Multi-task cooperative calibration method and device applied to robot and related equipment

Abstract

The disclosure provides a multi-task cooperative calibration method and device applied to a robot and related equipment, and relates to the technical field of robots, in particular to the technical field of robot perception and control. The specific implementation scheme is: generating a control instruction of a mechanical arm of a robot based on motion constraint information; the motion constraint information is used to control the mechanical arm to cover multiple postures in the motion process; in the process that the robot controls the motion of the mechanical arm in response to the control instruction, collecting respective preset data of at least two calibration tasks; and performing corresponding calibration operations based on the preset data of each calibration task.

Description

Technical Field

[0001] This disclosure relates to the field of robotics, and in particular to the fields of robot perception and control. Background Technology

[0002] To improve the accuracy of robot interaction with the environment, it is usually necessary to perform various calibration tasks on the robot, such as hand-eye calibration, tool calibration, and sensor calibration.

[0003] In a typical vision-guided robot system, sensor calibration aims to establish a precise mapping model between sensor measurement data and the physical world, thereby ensuring the accuracy of the perceived data. The core task of hand-eye calibration is to establish a geometric relationship between the visual perception coordinate system (such as cameras and tracking devices) and the robot's motion coordinate system (such as the base and end flange) within the robot system, enabling the visually observed target pose to be converted into executable motion commands for the robot. Tool calibration is used to establish a fixed transformation relationship between the tool's working point / coordinate system and the robot's end flange, allowing the robot to precisely manipulate the tool's end to perform tasks. These three aspects work together to form the geometric alignment foundation for the entire chain of robot operation, from visual perception and localization to manipulation, and are a prerequisite for the system to achieve high-precision operation.

[0004] Therefore, calibration tasks are important and essential, and how to improve the execution efficiency of calibration tasks is a concern in the industry. Summary of the Invention

[0005] This disclosure provides a multi-task collaborative calibration method, apparatus, and related equipment for robots to solve or alleviate one or more technical problems in the prior art.

[0006] Firstly, this disclosure provides a multi-task cooperative calibration method for robots, including: Based on motion constraint information, control commands for the robot's robotic arm are generated; the motion constraint information is used to control the robotic arm to cover multiple postures during movement. During the process of the robot responding to the control command to control the movement of the robotic arm, preset data for at least two calibration tasks are collected respectively. Based on the preset data for each calibration task, perform the corresponding calibration operations.

[0007] Secondly, this disclosure provides a multi-task collaborative calibration device for robots, comprising: The generation module is used to generate control commands for the robot's robotic arm based on motion constraint information; the motion constraint information is used to control the robotic arm to cover multiple postures during movement. The data acquisition module is used to acquire preset data for at least two calibration tasks during the process of the robot controlling the movement of the robotic arm in response to the control command. The execution module is used to perform corresponding calibration operations based on the preset data of each calibration task.

[0008] Thirdly, an electronic device is provided, comprising: At least one processor; and The memory is communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the methods of any embodiment of the present disclosure.

[0009] Fourthly, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause the computer to perform a method according to any embodiment of the present disclosure.

[0010] Fifthly, a computer program product is provided, including a computer program that, when executed by a processor, implements the method according to any embodiment of the present disclosure.

[0011] Sixthly, a multi-task calibration robot system is provided, comprising: A robotic arm with a tool and a tracking reference device mounted at its end; Tracking device, used to acquire the pose of the tracking reference device; and The controller includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the multi-task collaborative calibration method as described above.

[0012] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0013] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the various drawings denote the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings depict only some embodiments provided according to this disclosure and should not be construed as limiting the scope of this disclosure.

[0014] Figure 1 This is a schematic diagram of a multi-task cooperative calibration method for robots according to an embodiment of the present disclosure; Figure 2This is a flowchart illustrating a multi-task collaborative calibration method for robots according to an embodiment of the present disclosure; Figure 3 This is a schematic diagram of a multi-task collaborative calibration method for robots according to an embodiment of the present disclosure; Figure 4 This is a schematic diagram of a second motion trajectory according to an embodiment of the present disclosure; Figure 5 This is a schematic diagram of a robotic arm according to an embodiment of the present disclosure; Figure 6 This is a schematic diagram of a multi-task collaborative calibration device for robots according to an embodiment of the present disclosure; Figure 7 This is a block diagram of an electronic device used to implement the multi-task cooperative calibration method for robots according to the embodiments of this disclosure. Detailed Implementation

[0015] The present disclosure will now be described in further detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0016] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0017] The terms “first,” “second,” etc., used in this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion, such as including a series of steps or units. A method, system, product, or apparatus is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or apparatuses.

[0018] In related technologies, the calibration targets and methods differ for different calibration tasks of robots. Therefore, different calibration tasks need to be executed independently. This results in long calibration times and low calibration efficiency when multiple calibration tasks are involved. In addition, each calibration task requires the robotic arm to perform corresponding pose operations, which also leads to resource waste.

[0019] In view of this, this disclosure provides a multi-task collaborative calibration method for robots. This calibration method, by rationally designing the robot's manipulator's movement path, can collect pose data required for at least two calibration tasks during a single manipulator movement, thereby reducing resource waste caused by repetitive manipulator movements and improving calibration efficiency.

[0020] For ease of understanding, key terms used in this disclosure are explained, including: Hand-eye calibration establishes a geometric relationship between a visual perception coordinate system (such as a camera or tracking device) and a robot motion coordinate system (such as a base or end effector flange) within a robotic system. This relationship can be achieved in various ways. A fixed transformation relationship between the tracking device coordinate system (such as a camera) and the robot base (or end effector) coordinate system can be solved. The hand-eye calibration disclosed herein can also establish a transformation relationship between the end effector target and the robot end effector flange by defining a fixed spatial relationship. The relationship established by hand-eye calibration is a purely mechanical rigid connection, which remains constant during robot operation and is independent of the installation positions of the tracking device and the robotic arm base. Therefore, when the robotic arm is used to perform surgical procedures, the surgeon can freely adjust the positions of the tracking device and the robotic arm carriage during surgery without requiring system recalibration; the robot end effector pose can be calculated simply by the tracking device sensing the end effector target in real time.

[0021] Tool calibration is used to determine the fixed transformation relationship between the tool's operating position or operating coordinate system and a trackable reference marker on the tool (such as the first sub-tracking reference device mentioned later). By acquiring the pose of the reference marker in real time through tracking equipment, the spatial pose of the tool's operating position or operating coordinate system can be indirectly and accurately calculated. Common calibration methods include using precision calibration devices (such as calibration disks) or fitting through multiple points contacting the same fixed point.

[0022] Force sensor calibration requires zero-point calibration (tare) before use to eliminate inherent zero-point bias caused by circuitry and temperature. Furthermore, to eliminate the influence of tool weight on force sensor readings under different postures, load gravity compensation is necessary. Load gravity compensation requires obtaining parameters such as the tool's mass and center of gravity position, a process that can be achieved through the force sensor calibration described in this disclosure.

[0023] Angle Set: In this disclosure, the attitude of the robotic arm's end effector flange is described using Euler angles. Based on the robot's base coordinate system, the pitch angle is the angle of rotation of the end effector flange around the Y-axis of the base coordinate system, used to describe the end effector flange's head-up or head-down motion. The yaw angle is the angle of rotation of the end effector flange around the Z-axis of the base coordinate system, used to describe the end effector flange's rotation around its own axis. The roll angle is the angle of rotation of the end effector flange around the X-axis of the base coordinate system, used to describe the end effector flange's lateral tilting motion.

[0024] Range of angle variation: To avoid ambiguity, "range of variation" in this disclosure specifically refers to the absolute value of the difference between the maximum and minimum values ​​of the angle within the same motion trajectory, rather than the amount of change between the starting point and the ending point.

[0025] To facilitate understanding, the structure of the robot involved in the embodiments of this disclosure will first be described by way of example. It is understood that... Figure 1 The robot structure described herein is only for assisting in understanding the solutions provided in the embodiments of this disclosure and is not intended to limit the scope of the invention. Figure 1 As shown, the end effector of the robot's robotic arm 101 includes an end flange 102; a force sensor 103 is mounted on the end flange 102; a tool 104 is mounted on the other end of the force sensor 103 away from the robotic arm; the tool 104 is connected to a tracking reference device 105. To complete the calibration task, a tracking device 106 is needed to collect data for the corresponding calibration task. The tracking device 106 can be fixedly mounted on the robotic arm or fixed to a separate display screen trolley in the external environment; this disclosure does not limit this.

[0026] In practice, the matching of the tracking device and the tracking reference device follows specific physical principles and signal interaction mechanisms. Robot calibration can be performed using either visual navigation or non-visual navigation. For example, for visual navigation, optically identifiable markers such as reflectors, light-emitting points, reflective spheres, or specific patterns can be used as tracking reference devices, with the corresponding tracking device being a visual sensor (such as a camera). Non-visual navigation methods can include electromagnetic navigation, ultrasonic navigation, laser navigation, or other devices capable of providing six-degree-of-freedom pose measurement. For instance, an electromagnetic transmitter can be used as a tracking device, with an electromagnetic sensor as its corresponding reference device; ultrasonic tracking devices are typically paired with an ultrasonic transmitter or a dedicated reflective target, using ultrasonic time-of-flight for ranging and positioning; laser tracking devices require a laser reflective target (such as a cooperative target or prism), achieving high-precision tracking through laser interferometry and angle measurement. The core function of each tracking reference device is to be stably and accurately detected and analyzed by its corresponding type of tracking device, thus forming a complete six-degree-of-freedom measurement chain.

