A method and device for calibrating a robotic arm system for spatial motion trajectory simulation

Abstract

The present invention discloses a calibration method and device for a robotic arm system for spatial motion trajectory simulation, which belongs to the field of robotic arm systems. It includes an ultra-wideband positioning network, which is composed of several ultra-wideband anchor modules; a six-degree-of-freedom robotic arm system, each joint of which is integrated with an ultra-wideband master antenna and an inertial measurement unit; a binocular vision module, which is installed on the end joint of the robotic arm and establishes a spatial association with the robotic arm coordinate system through a rigid connection; a multi-source data fusion module, which is used to fuse and solve the acceleration, angular velocity, and attitude quaternion collected by the inertial measurement unit with the ultra-wideband positioning result; a calibration module, which includes a checkerboard calibration plate and a motion parameter solution module, and the motion parameter solution module is used to complete the calibration of the joint motion parameters of the robotic arm according to the measured value of each joint motion trajectory and the true value of the motion trajectory. The present invention can significantly improve the positioning accuracy and stability, and provide a reliable data basis for the dynamic motion strategy adjustment of the robotic arm system.

Description

Technical Field

[0001] The present invention relates to the field of robotic arm systems, and in particular to a robotic arm system calibration method and device for spatial motion trajectory simulation. Background Art

[0002] Industrial robots, as key equipment in the automation upgrade of manufacturing, play a vital role, with applications across all segments of industrial production. To perform complex tasks, the robotic arm must be capable of six degrees of freedom (DOF) in three dimensions. This allows the arm to move and rotate in all directions, reach any target position, and adjust its posture. The emergence of six-DOF simulation technology has brought breakthroughs in robotic arm motion control.

[0003] However, there is an urgent need across various fields for simulating and measuring the trajectory of objects moving with six degrees of freedom (DOF), but current technologies face numerous challenges that need to be addressed. Traditional equipment based on optical imaging principles and inertial measurement units (IMUs) has limitations when dealing with objects moving with six degrees of freedom. Optical imaging equipment is susceptible to frame rate limitations, posture changes, and occlusions, leading to inaccurate trajectory capture and data loss. Inertial measurement units (IMUs) exhibit cumulative errors, increasing the longer the measurement is performed, making it difficult to achieve high-precision recording over long periods of time. These data challenges hinder the application and performance improvement of industrial robotics technology.

[0004] However, various fields have different requirements for the position trajectory (X, Y, Z) and posture trajectory (Roll, Pitch, There is an urgent need for trajectory simulation and measurement of six-degree-of-freedom (6DOF) systems. Current measurement equipment, primarily based on optical imaging, such as motion capture systems and industrial vision systems, relies on capturing markers / natural features or on inertial measurement units (IMUs) and positioning technologies such as ultra-wideband (UWB). These devices have significant inherent limitations in acquiring complete, high-precision 6DOF trajectory data. Optical devices are susceptible to frame rate limitations (resulting in loss of high-speed motion details and trajectory discontinuity), drastic attitude changes (causing feature blurring / deformation / movement out of the field of view, seriously impacting attitude accuracy), and environmental / self-occlusion (resulting in missing key points, data frame loss, or calculation failure), resulting in inaccurate and discontinuous trajectories. IMUs are fundamentally constrained by inherent sensor noise, bias drift, and scale factor errors that accumulate and significantly amplify over time during the integration process. This makes them unable to independently achieve long-term, high-precision recording, and errors increase rapidly with operating time. Ultra-wideband (UWB) positioning accuracy is affected by multipath and non-line-of-sight (NLOS) environments, making it difficult to provide high-precision attitude information, and thus has limited improvement in achieving complete 6DOF trajectory measurement. The limitations of these technologies result in serious deficiencies in the measurement data in terms of accuracy, continuity, robustness and long-term stability. Summary of the Invention

[0005] In order to solve the problems of accuracy and stability in simulation and measurement of joint motion during calibration of a robotic arm system, the present invention proposes a robotic arm system calibration method and device for spatial motion trajectory simulation.

