Robot control method, device, apparatus, and storage medium

Abstract

The application relates to a robot control method, device, equipment and storage medium. The method comprises the following steps: acquiring a current pose, a current action type and a target pose corresponding to the current action type of a target robot; determining a motion control strategy of the target robot according to the current pose, the current action type and the target pose; controlling the motion of the target robot according to the motion control strategy; and switching the current action type when the action switching condition corresponding to the current action type is met. The above method guarantees the accuracy and reliability of the pipe gallery support installation process.

Description

Technical Field

[0001] This application relates to the field of robot control technology, and in particular to a robot control method, apparatus, device, and storage medium. Background Technology

[0002] With the development of robot control technology, robots are widely used in complex operation scenarios such as cable bracket installation in utility tunnels. By replacing manual labor in the gripping, positioning, and installation of brackets, robots can effectively improve work efficiency and safety.

[0003] However, related technologies often utilize unified control logic and fixed parameter configurations to control robots for the installation of utility tunnel supports. Since the installation process involves multiple action types, this method is prone to problems such as the accumulation of operational errors, affecting the accuracy of the utility tunnel support installation. Summary of the Invention

[0004] Therefore, it is necessary to provide a robot control method, device, equipment, and storage medium to address the aforementioned technical problems and improve the installation accuracy of pipe rack supports.

[0005] In a first aspect, this application provides a robot control method, including:

[0006] Obtain the target robot's current pose, current action type, and the target pose corresponding to the current action type;

[0007] Based on the current pose, current action type, and target pose, determine the motion control strategy for the target robot;

[0008] Control the movement of the target robot according to the motion control strategy;

[0009] If the action switching conditions corresponding to the current action type are met, switch the current action type.

[0010] In one embodiment, determining the motion control strategy of the target robot based on the current pose, the current action type, and the target pose includes: determining the target motion trajectory of the current action type based on the current pose and the target pose; and determining the motion control strategy of the target robot based on the current action type and the target motion trajectory corresponding to the current action type.

[0011] In one embodiment, a motion control strategy for the target robot is determined based on the current action type and the target motion trajectory corresponding to the current action type. This includes: when the current action type is a grasping action, a first control strategy is determined, which includes controlling the robot's robotic arm to move along the target motion trajectory corresponding to the grasping action, and controlling the robotic arm to grip the component to be installed when it reaches a preset grasping area; when the current action type is a positioning action, a second control strategy is determined, which includes controlling the robotic arm gripping the component to be installed to move along the target motion trajectory corresponding to the positioning action, and reducing the robotic arm's speed when it reaches a preset positioning area; and when the current action type is an installation action, a third control strategy is determined, which includes maintaining the contact force at the end of the robotic arm within a preset contact force range and controlling the robotic arm gripping the component to be installed to make micro-adjustments along the target motion trajectory corresponding to the installation action.

[0012] In one embodiment, controlling the movement of a target robot according to a motion control strategy includes: acquiring disturbance data of the target robot; determining a target control command based on the disturbance data and the motion control strategy; and controlling the movement of the target robot according to the target control command.

[0013] In one embodiment, the disturbance data includes load weight disturbance data and / or vibration disturbance data; based on the disturbance data and the motion control strategy, a target control command is determined, including any of the following: adjusting the target motion trajectory in the motion control strategy based on the load weight disturbance data, and determining the target control command based on the adjusted motion control strategy; determining an initial control command based on the motion control strategy, determining a vibration suppression command based on the vibration disturbance data, and determining a target control command based on the initial control command and the vibration suppression command; adjusting the target motion trajectory in the motion control strategy based on the load weight disturbance data, and determining an initial control command based on the adjusted motion control strategy, determining a vibration suppression command based on the vibration disturbance data, and determining a target control command based on the initial control command and the vibration suppression command.

[0014] In one embodiment, the method further includes: acquiring task data and sensing data of the target robot, wherein the sensing data includes at least one of visual sensing data, force sensing data and motion sensing data; and determining the target pose and the current pose of the target robot based on the task data and sensing data.

[0015] In one embodiment, the action switching condition includes the pose difference between the current pose and the target pose being less than a preset difference corresponding to the current action type.

