hard constraint to ensure that the optimization results conform to the actual coordinate system relationship. Simultaneously, the stability and convergence speed of the optimization process can be controlled by setting reasonable optimization parameters (such as iteration step size and convergence threshold).

[0064] To implement the above calibration method, this invention also provides a robot system overall calibration system based on multi-sensor collaboration. This system includes a reference coordinate system construction module, a sensor collaboration management module, and a precision optimization control module.

[0065] The reference coordinate system construction module is used to perform ground reference point calibration, robot base coordinate system establishment, and rigid transformation matrix calculation between the two reference coordinate systems. Specifically, this module includes a ground reference point identification unit, a robot base calibration unit, and a transformation matrix calculation unit. The ground reference point identification unit can identify the coordinates of ground reference points using a laser rangefinder or visual markers; the robot base calibration unit establishes the robot base coordinate system based on the geometric features of the mechanical interface; and the transformation matrix calculation unit solves for the rigid transformation parameters based on the robot end-effector pose matching data.

[0066] The sensor collaboration management module is used to construct a multi-sensor network topology, enabling synchronous data acquisition and spatial consistency verification. This module can associate vision sensors with the ground world coordinate system and LiDAR with the robot's base coordinate system based on the sensor's measurement type and installation location. Simultaneously, it defines the hierarchical relationship between master and slave sensors and configures data dependency paths to achieve synchronous acquisition and transmission of multi-source data.

[0067] The accuracy optimization control module is used to dynamically adjust the strategy based on the reliability weights, and to construct and iteratively optimize the joint error equation. This module can dynamically evaluate the reliability weights of sensors based on the spatial consistency error of the measurement data from each sensor, and adjust the weights according to the variance of the error distribution. Simultaneously, it constructs a joint error equation constrained by a dual-reference coordinate system, and iteratively optimizes it using a nonlinear least squares algorithm, ultimately outputting the overall system calibration parameters.

[0068] In practical applications, this calibration system can be implemented through a combination of hardware devices (such as sensors and measuring equipment) and software algorithms (such as data synchronization algorithms and error optimization algorithms). For example, multiple sensors (such as cameras and LiDAR) can be installed in the robot system, and computer programs can be used to process and optimize the sensor data. In this way, efficient calibration and optimization of the robot system can be achieved.

[0069] Furthermore, to improve the system's scalability and flexibility, a modular design can be introduced into the calibration system. For example, the reference coordinate system construction module, sensor collaborative management module, and accuracy optimization control module can be designed as independent modules, each of which can be customized and optimized according to different application scenarios. Simultaneously, through interfaces and communication mechanisms between modules, the overall collaborative operation of the system can be achieved.

[0070] To further improve the automation and scalability of the calibration method, this invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, this program can implement the aforementioned overall calibration method for a robot system based on multi-sensor collaboration.

[0071] Specifically, the computer program may include multiple modules, each corresponding to a step in the calibration method. For example, it may include a reference coordinate system construction module, a sensor collaborative management module, a data synchronization module, a reliability assessment module, and an error optimization module. Each module can implement its function through corresponding algorithms and data structures, and communicate and collaborate with other modules through a program interface.

[0072] In program implementation, various programming languages ​​and development tools can be used. For example, programs can be developed using programming languages ​​such as C++ and Python, and developed and debugged using open-source computer vision libraries (such as OpenCV) and robot operating systems (such as ROS). In this way, efficient calibration and optimization of the robot system can be achieved.

[0073] Furthermore, to improve the program's portability and compatibility, it can be tested and optimized on different hardware platforms. For example, it can be tested in industrial robot systems to verify the program's performance and reliability. Simultaneously, the program's efficiency and accuracy can be improved by optimizing its algorithms and data structures.

[0074] In practical applications, this computer-readable storage medium can be stored in the robot system's controller or in a computer device connected to the robot system. By loading and running this program, automatic calibration and optimization of the robot system can be achieved, improving the system's operating efficiency and accuracy.

[0075] In summary, this invention provides an efficient and accurate overall calibration method and system for robot systems by establishing a dual-reference coordinate system, constructing a multi-sensor network, achieving synchronous data acquisition and transmission, dynamically evaluating sensor reliability, and optimizing error equations. This method and system effectively solve the problems of existing calibration methods, such as reliance on the accuracy of a single sensor, lack of a system-level overall calibration scheme, and complex calibration processes. It has broad application prospects and significant practical value.

[0076] As is known from common technical knowledge, this invention can be implemented through other embodiments that do not depart from its spirit or essential characteristics. Therefore, the disclosed embodiments described above are merely illustrative in all respects and are not the only ones. All modifications within the scope of this invention or its equivalents are included in this invention.