[0006] The technical solution adopted in the present invention is as follows:

[0007] In a first aspect, the present invention provides a robotic arm system calibration device for spatial motion trajectory simulation, comprising:

[0008] The ultra-wideband positioning network consists of several ultra-wideband anchor modules arranged on a substrate in the experimental scene. Each ultra-wideband anchor module includes a retractable bracket, an ultra-wideband anchor antenna mounted thereon, and a first control module. The ultra-wideband anchor antenna is unobstructed and provides full-area signal coverage.

[0009] A six-degree-of-freedom robotic arm system, comprising a movable base plate, a control panel, and a robotic arm consisting of six joints connected in series, each of which integrates an ultra-wideband master control antenna, an inertial measurement unit, and a second control module;

[0010] The binocular vision module is installed on the end joint of the robotic arm and establishes a spatial association with the robotic arm coordinate system through a rigid connection;

[0011] The multi-source data fusion module is used to fuse the acceleration, angular velocity, attitude quaternion collected by the inertial measurement unit with the ultra-wideband positioning results. Its state vector is defined as ,in is the error state vector, is the accelerometer bias error in the inertial measurement unit, is the gyroscope bias error in the inertial measurement unit, is the attitude angle error;

[0012] The calibration module includes a checkerboard calibration plate and a motion parameter calculation module. The checkerboard calibration plate is fixed on the experimental scene substrate to ensure that it is always within the visual range of the binocular vision module to realize the real-time ranging function of the binocular vision module; the motion parameter calculation module is used to complete the joint motion parameter calibration of the robot arm based on the measured value of each joint motion trajectory and the true value of the motion trajectory.

[0013] Furthermore, the ultra-wideband anchor antenna is fixed to the top of the retractable bracket through a pan-tilt platform, and the pan-tilt platform has an azimuth adjustment function.

[0014] Furthermore, the sampling frequencies of the ultra-wideband master antenna and the inertial measurement unit are the same.

[0015] Furthermore, the number of ultra-wideband anchor modules is not less than 3.

[0016] In a second aspect, the present invention proposes a calibration method for a robotic arm system calibration device based on the above-mentioned spatial motion trajectory simulation, comprising the following steps:

[0017] Step 1: Arrange an ultra-wideband positioning network around the motion scene substrate, start the robotic arm and set the joint motion parameters, including the starting position, end position, and motion speed; calibrate the binocular camera;

[0018] Step 2: In real time, the inertial measurement data and ultra-wideband positioning results are fused through indirect Kalman filtering during the movement of the manipulator. The state vector is defined as , the indirect Kalman filter observation residual equation is defined as ;in, Indicates the current ultra-wideband positioning result, Indicates the next moment position predicted based on the current inertial measurement data, that is, the inertial measurement predicted position before correction. is the current inertial measurement noise;

[0019] Using inertial measurement data as control input, the prior error state estimate is calculated through the state transfer equation. The a posteriori error state estimate is calculated based on the prior error state estimate and the observation residual equation. The inertial measurement data is updated based on the accelerometer bias error, gyroscope bias error, and attitude angle error in the a posteriori error state estimate. The final posture is calculated using the updated inertial measurement data to form the motion trajectory measurement value of each joint.

[0020] Step 3: Use the calibrated binocular camera to obtain the target distance information from the checkerboard calibration plate in real time during the movement of the robot arm to obtain the true value of the motion trajectory of the robot arm's end joint, and infer the true value of the motion trajectory of the remaining joints;

[0021] Step 4: Complete the calibration of the joint motion parameters of the robotic arm based on the measured values ​​of each joint motion trajectory and the true value of the motion trajectory.

[0022] Furthermore, the formulas for the prior error state estimation and the posterior error state estimation are as follows:

[0023]

[0024]

[0025] in, represents the prior error state estimate, represents the posterior error state estimate, represents the nonlinear state transfer equation, represents the state vector at time k-1, represents the k-1 moment process noise, represents the current Kalman gain value, represents the observation matrix.

[0026] Furthermore, the updating of the inertial measurement data refers to adding the accelerometer bias and gyroscope bias in the inertial measurement data to the accelerometer bias error and gyroscope bias error, respectively, to obtain the updated accelerometer bias and gyroscope bias; and Perform quaternion multiplication as the updated attitude quaternion; where, represents the attitude angle error in the posterior error state estimate.