[0027] In this embodiment of the disclosure, a multi-task cooperative calibration method for robots is proposed, such as... Figure 2 As shown, it includes the following: S201, based on motion constraint information, generates control commands for the robot's robotic arm; the motion constraint information is used to control the robotic arm to cover multiple postures during movement.

[0028] The motion constraint information is a set of parameters describing the pose changes that the robot's end effector flange needs to satisfy during its movement. By designing the motion constraint information, the robotic arm can simultaneously meet the pose and data requirements of different calibration tasks during a single motion process.

[0029] Control commands are sequences of joint angles or end-effector trajectory points calculated based on motion constraint information, which drive the robot to move along a predetermined path.

[0030] The specific requirements for the posture covered by the robotic arm are determined based on the specific calibration task to be completed. These requirements are all included in the motion constraint information.

[0031] S202, during the process of the robot responding to control commands to control the movement of the robotic arm, collect preset data for at least two calibration tasks.

[0032] Since different calibration tasks require different data content and the calibration calculation process is different, the preset data is the data that needs to be collected in advance for different calibration tasks.

[0033] S203, based on the preset data of each calibration task, performs the corresponding calibration operation.

[0034] After collecting preset data for at least two calibration tasks, the system automatically selects and executes the corresponding calibration operation based on the type of data collected and the pre-set task identifier of the calibration task, so as to complete the calibration operation of multiple calibration tasks simultaneously.

[0035] In this embodiment, motion constraint information is used to control the robotic arm to cover the poses required for different calibration tasks during the same motion process, so that preset data for multiple calibration tasks can be collected simultaneously during one motion. Based on the collected preset data, corresponding calibration operations are performed to achieve collaborative calibration of multiple calibration tasks. Compared with the calibration methods in related technologies where each calibration task is executed independently and step by step, the method provided by this embodiment does not require repeatedly planning multiple dedicated trajectories for the robotic arm (each dedicated trajectory completes one calibration task), nor does it require multiple start-stop operations of the robot or manual switching of calibration modes. This can significantly shorten the total calibration time, improve calibration efficiency, and save equipment resources.

[0036] In some embodiments, at least two calibration tasks include at least two of the following: hand-eye calibration task, tool calibration task, and force sensor calibration task; In order for the robotic arm to cover multiple postures during its movement, these multiple postures include the tracking reference device being within a preset sensing range of the tracking device required to perform at least two calibration tasks, and the variation range of at least one target angle from the following set of angles of the robotic arm's end flange being greater than or equal to the corresponding preset angle: The set of angles includes: pitch angle, roll angle, and yaw angle.

[0037] In this embodiment, by constraining the tracking reference device to always remain within the preset sensing range of the tracking device, the continuous validity of the collected data can be achieved, reducing the problem of data quality degradation caused by the tracking reference device going out of the preset sensing range, thereby improving the accuracy of the calibration results. By controlling the range of angle changes in the angle set, it is beneficial to meet the calibration requirements of various calibration tasks, so that the robotic arm can collect data that meets different calibration tasks during the same movement, and complete multiple calibration tasks simultaneously.

[0038] During implementation, the preset angle requirements corresponding to the aforementioned target angle shall be set in accordance with the requirements of the corresponding calibration task, such as the following conditions as required by the task: Condition 1) When at least two calibration tasks include a force sensor calibration task, the preset angle corresponding to the pitch angle variation range is greater than or equal to the first target angle. The first target angle is the minimum threshold of the pitch angle variation range required for the force sensor calibration task. Only when the pitch angle variation is large enough can the projection of the gravity vector in the force sensor's coordinate system change sufficiently, thus enabling accurate force sensor calibration.

[0039] Condition 2) If at least two calibration tasks include hand-eye calibration tasks, at least one of the following conditions must be met: The preset angle corresponding to the range of roll angle variation is greater than or equal to the second target angle. The preset angle corresponding to the range of yaw angle variation is greater than or equal to the angle of the third target; The preset angle corresponding to the range of pitch angle variation is greater than or equal to the fourth target angle.

[0040] The second target angle is the minimum threshold for the range of roll angle variation in the hand-eye calibration task. Roll motion can provide different angles to enhance the diversity of preset data required for the hand-eye calibration task, thereby improving the calibration accuracy of the hand-eye calibration task.

[0041] The third target angle is the minimum threshold of the yaw angle variation range in the hand-eye calibration task.

[0042] The fourth target angle is the minimum threshold for the range of pitch angle variation in the hand-eye calibration task.

[0043] In the force sensor calibration task, the first target angle can be greater than the second, third, or fourth target angle in the hand-eye calibration task, so as to allow the pitch angle to vary over a wide range and meet the pitch attitude change requirements of the force sensor.

[0044] Condition 3) If at least two calibration tasks include hand-eye calibration and force sensor calibration, then the preset angle corresponding to the pitch angle variation range is greater than or equal to the first target angle, the preset angle corresponding to the roll angle variation range is greater than or equal to the second target angle, and / or the preset angle corresponding to the yaw angle variation range is greater than or equal to the third target angle.

[0045] Among them, provided that the pitch angle is greater than or equal to the first target angle, at least one of the roll angle and yaw angle must meet the corresponding requirements.

[0046] The first target angle is greater than the second or third target angle. Since tool calibration only requires that the tracking reference device has multiple pose changes in the angle set within the preset sensing range of the tracking equipment, condition 3) can adapt to the data acquisition conditions of three calibration tasks.

[0047] The most preferred condition in condition 3) is that the preset angle corresponding to the pitch angle variation range is greater than or equal to the first target angle, the preset angle corresponding to the roll angle variation range is greater than or equal to the second target angle, the preset angle corresponding to the yaw angle variation range is greater than or equal to the third target angle, and the first target angle is greater than the second or third target angle. This fully satisfies the requirements of the three calibration methods.

[0048] The tracking reference device includes a first sub-tracking reference device on a tool mounted at the end effector of a robotic arm, and / or a second sub-tracking reference device independent of the tool. Taking visual navigation as an example, the first sub-tracking reference device can be a visual marker on the tool, such as an end-effector fixed to the tool. This end-effector typically consists of a set of reflective sheets, reflective spheres, luminous dots, or high-contrast patterns with a specific geometric arrangement. The second sub-tracking reference device can be a visual marker required for tool positioning, such as pinpoint positioning, for example, a calibration disk. A calibration disk is a precisely sized and shaped circular plate with multiple positioning marks arranged in a specific geometric relationship, or with a specific pattern that is orientation-specific. The most common form of positioning marks is an array of circular reflective sheets, and the most common form of orientation-specific patterns is a QR code.

[0049] The tracking reference device needs to be within the preset sensing range of the tracking equipment. This preset sensing range is the spatial area (such as the camera's field of view, the effective range of the electromagnetic field, etc.) that the tracking equipment can stably and accurately identify. During the calibration process, it is necessary to ensure that the tracking reference device remains within this range to ensure the validity of the collected data and to guarantee the calibration accuracy.

[0050] The angle set consists of three angles used to describe the attitude of the end flange. In implementation, all angles in the angle set can be defined in the robot base coordinate system. In this embodiment, the robot base coordinate system is defined as follows: the Z-axis is perpendicular to the robot base mounting surface (vertically upward), and the X and Y axes are in the horizontal plane, as shown below. Figure 3 As shown, the horizontal plane (base XY plane) refers to the plane defined by the X-axis and Y-axis in the robot base coordinate system.

[0051] In this embodiment of the disclosure, for force sensor calibration and hand-eye calibration, by precisely controlling the threshold range of pitch angle, roll angle, and yaw angle, large-angle pitch motion and small-amplitude roll / yaw motion are integrated into the same trajectory, so as to accurately adapt to the attitude requirements of these two calibration tasks.

[0052] In some embodiments, control commands instruct the robotic arm to perform unidirectional pitch movements, such as "head up" or "head down" operations. Considering that patients or other medical equipment are usually located below the robotic arm in actual surgical or operational scenarios, the head-up movement mode is preferred in practice to avoid collisions or interference. It should be noted that the specific motion parameters of the head-up movement mode (e.g., the range and rate of pitch angle change) are not fixed, but can be adaptively designed according to the specific needs of the calibration task to be completed (e.g., force sensor calibration, hand-eye calibration).

[0053] In some implementations, the head-up motion mode can be implemented as follows: control commands are used to control the robotic arm to execute a first motion trajectory, which includes a first trajectory segment and a second trajectory segment; in the first trajectory segment, the pitch angle of the end flange moves from a first angle to a second angle; in the second trajectory segment, the pitch angle of the end flange moves from the second angle to a third angle; the angle difference between the first angle and the third angle is greater than or equal to a preset angle corresponding to the pitch angle. Thus, the head-up motion mode is adaptable to force sensor calibration and tool calibration. Of course, to further better adapt to hand-eye calibration, in at least a portion of the first trajectory segment, the yaw angle and / or roll angle of the end flange continuously change, and the range of change is greater than or equal to the corresponding preset angle. Therefore, the head-up motion mode can adapt to the data acquisition conditions of hand-eye calibration and force sensor calibration, and also meets the data acquisition conditions of the three calibration tasks.