[0016] Secondly, this application also provides a robot control device, comprising:

[0017] The first acquisition module is used to acquire the current pose, current action type, and target pose corresponding to the current action type of the target robot.

[0018] The determination module is used to determine the motion control strategy of the target robot based on the current pose, the current action type, and the target pose.

[0019] The control module is used to control the movement of the target robot according to the motion control strategy;

[0020] The switching module is used to switch the current action type when the action switching conditions corresponding to the current action type are met.

[0021] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0022] Obtain the target robot's current pose, current action type, and the target pose corresponding to the current action type;

[0023] Based on the current pose, current action type, and target pose, determine the motion control strategy for the target robot;

[0024] Control the movement of the target robot according to the motion control strategy;

[0025] If the action switching conditions corresponding to the current action type are met, switch the current action type.

[0026] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0027] Obtain the target robot's current pose, current action type, and the target pose corresponding to the current action type;

[0028] Based on the current pose, current action type, and target pose, determine the motion control strategy for the target robot;

[0029] Control the movement of the target robot according to the motion control strategy;

[0030] If the action switching conditions corresponding to the current action type are met, switch the current action type.

[0031] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0032] Obtain the target robot's current pose, current action type, and the target pose corresponding to the current action type;

[0033] Based on the current pose, current action type, and target pose, determine the motion control strategy for the target robot;

[0034] Control the movement of the target robot according to the motion control strategy;

[0035] If the action switching conditions corresponding to the current action type are met, switch the current action type.

[0036] The aforementioned robot control method, device, equipment, and storage medium acquire the target robot's current pose, current action type, and the target pose corresponding to the current action type. This allows the system to associate the current action type as an independent dimension with the pose information, enabling the system to distinguish the different requirements of different motion types on robot movement. This provides a data foundation for subsequently constructing differentiated motion control strategies. By determining the target robot's motion control strategy based on the current pose, current action type, and target pose, the system combines pose information with the current action type to generate a motion control strategy matching the current work stage, avoiding the accumulation of operational errors caused by using a unified control logic. By controlling the target robot's movement according to the motion control strategy and switching the current action type when the corresponding action switching conditions are met, the system avoids work interruptions or intermediate operational errors caused by improper action switching timing, ensuring the accuracy and reliability of the pipe gallery support installation process. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is an application environment diagram of the robot control method in one embodiment;

[0039] Figure 2 This is a flowchart illustrating a robot control method in one embodiment;

[0040] Figure 3 This is a flowchart illustrating the control steps of the target robot in one embodiment;

[0041] Figure 4 This is a flowchart of multimodal perception and operation status fusion in one embodiment;

[0042] Figure 5 This is a flowchart illustrating the robot control method in another embodiment;

[0043] Figure 6 This is a structural block diagram of the robot's control device in one embodiment;

[0044] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0046] The robot control method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, the input terminal of the controller 101 is connected to the sensor unit 102, and the output terminal of the controller 101 is connected to the execution unit 103. The sensor unit 102 is used to collect at least one of the following: the current pose of the target robot, the current action type, and the target pose corresponding to the current action type. The execution unit 103 may include multiple robotic arms, and the controller 101 can control the movement of each robotic arm to collaboratively achieve actions such as grasping, positioning, or installation.

[0047] For example, the target robot can be a multi-arm collaborative robot, such as a pipe rack support installation robot. The pipe rack support installation robot can perform actions such as grasping, lifting, positioning, hole alignment, installation, and fastening of pipe rack supports through the coordinated control of multiple robotic arms.

[0048] Robots are widely used in complex operations such as the installation of cable supports in utility tunnels. Replacing manual labor with robots for grasping, positioning, and installing supports effectively improves work efficiency and safety. However, related technologies often utilize unified control logic and fixed parameter configurations to control the robot for utility tunnel support installation. Since the installation process involves multiple action types, this method is prone to problems such as cumulative operational errors, decreased installation accuracy, abnormal operation interruptions, or intermediate operational issues, affecting the accuracy and reliability of utility tunnel support installation.