[0027] Furthermore, the process of calibrating the binocular camera includes:

[0028] Fix the checkerboard calibration plate on the experimental scene substrate;

[0029] Keep the base joint connecting the bottom of the robot arm to the base plate fixed, drive the remaining joints to perform several sets of posture transformations and synchronously collect calibration plate images;

[0030] Zhang's calibration method is used to solve the spatial transformation matrix of the binocular camera relative to the robotic arm base coordinate system.

[0031] Furthermore, the process of obtaining ultra-wideband positioning results includes:

[0032] The ultra-wideband main control antenna on the robotic arm acts as a mobile signal source, actively transmitting pulse signals to the ultra-wideband anchor antenna in the ultra-wideband anchor module; the ultra-wideband anchor antenna acts as a spatial position reference point, receiving the signal transmitted by the ultra-wideband main control antenna on the robotic arm, calculating the distance information in the first control module, and sending the calculated distance information back to the ultra-wideband main control antenna, and calculating the ultra-wideband positioning result in the second control module.

[0033] Furthermore, during the movement of the robotic arm, the horizontal movement of the robotic arm is achieved through the movable base plate.

[0034] The beneficial effects of the present invention are:

[0035] The present invention realizes efficient six-degree-of-freedom motion simulation and measurement of each joint of the robotic arm system by integrating multi-module collaborative control and indirect Kalman filter data fusion technology. Combined with the binocular vision real-time ranging technology optimized by Zhang calibration, the calibration of the robotic arm system is realized, which effectively overcomes equipment errors and environmental interference, significantly improves positioning accuracy and stability, and provides a reliable data basis for the dynamic motion strategy adjustment of the robotic arm system. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a schematic diagram of a robotic arm system calibration device for spatial motion trajectory simulation shown in this embodiment;

[0037] Figure 2 This is a schematic diagram of the robotic arm and binocular camera;

[0038] Figure 3 This is a schematic diagram of the joints of the robotic arm;

[0039] Figure 4 It is a flow chart of the calibration method of the robotic arm system calibration device;

[0040] Figure 5 is the motion trajectory result diagram;

[0041] In the figure: 1-base plate, 2-movable base plate, 3-control panel, 4-retractable bracket, 5-ultra-wideband anchor antenna, 6-robotic arm, 61-first joint, 62-second joint, 63-third joint, 64-fourth joint, 65-fifth joint, 66-sixth joint, 7-chessboard calibration plate, 8-binocular camera, 9-inertial measurement unit, 10-ultra-wideband main control antenna, 11-servo. DETAILED DESCRIPTION

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0043] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features indicated. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0044] In the description of the present invention, it should be noted that, unless otherwise specified or limited, the terms "mounted" and "connected" should be understood in a broad sense. For example, they can refer to fixed connection, detachable connection, or integral connection; mechanical connection, electrical connection; direct connection, indirect connection through an intermediate medium, or internal communication between two components. Those skilled in the art will be able to understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0045] The accompanying drawings show various structural schematic diagrams of the embodiments disclosed in the present invention. These drawings are not drawn to scale, and some details are exaggerated and may be omitted for the purpose of clarity.

[0046] like Figure 1 The figure shows a calibration device for a robotic arm system that simulates spatial motion trajectory, which mainly includes:

[0047] The ultra-wideband positioning network is composed of several ultra-wideband anchor modules arranged on the experimental scene substrate 1. Each ultra-wideband anchor module includes a retractable bracket 4 and an ultra-wideband anchor antenna 5 installed thereon, and a first control module. The ultra-wideband anchor antenna is unobstructed and has full-area signal coverage. Here, the ultra-wideband anchor antenna is fixed to the top of the retractable bracket through a pan-tilt platform, and the pan-tilt platform has an azimuth adjustment function. The core function of the retractable bracket and the pan-tilt platform is to provide a stable, height-adjustable support platform for the ultra-wideband anchor antenna, ensuring that the ultra-wideband anchor antenna is always in the optimal working position to ensure the stability and integrity of signal acquisition. The first control module collects signals through the ultra-wideband anchor antenna. The antenna must always be maintained in an unobstructed condition to avoid signal attenuation or interruption due to obstruction, thereby effectively improving the efficiency and accuracy of signal acquisition.