[0054] In practice, the preset angles for each angle in the head-up motion mode can be determined based on the actual calibration task combination. To meet the force sensor calibration task, the preset angles corresponding to the pitch angle in the head-up motion mode are set according to condition 1) mentioned above. To meet the hand-eye calibration task, each preset angle is set according to condition 2) mentioned above. If tool calibration is included, at least the tracking reference device must have multiple pose changes in the angle set within the preset sensing range of the tracking device. To simultaneously meet both force sensor calibration and hand-eye calibration tasks, each preset angle is set according to condition 3) mentioned above.

[0055] In some embodiments, force sensor calibration requires a wide range of pitch angle variations without relying on the measurement accuracy of the tracking device. Therefore, the angle difference between the first and third angles must be greater than or equal to a preset angle corresponding to the pitch angle to meet the calibration requirements of the force sensor, which necessitate a wide range of pitch angles. Thus, force sensor calibration can fully utilize data from the entire motion trajectory.

[0056] In some embodiments, the hand-eye calibration task requires the tracking device to provide high-precision pose measurement, and also requires multiple pose changes. Therefore, the data collected in the first trajectory segment is used as the data required for the hand-eye calibration task. At the same time, the yaw angle and / or roll angle of the end flange continuously change to provide the multiple pose changes required for the hand-eye calibration task.

[0057] In this embodiment, by controlling the pitch angle change in segments, the motion trajectory exhibits different motion characteristics during periods of smaller and larger pitch angles. This facilitates differentiated processing or selective use of the collected data based on the motion characteristics at different stages. Furthermore, by superimposing continuous small-amplitude rotations of yaw and / or roll during the pitch angle change process, the end flange undergoes both pitch and rotational motion simultaneously. This allows for the simultaneous acquisition of a wide range of attitude changes and small-amplitude attitude disturbances within a single continuous motion, enriching the attitude diversity of the end flange during the motion. In summary, the same motion trajectory can simultaneously provide multiple attitude changes of different amplitudes and dimensions, laying the foundation for subsequent calibration tasks with different requirements using the data collected from this trajectory. This eliminates the need to plan multiple independent trajectories to meet different attitude requirements, thereby improving data acquisition efficiency.

[0058] In addition to the aforementioned head-up motion mode, in this embodiment of the present disclosure, the control command is used to control the robotic arm to execute a second motion trajectory, which conforms to a spatial curve motion mode. The spatial curve motion mode means that the motion path of the end flange in three-dimensional space is composed of at least one continuous spatial line, such as a smooth curve segment and / or a broken line formed by connecting multiple straight line segments end to end.

[0059] To facilitate control of the robotic arm's movement, this spatial curve motion mode can be a polyhedral motion mode.

[0060] For example, the polyhedral motion mode controls the robotic arm to perform polyhedral operations based on control commands. Specifically: the second motion trajectory in the polyhedral motion mode includes a third trajectory segment and a fourth trajectory segment; the third trajectory segment includes at least one first side; the fourth trajectory segment includes at least one second side; when multiple first sides are included, the length of each first side follows the definition of the side length of the polyhedron; when multiple second sides are included, the length of each second side follows the definition of the side length of the polyhedron; in the third trajectory segment, the end flange performs a translation operation along the first side, and simultaneously, the pitch angle change of the end flange is greater than a first preset angle; in the fourth trajectory segment, the end flange performs a translation operation along the second side, and simultaneously, the pitch angle change of the end flange is greater than a second preset angle; in the second motion trajectory, the range of pitch angle change of the end flange is greater than or equal to the preset angle corresponding to the pitch angle; thus, the spatial curve motion mode can be adapted for force sensor calibration and tool calibration. Of course, in order to better adapt to hand-eye calibration, the yaw angle and / or roll angle of the end flange continuously change in at least a portion of the second motion trajectory, and the range of change is greater than or equal to the corresponding preset angle. Therefore, the spatial curve motion mode can adapt to the data acquisition conditions of hand-eye calibration and force sensor calibration, and also meets the data acquisition conditions of the three calibration tasks.

[0061] Similarly, during implementation, the set angles of each angle in the polyhedral motion mode can be determined according to the actual calibration task combination. To meet the force sensor calibration task, the preset angles corresponding to the pitch angle in the polyhedral motion mode are set according to the aforementioned condition 1). To meet the hand-eye calibration task, each preset angle is set according to the aforementioned condition 2). If tool calibration is included, at least the tracking reference device must be within the preset sensing range of the tracking device, and the angles in the angle set must have multiple pose changes. If both force sensor calibration and hand-eye calibration tasks are met simultaneously, each preset angle is set according to the aforementioned condition 3).

[0062] The third trajectory segment is the first set of edges in the polyhedral trajectory. In the third trajectory segment, the end flange performs a translation operation along the first edge, while the pitch angle continuously changes, and the range of pitch angle change is greater than the first set angle.

[0063] The fourth trajectory segment is the subsequent edge in the polyhedral trajectory. In this segment, the end flange performs a translation operation along the second edge, while the pitch angle continuously changes, and the range of pitch angle change is greater than the second set angle.

[0064] For example, taking a cuboid as an example, the second motion trajectory is as follows: Figure 4As shown, the robotic arm's end effector first performs a lifting motion, which is the preparatory trajectory segment, preparing for the subsequent polyhedral motion. The third trajectory segment is then executed, which includes the two first edges: L1 and L2.

[0065] On side L1, the robotic arm performs a translation operation (e.g., moves in a straight line along the direction of the arrow on L1), while the pitch angle of the end flange changes more than a first set angle (exemplary end flange pitching up 20°).

[0066] On side L2, continue to perform translation operation, while the pitch angle of the end flange changes more than the first set angle (for example, the end flange tilts down by 40°, and the tilting range of 40° also satisfies the requirement of being greater than the first set angle).

[0067] Throughout the entire trajectory segment, the end flange rotates back and forth by a certain amount, that is, the yaw angle and / or roll angle continuously change, and the range of change is greater than the corresponding preset angle (e.g., 20°, which can be continuously changed back and forth between -10° and 10°).

[0068] Then the fourth trajectory segment is executed, which contains two edges: L3 and L4.

[0069] On side L3, the robotic arm performs a translation operation, while the end flange tilts up by 40° (the pitch angle variation range is greater than the second set angle).

[0070] On side L4, a translation operation is performed, while the pitch angle of the end flange changes within a range greater than the second set angle (for example, the end flange is tilted down by 20°).

[0071] In this fourth trajectory segment, the end flange continuously rotates back and forth by varying the yaw and / or roll angles. The yaw and / or roll angles can also be continuously varied at least in a portion of the third or fourth trajectory segment. After completing L4, the robotic arm descends back to its original position.

[0072] In this embodiment, the polyhedral motion mode allows the end flange to traverse multiple different positions in space, providing rich translational motion characteristics. Furthermore, the total change in pitch angle is distributed across multiple trajectory segments, resulting in a smooth motion process. Additionally, continuous small-amplitude rotations of yaw and / or roll are superimposed during translational motion, enabling the end flange to perform rotational motion while primarily translating and pitching. This allows for the simultaneous acquisition of a wide range of attitude changes, small-amplitude attitude disturbances, and spatial position movements within a single continuous motion. This enables the same motion trajectory to simultaneously provide multiple attitude changes and spatial position changes of different amplitudes and dimensions, facilitating the acquisition of preset data required for different calibration tasks.

[0073] In some embodiments, when both the third trajectory segment and the fourth trajectory segment include at least two edges, any of the following conditions are met: Condition 3) In the intermediate trajectory segment formed by the last first side of the third trajectory segment and the first second side of the fourth trajectory segment, the pitch angle of the end flange varies within a range greater than the preset angle corresponding to the pitch angle. Condition 4) The trajectory formed by the last first side of the third trajectory segment and the trajectory formed by the first second side of the fourth trajectory segment both satisfy the condition that the pitch angle of the end flange varies within a range greater than the preset angle corresponding to the pitch angle.

[0074] During implementation, in the intermediate trajectory segment (continuous motion) formed by the last first side of the third trajectory segment and the first second side of the fourth trajectory segment, the pitch angle variation range is greater than or equal to the preset angle. This means that during the continuous process of transitioning from the third trajectory segment to the fourth trajectory segment, the pitch angle undergoes a sufficiently large continuous change.

[0075] For example, such as Figure 4 As shown, on trajectory L2, the end flange gradually lowers by 20°. On trajectory L3, the end flange continues to gradually lower by 20°. That is, the pitch angle variation range of this intermediate trajectory segment (continuous movement of L2 + L3) is ≥40°. If the first set angle for the force sensor calibration task is 40°, this movement mode can especially satisfy the aforementioned condition 1) to adapt to the force sensor calibration task.

[0076] During implementation, the pitch angle variation range of the end flange on the last side of the third trajectory segment is greater than or equal to the preset angle, and the pitch angle variation range on the first side of the fourth trajectory segment is also greater than or equal to the preset angle. This means that the end flange independently provides a sufficiently large continuous pitch angle variation on these two sides.