[0049] In the robot control method provided in this application, the sensor unit 102 is used to collect the current pose, current action type, and target pose corresponding to the current action type of the target robot. The controller 101 is used to determine the motion control strategy of the target robot based on the current pose, current action type, and target pose; according to the motion control strategy, it controls the execution unit 103 of the target robot to move, and switches the current action type when the action switching conditions corresponding to the current action type are met. Specifically, by acquiring the current pose, current action type, and target pose corresponding to the current action type of the target robot, the current action type is associated with the pose information as an independent dimension, enabling the system to distinguish the different requirements of different motion types on robot motion, providing a data foundation for subsequently constructing differentiated motion control strategies. By determining the motion control strategy of the target robot based on the current pose, current action type, and target pose, the pose information is combined with the current action type to generate a motion control strategy matching the current operation stage, avoiding the accumulation of operational errors caused by using a unified control logic. By controlling the target robot's movement according to the motion control strategy, and switching the current action type when the corresponding action switching conditions are met, the operation interruption or intermediate operation error caused by improper action switching timing is avoided, thus ensuring the accuracy and reliability of the pipe gallery support installation process.

[0050] In one exemplary embodiment, such as Figure 2 As shown, a robot control method is provided, which can be applied to... Figure 1 Taking controller 101 as an example, the following steps are included:

[0051] S201. Obtain the current pose, current action type, and target pose corresponding to the current action type of the target robot.

[0052] Here, the current pose can be understood as the pose and attitude of the target robot at the current moment or in the current control cycle; the current action type can be understood as the action category corresponding to the installation operation stage that the target robot is currently performing, and the current action type can include at least one of grasping action type, localization action type, and installation operation type; the target pose corresponding to the current action type can be understood as the expected position and expected attitude that the target robot should reach when the current action type is completed.

[0053] In some embodiments, task data and sensing data of the target robot can be acquired, and the sensing data includes at least one of visual sensing data, force sensing data, and motion sensing data; based on the task data and sensing data, the target pose and the current pose of the target robot are determined. The current pose may include the pose of the target robot's robotic arm and / or the pose of the pipe rack support held by the robotic arm.

[0054] For example, visual sensor data may include image data acquired by a visual sensor, from which the pose of the target robot's manipulator arm, the pose of the pipe rack support, and the installation reference pose can be extracted. Force sensing data may include end-effector force data acquired by a force / torque sensor, which may include force contact state data and force slip state data. Motion sensing data may include motion sensing data acquired by an encoder and vibration data acquired by an IMU (Inertial Measurement Unit), which may include joint position and joint velocity data, and vibration data may include vibration data of the manipulator arm or its end effector. Optionally, the pose of the pipe rack support and the installation reference pose can be determined by identifying mounting surfaces, holes, reference lines, or embedded parts in the image data. It is understood that when the pipe rack support is not held by the target robot, the pose of the pipe rack support is the position and orientation of the pipe rack support in the current task scenario; when the pipe rack support is held by the target robot, the pose of the pipe rack support is the position and orientation of the pipe rack support held by the manipulator arm.

[0055] For example, the task data may include at least one of the following: the theoretical installation position of the pipe rack support, the pipe rack support parameters, the clamping area data of the pipe rack support, and the preset distance between the positioning position and the installation position. Optionally, the pipe rack support parameters may include support layout parameters and support model parameters.

[0056] To facilitate understanding, the processes for determining the target pose and the current pose are illustrated below. It should be noted that this should not be interpreted as a limitation on the specific determination method.

[0057] For example, when the current action type is a grasping action, the target pose corresponding to the grasping action type, i.e. the grasping pose, can be determined based on the pose of the pipe rack support identified by the sensor, the pipe rack support parameters in the task data, and the support clamping area data.

[0058] For example, when the current action type is an installation action, the target pose, i.e., the installation pose, can be determined based on the installation reference pose identified by the sensors and the theoretical installation position in the task data. Optionally, the theoretical installation position can be used as the initial target pose corresponding to the installation action type; during the movement of the target robot, sensor data is continuously acquired, and the theoretical installation position is continuously corrected based on the sensor data to obtain the target pose corresponding to the installation action type. Of course, the installation reference pose identified by the sensors can also be directly used as the target pose of the installation action type; this application does not impose any limitations on this.