[0048] The six-degree-of-freedom robotic arm system includes a movable base plate 2, a control panel 3, and a robotic arm 6 composed of six joints connected in series. Figure 3 As shown, each joint is integrated with an ultra-wideband master control antenna 10, an inertial measurement unit 9 and a second control module;

[0049] The binocular vision module is installed on the end joint of the robotic arm and establishes a spatial association with the robotic arm coordinate system through a rigid connection;

[0050] The multi-source data fusion module is used to fuse the acceleration, angular velocity, attitude quaternion collected by the inertial measurement unit with the ultra-wideband positioning results. Its state vector is defined as ,in is the error state vector, is the accelerometer bias error in the inertial measurement unit, is the gyroscope bias error in the inertial measurement unit, is the attitude angle error;

[0051] The calibration module includes a checkerboard calibration plate 7 and a motion parameter calculation module. The checkerboard calibration plate is fixed on the experimental scene substrate to ensure that it is always within the visual range of the binocular vision module to realize the real-time ranging function of the binocular vision module; the motion parameter calculation module is used to complete the joint motion parameter calibration of the robotic arm based on the measured value of each joint motion trajectory and the true value of the motion trajectory.

[0052] In a specific implementation of the present invention, the sampling frequencies of the ultra-wideband master antenna and the inertial measurement unit are the same.

[0053] The number of ultra-wideband anchor point modules is not less than 3. In this embodiment, 4 ultra-wideband anchor point modules are used.

[0054] like Figure 1 As shown, the entire device includes: a substrate 1, a movable base plate 2, a control panel 3, a retractable bracket 4, an ultra-wideband anchor antenna 5, a robotic arm 6, a checkerboard calibration plate 7, and a binocular camera 8; Figure 2 As shown, the robotic arm 6 includes a first joint 61, a second joint 62, a third joint 63, a fourth joint 64, a fifth joint 65, and a sixth joint 66, and a binocular camera 8 is installed at the end of the robotic arm.

[0055] Universal wheels are installed at the bottom of the movable base plate 2, allowing the robotic arm to easily reposition between different workstations and enhancing the overall maneuverability of the entire device. A control panel is mounted directly above the movable base plate. This integrated control button, display screen, and user interface allow for easy parameter setting, function selection, and motion control operations through a simple, clear interface. This provides powerful operational support for efficient and stable operation of the device, enabling each function to operate with high efficiency. The robotic arm's six degrees of freedom (DOF) is primarily achieved through six integrated ultra-wideband control antennas for precise positioning and an inertial measurement unit (IMU) for real-time motion sensing. Precise control of each joint is achieved through servos 11 at each joint. Joint I, which bears the weight of the robotic arm and enables left-right rotation, is mounted and fixed directly above the movable base plate. Following a strict bottom-to-top assembly sequence, each joint is assembled and fixed to its corresponding second control module to form the robotic arm. Each joint is precisely positioned and debugged to ensure precise angular changes and flexible movement during motion. The corresponding second control module, through precise circuit connections, closely cooperates with the joints and, according to a pre-set control program, precisely regulates joint motion, enabling the robotic arm to perform complex six-degree-of-freedom movements and meet diverse application requirements. A binocular camera 8 is mounted above the last joint and calibrated with high precision using the Zhang calibration method. This overcomes positioning errors caused by factors such as device installation errors, lens distortion, and ambient light interference, improving measurement accuracy and stability.

[0056] When operating the calibration device, the corresponding joint motion trajectory can be flexibly selected and set through each second control module. Accurately set the starting position and end position of the robotic arm. According to actual needs, the movement speed is set independently. In order to analyze the characteristics of a specific trajectory under different motion conditions and ensure the integrity and accuracy of the acquired data, the user can set a variety of different movement speeds to analyze the trajectory. In addition, the sampling frequency of the first and second control modules must be set. The setting of the sampling frequency determines the number of times the control module collects data per unit time. By reasonably setting the sampling frequency, the control module can accurately collect data at the set time interval during the joint movement process. Here, the second control module is also used to solve the ultra-wideband positioning results, and the first control module is also used to solve the distance information based on the received signal transmitted by the ultra-wideband main control antenna on the robotic arm.