[0077] For example, such as Figure 4 As shown, on L2, the end flange gradually lowers by 40°. With a preset angle of 40°, the pitch angle variation range of the last first side of the third trajectory segment in condition 2) is greater than the preset angle. On L3, the end flange gradually raises by 40°, meaning the pitch angle variation range of the first second side of the fourth trajectory segment in condition 2) is also greater than the preset angle. If the first target angle required for force sensor calibration is 40°, then this trajectory meets the data requirements for force sensor calibration.

[0078] It is understood that the aforementioned values ​​are illustrative and this disclosure does not limit them.

[0079] In this embodiment of the present disclosure, a continuous pitch angle change of not less than a preset angle is provided in the transition region between the third trajectory segment and the fourth trajectory segment, thereby realizing pitch angle control within the motion range to meet the data requirements of the calibration task.

[0080] In summary, for the first and second motion trajectories, in order to better meet the data requirements of hand-eye calibration, in at least a portion of the motion trajectory of the robotic arm, the yaw angle and / or roll angle of the end flange continuously change, and the range of change is greater than or equal to the corresponding preset angle.

[0081] After introducing the motion trajectory of the robotic arm, the data acquisition and calibration operations for different calibration tasks will be described below. It should be noted that the multi-task calibration method disclosed herein is not limited to the specific calibration algorithms disclosed in the embodiments (such as end-effector-flange solving, visual calibration tool calibration, gravity compensation calibration, etc.). Any algorithm that can complete calibration using data synchronously acquired during the robotic arm's movement, including but not limited to traditional hand-eye calibration (camera-end-effector), tool tip fitting with multi-point contact fixed points, etc., can be replaced or integrated into this disclosure, as long as the motion trajectory can simultaneously provide the input data requirements for each calibration task.

[0082] 1) Hand-eye calibration In some embodiments, during the process of the robot controlling the movement of the robotic arm in response to control commands, preset data for at least two calibration tasks are collected, including: In step A1, if at least two calibration tasks include a hand-eye calibration task, preset data for the hand-eye calibration task is collected during the first trajectory portion of the robotic arm's movement.

[0083] Within the first trajectory section, preset data for the hand-eye calibration task are collected. The first trajectory section is either the aforementioned first trajectory segment or the third trajectory segment.

[0084] Step A2: When the robotic arm moves to the second trajectory section, stop collecting the preset data for the hand-eye calibration task; the first trajectory section is located before the second trajectory section.

[0085] The second trajectory section is either the aforementioned second or fourth trajectory segment. Specifically, if sufficient data for the hand-eye calibration task has been collected in the first trajectory section, the collection of preset data required for the hand-eye calibration task can be stopped in the second trajectory section to avoid overfitting.

[0086] However, it is understandable that while collecting the preset data corresponding to hand-eye calibration, data from other calibration tasks can still be collected simultaneously to facilitate the data accumulation for those other calibration tasks.

[0087] In this embodiment of the disclosure, by limiting the collection of hand-eye calibration data to the first trajectory portion and stopping the collection in the second trajectory portion, sufficient preset data required for the hand-eye calibration task is obtained, thereby reducing the decrease in accuracy of the hand-eye calibration task caused by overfitting.

[0088] In some embodiments, when the tracking reference device is a visual navigation and positioning device, before generating control commands for the robot's robotic arm based on motion constraint information, if it is detected that the angle between the plane of the tracking reference device and the visual navigation and positioning device in the current posture is greater than a preset sensing angle, an adaptive operation is initiated. Specifically, this can be implemented by performing a preset adaptive operation to adjust the visual navigation and positioning device to the preset sensing angle of the tracking device.

[0089] During implementation, the robotic arm can be controlled to move within a small range, and a small amount of data can be collected during the movement to complete the initial hand-eye calibration operation, so as to determine the approximate position (estimated position) of the visual navigation and positioning device at the end of the robotic arm. Based on this approximate position, the robotic arm is adjusted so that the visual navigation and positioning device is at the preset sensing angle of the tracking device. This preset sensing angle is an angle range defined in 90°.

[0090] Small-scale movements can be defined according to requirements and are not limited here, such as a range of ±n°, where n is a value greater than 0.

[0091] In this embodiment of the disclosure, when calibration is performed based on vision, the tracking reference device is adjusted to the optimal perception angle facing the tracking device through adaptive operation, so that the visual data initially acquired in the subsequent calibration motion has high quality, thereby improving the calibration accuracy of subsequent calibration tasks.

[0092] In cases where at least two calibration tasks include a hand-eye calibration task, the hand-eye calibration task is performed based on the first sub-tracking reference device; The preset data for each sampling point in the hand-eye calibration task includes: The first tracking data is used to record the pose of the first sub-tracking reference device in the tracking device coordinate system; Robotic arm data is used to record the pose of the robotic arm's flange center in the robot's base coordinate system.

[0093] Accordingly, for hand-eye calibration tasks, based on the preset data for each calibration task, corresponding calibration operations are performed, including: Step B1: For the hand-eye calibration task, construct multiple sets of sampling point pairs based on the multiple sampling points of the hand-eye calibration task.

[0094] During implementation, while the robotic arm executes a preset motion trajectory (such as a first motion trajectory and a second motion trajectory), sampling points can be collected simultaneously, and these collected sampling points can be processed into sampling point pairs. Assuming there are N+2 sampling points, each sampling point is denoted as... .in, For the i-th first tracking data, This represents the data for the i-th robotic arm, where 0 < i < N+2, and i and N are positive integers. Each pair of sampling points consists of two distinct sampling points. Typically, adjacent sampling points are selected. Alternatively, sampling point pairs can be constructed by selecting non-adjacent points with significant motion changes; this disclosure does not limit this to any particular method.

[0095] Step B2: Based on the relative motion of each sampling point pair in multiple sets of sampling point pairs, determine the calibration parameters for the hand-eye calibration task.

[0096] Relative motion represents the change in pose of the robot end flange from one sampling point to another.

[0097] The calibration parameters for the hand-eye calibration task are fixed transformations to be solved, including the rotation matrix and translation vector between the first sub-tracking reference device and the flange.

[0098] In some embodiments, determining the calibration parameters for the hand-eye calibration task based on the relative motion of each sampling point pair in multiple sets of sampling point pairs can be implemented as follows: Given AX = XB, find X using the Park-Martin method. Here, A represents the relative motion of the flange in the robot arm coordinate system (from time i to time j, i.e., the relative motion determined by the two sampling points within the sampling point pair), B represents the relative motion of the first tracking reference device in the tracking device coordinate system (from time i to time j), and X represents the fixed transformation relationship between the robot arm end flange coordinate system and the first tracking reference device coordinate system fixed thereon.

[0099] In this embodiment, by constructing sampling point pairs and calculating relative motion, systematic errors in absolute pose measurement (such as inaccurate tracking device parameters and robot base calibration errors) are eliminated, further improving calibration accuracy. Furthermore, the data required for hand-eye calibration (first tracking data, robotic arm data) are synchronously collected with other calibration tasks along the same motion trajectory, requiring no additional motion and enabling collaborative calibration of multiple tasks.

[0100] In this embodiment of the disclosure, the sampling points for the hand-eye calibration task can be collected and the calibration operation performed simultaneously, or the calibration operation can be performed after collecting a predetermined number of sampling points to solve for the rotation and translation parameters, and then optimization can be performed based on the newly collected sampling points. For example, a certain number of sampling point pairs can be collected first to solve for the rotation and translation parameters, and then optimization can be performed based on the subsequently collected sampling points.

[0101] In some embodiments, for hand-eye calibration tasks, to further improve the accuracy of calibration results, before constructing multiple sets of sampling point pairs based on multiple sampling points of the hand-eye calibration task, regardless of when the sampling points are collected, the method further includes: screening sampling point pairs that meet the qualification conditions to obtain valid multiple sets of sampling point pairs; the qualification conditions include that the relative motion distance expressed by the sampling point pair is greater than the target distance, and the relative deflection angle is greater than the angle threshold.

[0102] The relative motion distance can represent the translational distance between the center of the end flange of the robotic arm (or the center of the first sub-tracking reference device) of the two sampling points in a sampling point pair, which can be exemplified as the Euclidean distance in three-dimensional space. Only sampling point pairs whose relative motion distance is greater than the target distance are considered to provide effective translational constraints.

[0103] During implementation, the translation vectors at two time points are extracted from the robotic arm data (or the first tracking data) of the sampling point pair, and their Euclidean distance is calculated. For example, if the center positions of the end flanges at the i-th and j-th sampling points are respectively... and The relative motion distance between the i-th sampling point and the j-th sampling point is... .

[0104] The relative deflection angle refers to the angle of attitude change of the center of the end flange of the robotic arm between two sampling points in a sampling point pair. It usually refers to the rotation angle in the axis angle representation corresponding to the rotation matrix. Only sampling point pairs with a relative deflection angle greater than the angle threshold are considered to provide effective rotation constraints.

[0105] During implementation, the rotation matrix at two time points is extracted from the robotic arm data (or the first tracking data) in this sampling point pair. and Calculate the relative rotation matrix Then, based on the conversion relationship between the trace (the sum of the diagonal elements) and the rotation angle, the rotation angle is obtained. .