[0059] For example, when the current action type is a positioning action type, the target pose corresponding to the positioning action type, i.e. the positioning pose, can be obtained by moving outward a preset distance along the mounting surface based on the mounting pose.

[0060] For example, different sensor data can be normalized, and then fused according to the assigned weights of each sensor data point. Based on the task data and the fused sensor data, the current pose of the target robot can be determined. Understandably, by utilizing multi-dimensional perceptual information to complementarily correct the uncertainties of a single sensor, inaccurate pose determination can be avoided due to the failure of a single sensor or excessive deviation between the actual scene and the expected motion state of the theoretical task.

[0061] In some embodiments, the target robot's task data and sensor data can be input into the estimation model to obtain the target robot's current pose, current action type, and target pose corresponding to the current action type. The estimation model can be a traditional machine learning model or a neural network model, and this application does not impose any limitations on it.

[0062] S202. Determine the motion control strategy for the target robot based on the current pose, current action type, and target pose.

[0063] In some embodiments, the target motion trajectory of the current action type can be determined based on the current pose and the target pose; and the motion control strategy of the target robot can be determined based on the current action type and the target motion trajectory corresponding to the current action type.

[0064] In some embodiments, the current pose can be used as the starting point of the trajectory, the target pose can be used as the ending point of the trajectory, and the target motion trajectory can be constructed based on preset constraints.

[0065] For example, for a grasping action type, a grasping trajectory is constructed by using the current pose corresponding to the grasping mode as the trajectory starting point and the target pose corresponding to the grasping mode as the trajectory ending point. The target pose of the grasping mode is also the grasping pose corresponding to the preset grasping area. For example, for a positioning action type, a positioning trajectory is constructed by using the current pose corresponding to the positioning mode as the trajectory starting point and the target pose corresponding to the positioning mode as the trajectory ending point. The target pose of the positioning mode is also the pre-aligned pose at a preset distance from the target installation position (the target pose corresponding to the installation action type). For example, for an installation action type, a positioning trajectory is constructed by using the current pose corresponding to the installation mode as the trajectory starting point and the target pose corresponding to the installation mode as the trajectory ending point.

[0066] For example, the current pose and the target pose can be updated every preset time interval or preset control cycle to update the target motion trajectory.

[0067] For example, the target motion trajectory can be generated based on a preset motion template and online-generated parameters. Optionally, motion templates can be pre-stored, such as basic motion logic like approaching the support, closing the gripper, moving the support, low-speed alignment, approaching along the normal direction, micro-motion alignment, and maintaining contact force. During each actual operation, based on task data and the current and target poses identified by sensors, the specific trajectory points, velocity, acceleration, clamping force, and compensation amount are calculated online to construct the target motion trajectory. Based on the above construction method, the stability of the motion template is preserved, while avoiding the problem that fixed trajectories cannot adapt to different support positions, different installation references, and on-site deviations.

[0068] For example, during the generation of the target motion trajectory, a smooth trajectory function can be used to connect the trajectory start and end points. Optionally, a fifth-order polynomial or S-shaped velocity curve can be used to generate the target motion trajectory, ensuring continuous changes in position, attitude, velocity, and acceleration, thus avoiding impacts on the robotic arm during mode switching or when approaching the mounting surface. The generated target motion trajectory may include a reference position P. ref (t), Reference attitude R ref (t), reference speed V ref (t) and reference acceleration A ref (t), where t represents the trajectory execution time.

[0069] In some embodiments, when the current action type is a grasping action type, the motion control strategy can be determined as a first control strategy. The first control strategy includes controlling the robotic arm of the target robot to move along the target motion trajectory corresponding to the grasping action type, and controlling the robotic arm to grip the component to be installed when the robotic arm moves to the preset grasping area.

[0070] In some embodiments, when the current action type is a positioning action type, the motion control strategy can be determined as a second control strategy. The second control strategy includes controlling the robotic arm holding the component to be installed to move along the target motion trajectory corresponding to the positioning action type, and reducing the movement speed of the robotic arm when the robotic arm moves to a preset positioning area.