[0057] In one specific embodiment of the present invention, the control panel, as a human-machine interaction component, integrates a multimodal operation terminal and is primarily composed of three parts: a hardware control module, a visual interaction interface, and a system expansion interface. The hardware control module is equipped with an emergency stop protection device, an electromechanical robotic arm operating mode selection switch, and a UWB positioning system start-stop control unit to ensure safe and reliable system operation. The visual interaction interface utilizes a high-resolution touchscreen display that supports multi-touch interaction based on capacitive touch technology, enabling real-time visualization of motion parameters and simultaneous display of robotic arm posture data, UWB signal strength distribution, and more. The system expansion interface is equipped with industrial-grade communication interfaces (Gigabit Ethernet / IEEE 802.3, RS-422 / TIA-485) and high-speed data transmission channels (USB 3.2 Gen 2), and includes a programmable SDK expansion slot to support hot-swappable algorithm modules. This design fully meets complex operational requirements through the organic integration of hardware control, data visualization, and system expansion.

[0058] like Figure 4 As shown, the calibration method of the robotic arm system calibration device using the above-mentioned spatial motion trajectory simulation mainly includes the following steps:

[0059] S1: Arrange an ultra-wideband positioning network around the motion scene substrate, start the robotic arm and set the joint motion parameters, including the starting position, end position and motion speed; calibrate the binocular camera;

[0060] In this embodiment, an experimental site that meets the experimental requirements must be selected. This experimental site is 5×5×8 meters, ensuring sufficient space for activities and good environmental conditions. Four ultra-wideband anchor module support frames are arranged in the experimental site to ensure that the entire experimental area is covered, so as to achieve comprehensive signal coverage and data acquisition. The six-degree-of-freedom robotic arm system is moved to a fixed position by controlling the universal wheels. Each module in the device must be given a stable power supply to ensure the stability and continuity of the equipment operation. When the equipment needs to be started, the control panel of the device can be used to achieve precise adjustment of the corresponding joint movement to complete various complex six-degree-of-freedom motion simulations. During the entire movement process, the control module, with its efficient data acquisition capabilities, transmits the real-time captured motion trajectory data to the computer in a high-speed, low-latency manner, providing data support for subsequent trajectory analysis.

[0061] Here, the robotic arm's starting position is 3 meters perpendicular to the movable base, and its final position is the ground. According to the experimental requirements, the robotic arm is required to execute a specific motion trajectory: first maintaining an initial vertical upward posture, then performing a horizontal projectile motion at a set initial velocity of 1 m / s. The control module's sampling frequency is set to 100 Hz. The sampling frequency setting determines the number of data acquisitions per unit time.

[0062] S2: In the process of manipulator movement, inertial measurement data and ultra-wideband positioning results are fused in real time through indirect Kalman filtering. The state vector is defined as , the indirect Kalman filter observation residual equation is defined as ;in, Indicates the current ultra-wideband positioning result, Indicates the next moment position predicted based on the current inertial measurement data, that is, the inertial measurement predicted position before correction. is the current inertial measurement noise;

[0063] Taking inertial measurement data as control input, the prior error state estimate is calculated through the state transfer equation. The posterior error state estimate is calculated based on the prior error state estimate and the observation residual equation. The inertial measurement data is updated according to the accelerometer bias error, gyroscope bias error and attitude angle error in the posterior error state estimate. The final posture is calculated using the updated inertial measurement data to form the motion trajectory measurement value of each joint.

[0064] S3: Use the calibrated binocular camera to obtain the target distance information from the checkerboard calibration plate in real time during the movement of the robot arm to obtain the true value of the motion trajectory of the robot arm's end joint, and infer the true value of the motion trajectory of the remaining joints;

[0065] S4: Complete the calibration of the joint motion parameters of the robotic arm based on the measured value of each joint motion trajectory and the true value of the motion trajectory.

[0066] In the present invention, during the movement of the robotic arm, the robotic arm is moved horizontally by a movable base plate.