[0106] During implementation, if the relative motion distance expressed by the sampling point is greater than the target distance and the relative deflection angle is greater than the angle threshold, then the hand-eye calibration task is performed based on the sampling point pair. If the relative motion distance expressed by the sampling point pair is not greater than the target distance or the relative deflection angle is not greater than the angle threshold, then the sampling point pair is discarded.

[0107] In this embodiment of the disclosure, by screening qualified sampling point pairs, invalid data that is dominated by measurement noise due to too small motion amplitude can be eliminated, so that each sampling point pair participating in the hand-eye calibration calculation has significant translation and rotation changes, thereby improving the accuracy and numerical stability of the calibration parameters of the hand-eye calibration task.

[0108] In some embodiments, after determining the rotation and translation parameters of the hand-eye calibration task using (N+1) sets of sampling point pairs, the method further includes: performing the following operations for each newly added sampling point pair after the (N+1) sets of sampling point pairs, until the number of elements in the latest target parameter set reaches the target number, where N is a positive integer greater than 1: Step C1: Add the sampling point pairs to the target parameter set to obtain the intermediate set; the target parameter set is used to collect (N+1) sets of sampling point pairs, as well as qualified point pairs that meet the stability requirements.

[0109] During implementation, (N+1) pairs of sampling points that meet the initial qualification criteria are selected. After stability verification confirms that they meet the stability requirements, these pairs of sampling points are added to the target parameter set, where all point pairs in the target parameter set meet the stability requirements. If new pairs of sampling points exist, an intermediate set is constructed based on the newly added pairs and the previously qualified pairs that have passed the stability requirements.

[0110] Step C2: Based on the relative motion of each sampling point pair in the intermediate set, determine the intermediate values ​​of the calibration parameters for the hand-eye calibration task.

[0111] In one embodiment, the intermediate set can be processed by referring to (N+1) sets of sampling points to solve for the rotation and translation parameters, thereby obtaining the intermediate values ​​of the calibration parameters. .

[0112] In another implementation, at least one set of sampling point pairs, including the newly added sampling point pairs, can be selected from the intermediate set, and the intermediate values ​​of the calibration parameters can be solved by referring to the processing method of (N+1) sets of sampling point pairs. .

[0113] Step C3: Determine the degree of difference between the intermediate value of the calibration parameter and the reference value of the calibration parameter. The reference value of the calibration parameter is determined based on the relative motion of each sampling point pair in the previous target parameter set when a new sampling point pair is added.

[0114] In one embodiment, the calibration parameter reference value is obtained based on the rotation and translation parameters of the original target parameter set (without adding new sampling point pairs); the rotation and translation parameters are substituted into the hand-eye calibration equation to obtain the calibration parameter reference value. .

[0115] The way the difference between the two can be expressed can be set according to the needs, such as distance, ratio, etc.

[0116] Step C4: If the difference is less than the difference threshold, determine the sampling point pair as a qualified point pair that meets the stability requirements, update the intermediate set to a new target parameter set, and update the mean of the intermediate value of the calibration parameter and the reference value of the calibration parameter to a new reference value of the calibration parameter; if the difference is greater than or equal to the difference threshold, discard the sampling point pair from the intermediate set. If the termination condition is not met, continue to introduce new sampling point pairs.

[0117] During implementation, if the difference is less than the difference threshold, it indicates that the newly added sampling point pair is consistent with the data of all existing sampling point pairs in the target parameter set; otherwise, it may be an outlier. In the case of outliers, the new sampling point pair will be filtered out and will not participate in the parameter solving for hand-eye calibration.

[0118] In this embodiment of the disclosure, due to various factors such as interference, some sampling point pairs may produce erroneous relative motion. By using a difference threshold, outliers can be effectively eliminated, ensuring that newly added sampling point pairs have good consistency with the sampling point pairs in the target parameter set, thereby improving the accuracy and robustness of the calibration parameters.

[0119] 2) Tool Calibration In some embodiments, when at least two calibration tasks include a tool calibration task, preset data for the tool calibration task is acquired in the first trajectory portion and the second trajectory portion.

[0120] The tool calibration task is completed based on the first sub-tracking reference device and the second sub-tracking reference device. Based on this, the preset data for each sampling point of the tool calibration task includes: The first tracking data is used to record the pose of the first sub-tracking reference device in the tracking device coordinate system; The second tracking data is used to record the pose of the second sub-tracking reference device in the tracking device coordinate system.

[0121] In practice, when visual navigation is used for calibration, the first sub-tracking reference device is the end target, and the second sub-tracking reference device is the calibration disk. The first tracking data provides the pose of the end target under the camera; the second tracking data provides the pose of the calibration disk under the camera.

[0122] In some embodiments, based on preset data of the tool calibration task, the corresponding calibration operation is performed, which can be implemented as follows: for multiple sampling points of the tool calibration task, the spatial transformation relationship between the tool's operating position and the first sub-tracking reference device is determined based on the spatial mapping relationship; the spatial mapping relationship is used to describe the known transformation relationship between the second sub-tracking reference device and the operating position, and is converted into a spatial transformation relationship through the first tracking data and the second tracking data in the sampling points; the mean value of the spatial transformation relationship obtained from multiple sampling points is calculated to obtain the calibration result of the tool calibration.

[0123] After obtaining the spatial transformation relationship of multiple sampling points based on the aforementioned method, the mean is calculated to obtain the calibration result of the tool calibration.

[0124] In this embodiment of the disclosure, compared with the traditional method that relies on manual touching of a fixed tip, this approach reduces errors caused by alignment during operation, further improving calibration accuracy and automation. Furthermore, since the first and second tracking data can be collected synchronously with other calibration tasks, there is no need to plan a separate motion trajectory for tool calibration, thereby simplifying the calibration process and shortening the overall time.

[0125] 3) Force sensor calibration In some embodiments, when at least two calibration tasks include a force sensor calibration task, preset data for the tool calibration task is acquired in the first trajectory portion and the second trajectory portion.

[0126] The force sensor calibration task includes a force parameter calibration task; the preset data for each sampling point in the force parameter calibration task includes: Force data is used to record the raw force readings collected by the force sensor; Gravity component data is used to record the components of the gravitational acceleration vector in the three directions of the force sensor coordinate system.

[0127] Among them, the force data refers to the raw force reading output by the force sensor at each sampling time, which is usually composed of three components ( , The gravity component data consists of three components of the gravitational acceleration vector calculated based on the current robot posture in the force sensor coordinate system. .

[0128] Based on the preset data for each calibration task, perform the corresponding calibration operations, including the following steps: Step D1: For the force parameter calibration task, obtain multiple sampling points for the force parameter calibration task.

[0129] During the execution of a preset motion trajectory by the robotic arm, force data and gravity component data are collected at a preset frequency. The preset frequency can be a fixed frequency or a periodic frequency, which can be determined based on the actual situation.

[0130] Step D2 involves minimizing the force residual using the least squares method to solve for the total load mass and zero-point offset of the force in the force parameter calibration task. The force residual is used to represent the difference between the theoretical sensed force and the force data measured by the force sensor at multiple sampling points. The theoretical sensed force is generated based on the gravity component data, total load mass, and zero-point offset of the force at each sampling point.

[0131] For each sampling point, based on the gravity component data, total load mass, and zero-point force bias of each sampling point, an additional force model for the force sensor is constructed. This model represents the data acquired by the force sensor under the load influence of the second sub-tracking reference device. It can be assumed that... represents the components of the gravitational acceleration vector in the three directions of the force sensor coordinate system; m is the total mass of the load. , , Indicates the zero-point offset of the force. , This represents the raw force reading acquired by the force sensor. When the additional force is zero, this model can be used for parameter calibration. Once the parameters are calibrated, this model can be used for real-time estimation of the additional force.

[0132] Based on the rotation matrix of the end effector flange coordinate system relative to the robot base coordinate system, and the pre-calibrated rotation matrix of the sensor coordinate system relative to the end effector flange coordinate system, the attitude of the force sensor in the base coordinate system is calculated. Based on the constructed additional force model, the unknown quantity (m, , , In practice, multiple sampling points are combined to form an overdetermined linear equation system. By minimizing the sum of squared force residuals at all sampling points, the total load mass and zero-point offset of the force in the force parameter calibration task can be obtained.

[0133] In this embodiment, calibration can be completed using the robot's own multi-pose motion, sharing the same motion trajectory with other calibration tasks, thus enabling efficient data reuse. By constructing an overdetermined equation system using multiple sampling points, the calibration accuracy of mass and bias can be improved, providing reliable parameters for accurate compensation of subsequent real-time contact forces.

[0134] In some embodiments, the total load mass in the force sensor calibration task includes the second sub-tracking reference device, while the operation task does not include the second sub-tracking reference device. In this case, the method further includes: Step E1: Subtract the known mass of the second sub-tracking reference device from the total load mass to obtain the load mass to be processed by the force sensor in the working state.

[0135] In the calibration task, the total load mass includes the mass of the second sub-tracking reference device. In the operation task, the second sub-tracking reference device needs to be removed first, that is, the total load mass is subtracted from the known mass of the second sub-tracking reference device to obtain the load mass to be processed in the operation state.