Claims

1. A method for overall calibration of a robot system based on multi-sensor collaboration, characterized in that, Includes the following steps, Step S1: Set up multiple reference points on the ground and calibrate the ground world coordinate system using space measurement equipment; A robot base coordinate system is established based on the mechanical interface features of the robot base; the spatial alignment relationship of the dual reference coordinate system is established by calculating the rigid transformation matrix between the ground world coordinate system and the robot base coordinate system. Step S2: Construct a multi-sensor network by associating at least one visual sensor and at least one lidar to the ground world coordinate system or the robot base coordinate system in the dual reference coordinate system, thereby forming a topological dependency relationship between the sensors. Step S3: Based on the dual reference coordinate system, synchronously collect measurement data of each sensor on the same target object, and unify the sensor data to the same coordinate system through the rigid transformation matrix; Step S4: Dynamically evaluate the reliability weight of each sensor based on the spatial consistency error of the measurement data of each sensor in a unified coordinate system. Step S5: Construct a joint error equation with the dual reference coordinate system as a constraint, iteratively optimize the error equation based on the least squares method, and combine the confidence weight to weighted fuse multi-sensor data to output the overall system calibration parameters.

2. The overall calibration method for a robot system based on multi-sensor collaboration according to claim 1, characterized in that, The calculation of the rigid transformation matrix in step S1 includes: The robot's end effector is controlled to perform pose measurements in both the ground world coordinate system and the robot's base coordinate system. An overdetermined set of equations is constructed based on two sets of pose data, and the rotation matrix and translation vector between the coordinate systems are solved by the least squares method. The rotation matrix and translation vector are combined into a rigid transformation matrix to complete the spatial alignment of the dual reference coordinate system.

3. The overall calibration method for a robot system based on multi-sensor collaboration according to claim 1, characterized in that, The construction of the sensor network in step S2 includes: Based on the measurement type and installation location of the sensors, the vision sensors are associated with the ground world coordinate system, and the lidar is associated with the robot base coordinate system; Define the hierarchical relationship between the master sensor and the slave sensor, and configure the data dependency path to achieve synchronous acquisition and transmission of multi-source data.

4. The overall calibration method for a robot system based on multi-sensor collaboration according to claim 1, characterized in that, The data synchronization in step S3 includes: Based on a unified spatial reference with dual reference coordinate systems, spatiotemporal registration is performed on image data from visual sensors and point cloud data from lidar. A timestamp alignment mechanism ensures temporal consistency of data from multiple sensors.

5. The overall calibration method for a robot system based on multi-sensor collaboration according to claim 1, characterized in that, The dynamic reliability assessment in step S4 includes: Calculate the positional deviation of the same target object in the measurement data of each sensor, and generate the local error distribution of each sensor; The reliability weight of the sensor is dynamically adjusted based on the variance of the error distribution; the larger the error variance, the lower the weight. When the reliability of a certain sensor falls below a preset threshold, the redundant sensor switching mechanism is triggered.

6. The overall calibration method for a robot system based on multi-sensor collaboration according to claim 1, characterized in that, The construction of the joint error equation in step S5 includes: Using the spatial alignment relationship of the dual reference coordinate systems as a constraint, a coupled model of robot body pose error and sensor measurement error is established. The confidence weight is introduced as a weighting coefficient for the error term to generate the objective optimization function.

7. The method for overall calibration of a robot system based on multi-sensor collaboration according to claim 6, characterized in that, The iterative optimization includes: The objective optimization function is solved in multiple rounds using a nonlinear least squares algorithm, and the robot body parameters and sensor calibration parameters are gradually corrected. In each iteration, the sensor confidence weights are updated based on the current optimization results, and the joint error equation is recalculated.

8. A robot system overall calibration system based on multi-sensor collaboration, based on the robot system overall calibration method based on multi-sensor collaboration as described in any one of claims 1-7, characterized in that, include: The reference coordinate system construction module is used to perform ground reference point calibration, robot base coordinate system establishment, and rigid transformation matrix calculation between two reference coordinate systems. The sensor collaboration management module is used to construct multi-sensor network topologies, achieve synchronous data acquisition, and verify spatial consistency. The accuracy optimization control module is used to construct and iteratively optimize the joint error equation for a strategy based on a dynamic adjustment of confidence weights.

9. The overall calibration system for a robot system based on multi-sensor collaboration according to claim 8, characterized in that, The reference coordinate system construction module includes: The ground reference point identification unit identifies the coordinates of ground reference points using a laser rangefinder or visual markers. The robot base calibration unit establishes the robot's base coordinate system based on the geometric features of the mechanical interface; The transformation matrix calculation unit solves for rigid transformation parameters based on robot end-effector pose matching data.

10. A computer-readable storage medium storing a computer program, characterized in that, When the program is executed by the processor, it implements a method for overall calibration of a robot system based on multi-sensor collaboration as described in any one of claims 1-7.

Images