[0067] Introduce the principles of inertial measurement positioning and ultra-wideband positioning:

[0068] Among them, the inertial measurement unit model is:

[0069]

[0070]

[0071] in, and are the acceleration and angular velocity of the inertial measurement unit, respectively; and are the real acceleration and angular velocity respectively; is the attitude quaternion The corresponding rotation matrix; and are the bias of the accelerometer and gyroscope respectively; and are the noise of the accelerometer and gyroscope, respectively.

[0072] Attitude and position prediction based on current inertial measurement data:

[0073]

[0074]

[0075]

[0076] in, represents quaternion multiplication, is the quaternion index mapping, is the angular velocity value currently measured by the gyroscope, is the current gyroscope bias, is the time interval, is the current attitude quaternion, is the predicted value of the attitude quaternion at the next moment, is the rotation matrix corresponding to the current posture quaternion, is the acceleration value currently measured by the accelerometer, is the current accelerometer bias, is the gravity vector, is the current speed, is the current location, Predict the position for the next moment.

[0077] The ultra-wideband model is as follows:

[0078]

[0079] in, The coordinates measured for the UWB module antenna, are the real coordinates, This is ultra-wideband measurement noise. During ultra-wideband positioning, the ultra-wideband master antenna on the manipulator acts as a mobile signal source, actively transmitting pulse signals to the ultra-wideband anchor antenna in the ultra-wideband anchor module. The ultra-wideband anchor antenna, acting as a spatial reference point, receives the signal transmitted by the ultra-wideband master antenna on the manipulator. The first control module calculates the distance information, which is then sent back to the ultra-wideband master antenna. The second control module then calculates the ultra-wideband positioning result.

[0080] The collected data is fused using indirect Kalman filtering to achieve accurate calculation of the motion trajectory, which can improve the overall measurement accuracy and reliability. In a specific implementation of the present invention, a process for real-time fusion of inertial measurement data and ultra-wideband positioning results is as follows:

[0081] (1) State space modeling

[0082] Define the state vector: ,in, is the error state vector, is the accelerometer bias error in the inertial measurement unit, is the gyroscope bias error in the inertial measurement unit, Different from the traditional direct pose estimation, this paper indirectly improves the accuracy by estimating the sensor error and solves the IMU integral drift problem.

[0083] (2) Prediction stage

[0084] Predict the pose at the next moment based on the inertial measurement data at moment k, including:

[0085] Posture Update:

[0086]

[0087] Speed ​​update:

[0088]

[0089] Location Updates:

[0090]

[0091] (3) Correction of inertial measurement data

[0092] State transition equation:

[0093]

[0094] Linearized state transfer equation:

[0095]

[0096] Covariance prediction:

[0097]

[0098] Define the observation residual equation:

[0099]

[0100] Using predicted observations based on prior states Linearize the observation matrix:

[0101]

[0102]

[0103] Kalman gain and update:

[0104]

[0105]

[0106]

[0107] in, represents the prior error state estimate, represents the posterior error state estimate, represents the Kalman gain value, represents the observation matrix, is the ultra-wideband measurement noise The covariance matrix of represents the nonlinear state transfer equation, represents the state vector at time k-1, represents the k-1 moment process noise, Is the control input, which is the original measurement value of IMU and ; is the prior error covariance matrix, is the posterior error covariance matrix of the previous moment, is the process noise covariance matrix, is the observation residual, Indicates the current ultra-wideband positioning result, Indicates the next moment position predicted based on the current inertial measurement data, that is, the inertial measurement predicted position before correction.

[0108] In this embodiment, the updating of the inertial measurement data refers to adding the accelerometer bias and gyroscope bias in the inertial measurement data to the accelerometer bias error and gyroscope bias error respectively to obtain the updated accelerometer bias and gyroscope bias; and updating the attitude quaternion in the inertial measurement data to obtain the updated attitude quaternion. Perform quaternion multiplication as the updated attitude quaternion; where, represents the attitude angle error in the posterior error state estimate.

[0109] Specifically, the bias correction formula is:

[0110]

[0111]

[0112] The posture correction formula is:

[0113]

[0114] Substitute the corrected bias and posture into the prediction stage formula and finally output the corrected position and posture .