[0136] Step E2: In the working state, acquire the real-time force measured by the force sensor.

[0137] During implementation, the real-time force in the working state is not affected by the mass of the second sub-tracking reference device.

[0138] Step E3: Based on the mass of the load to be processed and the zero-point bias of the force sensor, the real-time force is compensated to obtain the contact force under the working state.

[0139] Specifically, the contact force under operating conditions As shown in equation (1): (1) in, Indicates real-time force. Indicates the quality of the load to be processed. This indicates the zero-point offset of the force, and G represents the component of the gravitational acceleration vector in the current force sensor coordinate system, which is usually a three-dimensional vector. ).

[0140] In this embodiment, the mass of the second sub-tracking reference device increases the gravity signal, making the least-squares solution more stable and improving the accuracy of force parameter calibration. During the operation phase, the second sub-tracking reference device is removed, and based on its known mass and center-of-mass parameters, the corresponding parameters of the actual tool are extracted from the total load parameters obtained during calibration, enabling accurate calculation of the actual tool contact force during operation. Through the above calibration and parameter extraction process, seamless data and parameter integration from the calibration state to the operation state is achieved.

[0141] In some embodiments, the force sensor calibration task further includes a torque parameter calibration task; the preset data for each sampling point in the torque parameter calibration task includes: Torque data is used to record the raw torque readings collected by the force sensor; Gravity component data is used to record the components of the gravitational acceleration vector in the three directions of the force sensor coordinate system.

[0142] Based on the preset data for each calibration task, perform the corresponding calibration operations, including: Step F1: For the torque parameter calibration task, obtain multiple sampling points for the torque parameter calibration task.

[0143] During the execution of the preset motion trajectory by the robotic arm, torque data and gravity component data are collected at a preset frequency. The preset frequency has been described above and will not be repeated here. It should be noted that the sampling points for the torque parameter calibration task and the aforementioned force parameter calibration task are collected simultaneously, and the parameters collected at the same sampling point can simultaneously meet the requirements of these two calibration sub-tasks.

[0144] Step F2: For each sampling point, construct the theoretical torque vector generated by gravity based on the gravity component data, the total mass of the load, and the centroid position information to be solved; the centroid position information is the position offset vector of the load centroid relative to the end flange coordinate system.

[0145] For each sampling point, based on the gravity component data, the total load mass, and the centroid position information to be solved, an additional torque model generated by gravity is constructed. This model is based on the torque data collected by the force sensor under the load influence of the second sub-tracking reference device. represents the components of the gravitational acceleration vector in the three directions of the force sensor coordinate system; m is the total mass of the load. , , Indicates the zero-point offset of the force. , This represents the raw torque reading acquired by the force sensor. , , This represents the components of the position offset vector of the load's center of mass relative to the end flange coordinate system in three directions. When the additional force is zero, this model can be used to complete parameter calibration. Once the parameters are calibrated, this model can be used to estimate the additional torque in real time.

[0146] Step F3 obtains the centroid position information and the zero-point offset of the torque by minimizing the torque residual between the theoretical torque vector and the torque data of multiple sampling points.

[0147] Expanding the equation at each sampling point yields information about the unknown quantity ( , , , , , The linear equations of the sampled points are obtained. Multiple sampling points simultaneously form an overdetermined system of linear equations. By minimizing the torque residuals at all sampling points, the centroid position information and the zero-point offset of the torque can be obtained.

[0148] In this embodiment, the same trajectory can be shared with other calibration tasks to reduce operational complexity and shorten the total time. Multiple sampling points are used to construct an overdetermined system of equations and solve it using the least squares method to improve the calibration accuracy of the centroid position and torque offset. This provides reliable parameters for accurate compensation of subsequent real-time external torques, thereby enhancing the accuracy and safety of the robot's force control operations.

[0149] In some embodiments, where the total load mass in the force sensor calibration task includes the second sub-tracking reference device, but the task does not include the second sub-tracking reference device, the method further includes: Step G1: Subtract the known mass of the second sub-tracking reference device from the total load mass to obtain the load mass to be processed by the force sensor in the working state.

[0150] The method for determining the quality of the load to be processed has been explained above and will not be repeated here.

[0151] Step G2: In the working state, acquire the real-time torque measured by the force sensor.

[0152] During implementation, the real-time torque in the working state is the torque of the mass without a second tracking reference device.

[0153] Step G3: Based on the mass of the load to be processed, the centroid position information determined by the force sensor calibration task, and the zero-point offset of the force sensor torque, the real-time torque is compensated to obtain the contact torque under the working state.

[0154] During implementation, the change in the center of mass and the contact force under operating conditions need to be ignored. As shown in equation (2): (2) in, Indicates real-time torque. Indicates the quality of the load to be processed. This indicates the zero-point offset of the torque, and g represents the component of the gravitational acceleration vector in the current force sensor coordinate system, which is usually a three-dimensional vector. ), r represents the centroid position information.

[0155] In this embodiment, during the calibration phase, a second sub-tracking reference device with a known mass is used to enhance the excitation intensity of the gravitational torque signal, thereby improving the identification accuracy of the center of mass position and the zero-point offset of the torque. During the operation phase, the mass of the second sub-tracking reference device is automatically deducted, and the real-time torque is accurately compensated using the calibrated center of mass position and zero-point offset of the torque to obtain the true contact torque. This method achieves seamless switching from the calibration state to the operation state without replacing the sensor or recalibrating, significantly improving calibration efficiency.

[0156] In some embodiments, it is also necessary to verify and optimize the results obtained from the aforementioned calibration task: In one embodiment, the verification of the hand-eye calibration task can be implemented as follows: a robotic arm executes a predetermined short path under visual guidance, and the calibration result is judged as qualified based on whether the actual executed path matches the predetermined path.

[0157] Verification of force sensor calibration tasks: After completing the force sensor calibration (including force parameter calibration and torque parameter calibration), it is necessary to verify the accuracy of the calibration results. The core idea of ​​the verification is: under the condition of no external force contact, move the robot end effector to multiple different postures, and use the calibrated parameters (total load mass, center of mass position, force zero-point offset, torque zero-point offset) to perform gravity compensation and offset subtraction on the original sensor readings, and check whether the residual force and residual torque after compensation are close to zero.

[0158] The core idea for verifying tool calibration tasks is to use an independent external reference with known accuracy (such as a high-precision probe or a third-party vision measurement system) to compare the theoretical position of the tool tip displayed on the system interface with the actual physical position through physical contact or non-contact measurement, calculate the position error, and thus evaluate the accuracy of the tool calibration parameters. If the error under all postures is less than a preset accuracy threshold, the tool calibration result is considered valid.

[0159] For example, such as Figure 5 As shown, taking the head-up motion mode as an example, while simultaneously meeting the requirements of three calibration tasks, the robot includes a first joint, a second joint, a third joint, a fourth joint, a fifth joint, and a sixth joint. The fourth joint can perform the head-up motion to ensure that the pitch angle changes meet the force sensor calibration requirements; the fifth and sixth joints can both rotate back and forth to ensure that the roll and yaw angles change to meet the requirements of other calibration tasks.

[0160] Based on the same technical concept, this disclosure also provides a multi-task calibration robot system, including: A robotic arm with a tool and a tracking reference device mounted at its end; Tracking device, used to acquire the pose of the tracking reference device; and The controller includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the aforementioned multi-task cooperative calibration method for robots.

[0161] The robotic arm is the same as the robotic arm of the aforementioned robot, and the tracking device is the same as described above, so it will not be repeated here.

[0162] Based on the same technical concept, this disclosure proposes a multi-task collaborative calibration device 600 for robots, such as... Figure 6 As shown, it includes: The generation module 601 is used to generate control commands for the robot's robotic arm based on motion constraint information; the motion constraint information is used to control the robotic arm to cover multiple postures during movement. The data acquisition module 602 is used to acquire preset data for at least two calibration tasks during the process of the robot controlling the movement of the robotic arm in response to the control command. The execution module 603 is used to perform corresponding calibration operations based on the preset data of each calibration task.

[0163] In some embodiments, the at least two calibration tasks include at least two of the following: hand-eye calibration task, tool calibration task, and force sensor calibration task; The plurality of postures includes the tracking reference device required to perform the at least two calibration tasks being within a preset sensing range of the tracking device, and the variation range of at least one target angle from the following set of angles of the end flange of the robotic arm being greater than or equal to the corresponding preset angle: The set of angles includes: pitch angle, roll angle, and yaw angle; The tracking reference device includes a first sub-tracking reference device on a tool mounted at the end of the robotic arm, and / or a second sub-tracking reference device independent of the tool.

[0164] In some embodiments, when the at least two calibration tasks include a force sensor calibration task and a hand-eye calibration task, the following condition is met: The preset angle corresponding to the range of pitch angle changes is greater than or equal to the first target angle; The preset angle corresponding to the range of roll angle variation is greater than or equal to the second target angle, and / or the preset angle corresponding to the range of yaw angle variation is greater than or equal to the third target angle; The first target angle is greater than the second target angle and / or the third target angle.

[0165] In some embodiments, the control command is used to control the robotic arm to execute a first motion trajectory, the first motion trajectory including a first trajectory segment and a second trajectory segment; In the first trajectory segment, the pitch angle of the end flange moves from a first angle to a second angle; In the second trajectory segment, the pitch angle of the end flange moves from the second angle to the third angle; The angle difference between the first angle and the third angle is greater than or equal to the preset angle corresponding to the pitch angle.

[0166] In some embodiments, the control command is used to control the robotic arm to execute a second motion trajectory, the second motion trajectory conforming to a polyhedral motion pattern; the second motion trajectory in the polyhedral motion pattern includes a third trajectory segment and a fourth trajectory segment; the third trajectory segment includes at least one first side; the fourth trajectory segment includes at least one second side; when multiple first sides are included, the length of each first side follows the definition of the side length of a polyhedron; when multiple second sides are included, the length of each second side follows the definition of the side length of a polyhedron. In the third trajectory segment, the end flange performs a translation operation along the first side, and at the same time, the pitch angle of the end flange changes more than the first set angle. In the fourth trajectory segment, the end flange performs a translation operation along the second side, and at the same time, the pitch angle of the end flange changes more than the second set angle. In the second motion trajectory, the pitch angle of the end flange varies within a range greater than or equal to the preset angle corresponding to the pitch angle.

[0167] In some embodiments, for hand-eye calibration, in at least a portion of the movement trajectory of the robotic arm, the yaw angle and / or roll angle of the end flange continuously change, and the range of change is greater than or equal to the corresponding preset angle.

[0168] In some embodiments, the acquisition module includes: The data acquisition unit is used to acquire preset data of the hand-eye calibration task in the first trajectory part of the robotic arm's movement when the at least two calibration tasks include the hand-eye calibration task. The control unit is used to stop collecting preset data for the hand-eye calibration task when the robotic arm moves to the second trajectory section; the first trajectory section is located before the second trajectory section.

[0169] In some embodiments, when both the third trajectory segment and the fourth trajectory segment include at least two edges, any of the following conditions are met: In the intermediate trajectory segment formed by the last first side of the third trajectory segment and the first second side of the fourth trajectory segment, the pitch angle of the end flange varies within a range greater than the preset angle corresponding to the pitch angle. The trajectory formed by the last first side of the third trajectory segment and the trajectory formed by the first second side of the fourth trajectory segment both satisfy the condition that the pitch angle of the end flange varies within a range greater than the preset angle corresponding to the pitch angle.

[0170] In some embodiments, where the at least two calibration tasks include the hand-eye calibration task, the hand-eye calibration task is performed based on the first sub-tracking reference device; The preset data for each sampling point in the hand-eye calibration task includes: The first tracking data is used to record the pose of the first sub-tracking reference device in the coordinate system of the tracking device. Robotic arm data, used to record the pose of the flange center of the robotic arm in the robot's base coordinate system; The execution module includes: The first construction unit is used to construct multiple sets of sampling point pairs based on multiple sampling points of the hand-eye calibration task for the hand-eye calibration task. The determining unit is used to determine the calibration parameters of the hand-eye calibration task based on the relative motion of each sampling point pair in the multiple sets of sampling point pairs.

[0171] In some embodiments, before constructing multiple sets of sampling point pairs based on multiple sampling points of the hand-eye calibration task, a first filtering module is further included for: Select sampling point pairs that meet the qualification criteria to obtain multiple valid sampling point pairs; the qualification criteria include that the relative motion distance expressed by the sampling point pair is greater than the target distance and the relative deflection angle is greater than the angle threshold.

[0172] In some embodiments, the constructed multiple sets of sampling point pairs are (N+1) sets of sampling point pairs. After determining the rotation and translation parameters of the hand-eye calibration task using (N+1) sets of sampling point pairs, a second filtering module is further included, used for: For each new sampling point pair added after the (N+1) sets of sampling point pairs, perform the following operations until the number of elements in the latest target parameter set reaches the target number, where N is a positive integer greater than 1: The sampling point pairs are added to the target parameter set to obtain an intermediate set; the target parameter set is used to collect the (N+1) sets of sampling point pairs, as well as qualified point pairs that meet the stability requirements; Based on the relative motion of each sampling point pair in the intermediate set, the intermediate values ​​of the calibration parameters for the hand-eye calibration task are determined. The degree of difference between the intermediate value of the calibration parameter and the reference value of the calibration parameter is determined, wherein the reference value of the calibration parameter is determined based on the relative motion of the newly added sampling point to each sampling point pair in the previous target parameter set; If the difference is less than the difference threshold, the sampling point pair is determined to be a qualified point pair that meets the stability requirements, the intermediate set is updated to a new target parameter set, and the mean of the intermediate value of the calibration parameter and the reference value of the calibration parameter is updated to a new reference value of the calibration parameter. If the difference is greater than or equal to the difference threshold, the sample point pair is discarded from the intermediate set.

[0173] In some embodiments, where the at least two calibration tasks include a tool calibration task, the tool calibration task is performed based on the first sub-tracking reference device and the second sub-tracking reference device; The preset data for each sampling point in the tool calibration task includes: The first tracking data is used to record the pose of the first sub-tracking reference device in the coordinate system of the tracking device. The second tracking data is used to record the pose of the second sub-tracking reference device in the coordinate system of the tracking device.

[0174] In some embodiments, where the at least two calibration tasks include a force sensor calibration task, the force sensor calibration task includes a torque parameter calibration task; the preset data for each sampling point in the torque parameter calibration task includes: Torque data is used to record the raw torque readings collected by the force sensor; Gravity component data is used to record the components of the gravitational acceleration vector in the three directions of the force sensor coordinate system. The execution module includes: The acquisition unit is used to acquire multiple sampling points for the torque parameter calibration task. The second construction unit is used to construct a theoretical torque vector generated by gravity for each sampling point based on the gravity component data, the total mass of the load, and the centroid position information to be solved; the centroid position information is the position offset vector of the load centroid relative to the end flange coordinate system.

[0175] In some embodiments, where the total load mass in the force sensor calibration task includes the second sub-tracking reference device, but the operation task does not include the second sub-tracking reference device, the device further includes a calibration module for: By minimizing the torque residual between the theoretical torque vector and the torque data of the multiple sampling points, the centroid position information and the zero-point offset of the torque are obtained; The known mass of the second sub-tracking reference device is subtracted from the total mass of the load to obtain the mass of the load to be processed by the force sensor in the working state; In the operating state, the real-time torque measured by the force sensor is acquired. Based on the mass of the load to be processed, the centroid position information determined by the force sensor calibration task, and the zero-point offset of the torque of the force sensor, the real-time torque is compensated to obtain the contact torque under the working state.

[0176] In some embodiments, an adjustment module is further included, for: When the tracking reference device is a visual navigation and positioning device, before generating the control commands for the robot's robotic arm based on motion constraint information, the following steps are also included: Perform a preset adaptive operation to adjust the visual navigation and positioning device to a preset sensing angle of the tracking device.

[0177] The specific functions and examples of each module and submodule of the apparatus in this disclosure can be found in the relevant descriptions of the corresponding steps in the above method embodiments, and will not be repeated here.

[0178] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0179] Figure 7 This is a structural block diagram of an electronic device according to an embodiment of the present disclosure. Figure 7 As shown, the electronic device includes a memory 710 and a processor 720. The memory 710 stores a computer program that can run on the processor 720. The number of memories 710 and processors 720 can be one or more. The memory 710 can store one or more computer programs, which, when executed by the electronic device, cause the electronic device to perform the methods provided in the above-described method embodiments. The electronic device may also include a communication interface 730 for communicating with external devices and performing data exchange and transmission.

[0180] If the memory 710, processor 720, and communication interface 730 are implemented independently, they can be interconnected via a bus to communicate with each other. This bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 7 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0181] Optionally, in a specific implementation, if the memory 710, processor 720, and communication interface 730 are integrated on a single chip, then the memory 710, processor 720, and communication interface 730 can communicate with each other through an internal interface.

[0182] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. General-purpose processors can be microprocessors or any conventional processor. It is worth noting that the processor can be a processor supporting Advanced Reduced Instruction Set Machines (ARM) architecture.

[0183] Further, optionally, the aforementioned memory may include read-only memory and random access memory, and may also include non-volatile random access memory. The memory may be volatile or non-volatile, or may include both. Non-volatile memory may include read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory may include random access memory (RAM), which serves as an external cache. Many forms of RAM are available by way of example, but not limitation. Examples include Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate Synchronous DRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct RAMBUS RAM (DR RAM).

[0184] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this disclosure are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line, DSL) or wireless (e.g., infrared, Bluetooth, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer, or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Disks (SSDs)). It is worth noting that the computer-readable storage media mentioned in this disclosure can be non-volatile storage media; in other words, it can be non-transient storage media.

[0185] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0186] In the description of the embodiments of this disclosure, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0187] In the description of the embodiments disclosed herein, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone.

[0188] In the description of embodiments of this disclosure, 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 indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of embodiments of this disclosure, unless otherwise stated, "a plurality of" means two or more.

[0189] The above description is merely an exemplary embodiment of this disclosure and is not intended to limit this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the protection scope of this disclosure.