[0115] Figure 5 This is the result diagram of the motion trajectory of the end of the robotic arm in this embodiment. The black line is the result after correction by the present invention, which is very close to the true value represented by the green line, indicating that the present invention effectively overcomes equipment errors and environmental interference and significantly improves positioning accuracy and stability; the red line is the result of pure inertial measurement data before correction, which has obvious drift.

[0116] In a specific implementation of the present invention, the process of calibrating a binocular camera includes:

[0117] Fix the checkerboard calibration plate on the experimental scene substrate;

[0118] Keep the base joint connecting the bottom of the robot arm to the base plate fixed, drive the remaining joints to perform several sets of posture transformations and synchronously collect calibration plate images;

[0119] Zhang's calibration method is used to solve the spatial transformation matrix of the binocular camera relative to the robotic arm base coordinate system.

[0120] Here, high-precision calibration effectively eliminates positioning errors caused by factors such as equipment installation errors, lens distortion, and ambient light interference, significantly improving measurement accuracy and stability. Specifically, during calibration, a checkerboard calibration plate is positioned within the binocular camera's field of view, allowing it to be clearly seen through the camera. While capturing several sets of images, the calibration plate and the manipulator base remain stationary, while the remaining joints are controlled to perform 20 different pose transformations, simultaneously capturing images of the calibration plate and joint angle data. Based on Zhang's calibration method, the fixed pose matrix of the binocular camera in the base coordinate system is calculated using image feature points and joint angle data. This matrix fully represents the camera's three-dimensional position and spatial orientation. This camera pose is then calculated using a forward kinematic model, spatially mapping the end-point center coordinates and Euler angles to determine the rigid transformation matrix from the camera to the end-point of the manipulator. To ensure calibration quality, the entire process rigorously ensures that the calibration plate covers at least 80% of the camera's field of view and that data acquisition is strictly synchronized. After calibration, the robot arm can obtain real-time 3D information of the target through the binocular camera during movement, and use the transformation matrix to achieve accurate ranging in the mechanical coordinate system. The ranging result here will serve as the true value of the motion trajectory of the robot arm's end joint, and the true value of the motion trajectory of the remaining joints will be inferred.

[0121] The above examples are merely specific embodiments of the present invention. Obviously, the present invention is not limited to the above examples, and many variations are possible. All variations that can be directly derived or imagined by a person skilled in the art from the disclosure of the present invention should be considered to be within the scope of protection of the present invention.

Claims

1. A calibration device for a robotic arm system for spatial motion trajectory simulation, characterized in that: include: The ultra-wideband positioning network consists of several ultra-wideband anchor modules arranged on a substrate in the experimental scene. Each ultra-wideband anchor module includes a retractable bracket, an ultra-wideband anchor antenna mounted thereon, and a first control module. The ultra-wideband anchor antenna is unobstructed and provides full-area signal coverage. A six-degree-of-freedom robotic arm system, comprising a movable base plate, a control panel, and a robotic arm consisting of six joints connected in series, each of which integrates an ultra-wideband master control antenna, an inertial measurement unit, and a second control module; The binocular vision module is installed on the end joint of the robotic arm and establishes a spatial association with the robotic arm coordinate system through a rigid connection; The multi-source data fusion module is used to fuse the acceleration, angular velocity, attitude quaternion collected by the inertial measurement unit with the ultra-wideband positioning results. Its state vector is defined as ,in is the error state vector, is the accelerometer bias error in the inertial measurement unit, is the gyroscope bias error in the inertial measurement unit, is the attitude angle error; The calibration module includes a checkerboard calibration plate and a motion parameter calculation module. The checkerboard calibration plate is fixed on the experimental scene substrate to ensure that it is always within the visual range of the binocular vision module to realize the real-time ranging function of the binocular vision module; the motion parameter calculation module is used to complete the joint motion parameter calibration of the robot arm based on the measured value of each joint motion trajectory and the true value of the motion trajectory.

2. The robotic arm system calibration device for spatial motion trajectory simulation according to claim 1, characterized in that: The ultra-wideband anchor antenna is fixed to the top of the retractable bracket through a pan-tilt platform, and the pan-tilt platform has an azimuth adjustment function.