Claims

1. A multi-task cooperative calibration method for robots, comprising: Based on motion constraint information, control commands for the robot's robotic arm are generated; The motion constraint information is used to control the robotic arm to cover multiple postures during its movement; During the process of the robot responding to the control command to control the movement of the robotic arm, preset data for at least two calibration tasks are collected respectively. Based on the preset data for each calibration task, perform the corresponding calibration operations.

2. The method according to claim 1, wherein, The at least two calibration tasks include at least two of the following: hand-eye calibration task, tool calibration task, and force sensor calibration task; The plurality of postures includes the tracking reference device required to perform the at least two calibration tasks being within a preset sensing range of the tracking device, and the variation range of at least one target angle from the following set of angles of the end flange of the robotic arm being greater than or equal to the corresponding preset angle: The set of angles includes: pitch angle, roll angle, and yaw angle; The tracking reference device includes a first sub-tracking reference device on a tool mounted at the end of the robotic arm, and / or a second sub-tracking reference device independent of the tool.

3. The method according to claim 2, wherein, When the at least two calibration tasks include a force sensor calibration task and a hand-eye calibration task, the following conditions must be met: The preset angle corresponding to the range of pitch angle changes is greater than or equal to the first target angle; The preset angle corresponding to the range of roll angle variation is greater than or equal to the second target angle, and / or the preset angle corresponding to the range of yaw angle variation is greater than or equal to the third target angle; The first target angle is greater than the second target angle and / or the third target angle.

4. The method according to claim 2 or 3, wherein, The control command is used to control the robotic arm to execute a first motion trajectory, the first motion trajectory including a first trajectory segment and a second trajectory segment; In the first trajectory segment, the pitch angle of the end flange moves from a first angle to a second angle; In the second trajectory segment, the pitch angle of the end flange moves from the second angle to the third angle; The angle difference between the first angle and the third angle is greater than or equal to the preset angle corresponding to the pitch angle.

5. The method according to claim 2 or 3, wherein, The control command is used to control the robotic arm to execute a second motion trajectory, which conforms to a polyhedral motion pattern. The second motion trajectory in the polyhedral motion pattern includes a third trajectory segment and a fourth trajectory segment. The third trajectory segment includes at least one first side; the fourth trajectory segment includes at least one second side. When multiple first sides are included, the length of each first side follows the definition of the side length of a polyhedron. When multiple second sides are included, the length of each second side follows the definition of the side length of a polyhedron. In the third trajectory segment, the end flange performs a translation operation along the first side, and at the same time, the pitch angle of the end flange changes more than the first set angle. In the fourth trajectory segment, the end flange performs a translation operation along the second side, and at the same time, the pitch angle of the end flange changes more than the second set angle. In the second motion trajectory, the pitch angle of the end flange varies within a range greater than or equal to the preset angle corresponding to the pitch angle.

6. The method according to claim 4 or 5, wherein, For hand-eye calibration, in at least a portion of the movement trajectory of the robotic arm, the yaw angle and / or roll angle of the end flange continuously change, and the range of change is greater than or equal to the corresponding preset angle.

7. The method according to claim 4 or 5, wherein, During the process of the robot responding to the control command to control the movement of the robotic arm, preset data for at least two calibration tasks are collected, including: In the case where the at least two calibration tasks include a hand-eye calibration task, preset data for the hand-eye calibration task is collected during the first trajectory portion of the robotic arm's movement. When the robotic arm moves to the second trajectory section, it stops collecting the preset data for the hand-eye calibration task; the first trajectory section is located before the second trajectory section.

8. The method according to claim 5, wherein when both the third trajectory segment and the fourth trajectory segment include at least two edges, any of the following conditions are satisfied: In the intermediate trajectory segment formed by the last first side of the third trajectory segment and the first second side of the fourth trajectory segment, the pitch angle of the end flange varies within a range greater than the preset angle corresponding to the pitch angle. The trajectory formed by the last first side of the third trajectory segment and the trajectory formed by the first second side of the fourth trajectory segment both satisfy the condition that the pitch angle of the end flange varies within a range greater than the preset angle corresponding to the pitch angle.

9. The method according to claim 2, wherein, In the case where the at least two calibration tasks include the hand-eye calibration task, the hand-eye calibration task is completed based on the first sub-tracking reference device; The preset data for each sampling point in the hand-eye calibration task includes: The first tracking data is used to record the pose of the first sub-tracking reference device in the coordinate system of the tracking device. Robotic arm data, used to record the pose of the flange center of the robotic arm in the robot's base coordinate system; The step of performing corresponding calibration operations based on preset data for each calibration task includes: For the hand-eye calibration task, multiple sets of sampling point pairs are constructed based on multiple sampling points of the hand-eye calibration task; Based on the relative motion of each sampling point pair in the multiple sets of sampling point pairs, the calibration parameters of the hand-eye calibration task are determined.

10. The method according to claim 9, wherein, For the hand-eye calibration task, before constructing multiple sets of sampling point pairs based on multiple sampling points of the hand-eye calibration task, the method further includes: Select sampling point pairs that meet the qualification criteria to obtain multiple valid sampling point pairs; the qualification criteria include that the relative motion distance expressed by the sampling point pair is greater than the target distance and the relative deflection angle is greater than the angle threshold.

11. The method according to claim 9 or 10, wherein the constructed multiple sets of sampling point pairs are (N+1) sets of sampling point pairs, and after determining the rotation parameters and translation parameters of the hand-eye calibration task using (N+1) sets of sampling point pairs, the method further includes: For each new sampling point pair added after the (N+1) sets of sampling point pairs, perform the following operations until the number of elements in the latest target parameter set reaches the target number, where N is a positive integer greater than 1: The sampling point pairs are added to the target parameter set to obtain an intermediate set; the target parameter set is used to collect the (N+1) sets of sampling point pairs, as well as qualified point pairs that meet the stability requirements; Based on the relative motion of each sampling point pair in the intermediate set, the intermediate values ​​of the calibration parameters for the hand-eye calibration task are determined. The degree of difference between the intermediate value of the calibration parameter and the reference value of the calibration parameter is determined, wherein the reference value of the calibration parameter is determined based on the relative motion of the newly added sampling point to each sampling point pair in the previous target parameter set; If the difference is less than the difference threshold, the sampling point pair is determined to be a qualified point pair that meets the stability requirements, the intermediate set is updated to a new target parameter set, and the mean of the intermediate value of the calibration parameter and the reference value of the calibration parameter is updated to a new reference value of the calibration parameter. If the difference is greater than or equal to the difference threshold, the sample point pair is discarded from the intermediate set.

12. The method according to claim 2, wherein, In the case where the at least two calibration tasks include a tool calibration task, the tool calibration task is performed based on the first sub-tracking reference device and the second sub-tracking reference device; The preset data for each sampling point in the tool calibration task includes: The first tracking data is used to record the pose of the first sub-tracking reference device in the coordinate system of the tracking device. The second tracking data is used to record the pose of the second sub-tracking reference device in the coordinate system of the tracking device.

13. The method according to claim 1, wherein when the at least two calibration tasks include a force sensor calibration task, the force sensor calibration task includes a torque parameter calibration task; The preset data for each sampling point in the torque parameter calibration task includes: Torque data is used to record the raw torque readings collected by the force sensor; Gravity component data is used to record the components of the gravitational acceleration vector in the three directions of the force sensor coordinate system. The step of performing corresponding calibration operations based on preset data for each calibration task includes: For the torque parameter calibration task, multiple sampling points of the torque parameter calibration task are obtained; For each sampling point, a theoretical torque vector generated by gravity is constructed based on the gravity component data, the total mass of the load, and the centroid position information to be solved; the centroid position information is the position offset vector of the load centroid relative to the end flange coordinate system.

14. The method according to claim 13, wherein, When the total load mass in the force sensor calibration task includes the second sub-tracking reference device, but the operation task does not include the second sub-tracking reference device, the method further includes: By minimizing the torque residual between the theoretical torque vector and the torque data of the multiple sampling points, the centroid position information and the zero-point offset of the torque are obtained; The known mass of the second sub-tracking reference device is subtracted from the total mass of the load to obtain the mass of the load to be processed by the force sensor in the working state; In the operating state, the real-time torque measured by the force sensor is acquired. Based on the mass of the load to be processed, the centroid position information determined by the force sensor calibration task, and the zero-point offset of the torque of the force sensor, the real-time torque is compensated to obtain the contact torque under the working state.

15. The method according to claim 2, further comprising: When the tracking reference device is a visual navigation and positioning device, before generating the control commands for the robot's robotic arm based on motion constraint information, the following steps are also included: Perform a preset adaptive operation to adjust the visual navigation and positioning device to a preset sensing angle of the tracking device.

16. A multi-task collaborative calibration device for robots, comprising: The generation module is used to generate control commands for the robot's robotic arm based on motion constraint information. The motion constraint information is used to control the robotic arm to cover multiple postures during its movement; The data acquisition module is used to acquire preset data for at least two calibration tasks during the process of the robot controlling the movement of the robotic arm in response to the control command. The execution module is used to perform corresponding calibration operations based on the preset data of each calibration task.

17. An electronic device comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-15.

18. A multi-task calibration robot system, characterized in that, include: A robotic arm with a tool and a tracking reference device mounted at its end; A tracking device for acquiring the pose of the tracking reference device; as well as The controller includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the multi-task collaborative calibration method as described in any one of claims 1 to 15.

Images