3. The robotic arm system calibration device for spatial motion trajectory simulation according to claim 1, characterized in that: The sampling frequency of the UWB master antenna and the inertial measurement unit is the same.

4. The robotic arm system calibration device for spatial motion trajectory simulation according to claim 1, characterized in that: The number of ultra-wideband anchor modules shall not be less than 3.

5. A calibration method for a robotic arm system calibration device based on the spatial motion trajectory simulation according to any one of claims 1 to 4, characterized in that: The following steps are involved: Step 1: Arrange an ultra-wideband positioning network around the motion scene substrate, start the robotic arm and set the joint motion parameters, including the starting position, end position, and motion speed; calibrate the binocular camera; Step 2: In real time, the inertial measurement data and ultra-wideband positioning results are fused through indirect Kalman filtering during the movement of the manipulator. The state vector is defined as , the indirect Kalman filter observation residual equation is defined as ;in, Indicates the current ultra-wideband positioning result, Indicates the next moment position predicted based on the current inertial measurement data, that is, the inertial measurement predicted position before correction. is the current inertial measurement noise; Using inertial measurement data as control input, the prior error state estimate is calculated through the state transfer equation. The a posteriori error state estimate is calculated based on the prior error state estimate and the observation residual equation. The inertial measurement data is updated based on the accelerometer bias error, gyroscope bias error, and attitude angle error in the a posteriori error state estimate. The final posture is calculated using the updated inertial measurement data to form the motion trajectory measurement value of each joint. Step 3: Use the calibrated binocular camera to obtain the target distance information from the checkerboard calibration plate in real time during the movement of the robot arm to obtain the true value of the motion trajectory of the robot arm's end joint, and infer the true value of the motion trajectory of the remaining joints; Step 4: Complete the calibration of the joint motion parameters of the robotic arm based on the measured values ​​of each joint motion trajectory and the true value of the motion trajectory.

6. The calibration method of the robotic arm system calibration device for spatial motion trajectory simulation according to claim 5, characterized in that: The formulas for the prior error state estimate and the posterior error state estimate are as follows: ； ； in, represents the prior error state estimate, represents the posterior error state estimate, represents the nonlinear state transfer equation, represents the state vector at time k-1, represents the k-1 moment process noise, represents the current Kalman gain value, represents the observation matrix.

7. The calibration method of the robotic arm system calibration device for spatial motion trajectory simulation according to claim 5, characterized in that: The updating of the inertial measurement data refers to adding the accelerometer bias and gyroscope bias in the inertial measurement data to the accelerometer bias error and gyroscope bias error respectively to obtain the updated accelerometer bias and gyroscope bias; and Perform quaternion multiplication as the updated attitude quaternion; where, represents the attitude angle error in the posterior error state estimate.

8. The calibration method of the robotic arm system calibration device for spatial motion trajectory simulation according to claim 5, characterized in that: The process of calibrating a binocular camera includes: Fix the checkerboard calibration plate on the experimental scene substrate; Keep the base joint connecting the bottom of the robot arm to the base plate fixed, drive the remaining joints to perform several sets of posture transformations and synchronously collect calibration plate images; Zhang's calibration method is used to solve the spatial transformation matrix of the binocular camera relative to the robotic arm base coordinate system.

9. The calibration method of the robotic arm system calibration device for spatial motion trajectory simulation according to claim 5, characterized in that: The process of obtaining ultra-wideband positioning results includes: The ultra-wideband main control antenna on the robotic arm acts as a mobile signal source, actively transmitting pulse signals to the ultra-wideband anchor antenna in the ultra-wideband anchor module; the ultra-wideband anchor antenna acts as a spatial position reference point, receiving the signal transmitted by the ultra-wideband main control antenna on the robotic arm, calculating the distance information in the first control module, and sending the calculated distance information back to the ultra-wideband main control antenna, and calculating the ultra-wideband positioning result in the second control module.

10. The calibration method of the robotic arm system calibration device for spatial motion trajectory simulation according to claim 1, characterized in that: During the movement of the robotic arm, the horizontal movement of the robotic arm is achieved through the movable base plate.

Images