[0071] In some embodiments, when the current action type is an installation action type, the motion control strategy can be determined as a third control strategy. The third control strategy includes maintaining the contact force at the end of the robotic arm within a preset contact force range and controlling the robotic arm holding the component to be installed to make micro-adjustments along the target motion trajectory corresponding to the installation action type.

[0072] S203. Control the movement of the target robot according to the motion control strategy.

[0073] In some embodiments, a target control command can be generated according to a motion control strategy; and the target robot can be controlled to move according to the target control command.

[0074] In other embodiments, disturbance data of the target robot can be acquired; target control commands can be determined based on the disturbance data and motion control strategy; and the target robot can be controlled to move according to the target control commands.

[0075] In some embodiments, the target control command may include at least one of the following: end-effector velocity command, end-effector acceleration command, gripper opening and closing displacement command, gripper closing speed command, clamping force command, force control or impedance control parameters, joint torque feedforward compensation, end-effector pose command, and micro-motion step length of the robotic arm.

[0076] S204. If the action switching conditions corresponding to the current action type are met, switch the current action type.

[0077] In some embodiments, the action switching condition may include the pose difference between the current pose and the target pose being less than a preset difference corresponding to the current action type.

[0078] In some embodiments, the installation process of the utility tunnel supports is terminated when termination conditions are met. Termination conditions may include: the installation of a preset number of utility tunnel supports is completed; the support model of the current utility tunnel support is the preset last support model; and the installation of the current utility tunnel support is completed.

[0079] The robot control method provided in this application acquires the target robot's current pose, current action type, and the target pose corresponding to the current action type. It then associates the current action type as an independent dimension with the pose information, enabling the system to distinguish the different requirements of different motion types on robot movement. This provides a data foundation for subsequently constructing differentiated motion control strategies. By determining the target robot's motion control strategy based on the current pose, current action type, and target pose, the system combines pose information with the current action type to generate a motion control strategy matching the current work stage, avoiding the accumulation of operational errors caused by using a unified control logic. By controlling the target robot's movement according to the motion control strategy and switching the current action type when the corresponding action switching conditions are met, the system avoids work interruptions or intermediate operational errors caused by improper action switching timing, ensuring the accuracy and reliability of the pipe gallery support installation process.

[0080] Based on the technical solutions of the above embodiments, this application also provides an optional embodiment in which the control steps of the target robot are refined.

[0081] refer to Figure 3 The control steps of the target robot are shown, including:

[0082] S301. Obtain the disturbance data of the target robot.

[0083] For example, the disturbance data includes load weight disturbance data and / or vibration disturbance data. The load weight disturbance data is used to characterize the low-frequency sinking trend of the robotic arm end effector caused by weight deviation of the pipe rack support; the vibration disturbance data is used to characterize the high-frequency disturbance of the robotic arm caused by equipment vibration or environmental vibration. Optionally, the disturbance data may also include installation contact disturbance data to characterize contact disturbances during the installation contact process.

[0084] For example, initial disturbance data can be extracted from visual sensing data acquired by a visual sensor, force sensing data acquired by a force sensor, and motion sensing data acquired by a motion sensor (such as an encoder or IMU). Based on each initial disturbance data, load weight disturbance data, vibration disturbance data, and contact disturbance data can be determined. Optionally, vibration disturbance data can be determined based on vibration acceleration and angular velocity changes acquired by the IMU and residuals extracted by the encoder. Optionally, load weight disturbance data can be determined based on force / torque data, joint torque estimation, and current pose changes. Optionally, contact disturbance data can be determined based on force / torque data changes and visual pose deviations.

[0085] For example, to facilitate subsequent control, the current operation state can be represented as S(k), where k represents the k-th control cycle. S(k) can include the current pose of the robotic arm's end effector. Current position of the support body Corrected target pose End force Clamping force Vibration observation and current operating mode .

[0086] in, This represents the current pose of the robotic arm's end effector during the k-th control cycle; This represents the pose of the support body identified in the k-th control cycle; This represents the target pose of the i-th mounting position after being sensed and corrected by the field sensors. This indicates the external force acting on the end effector of the robotic arm; This indicates the current gripping force of the robotic arm's gripper; This represents the disturbance data obtained from actual observation; That is, the current action type, indicating whether the current action is in grab mode, location mode, or installation mode.

[0087] refer to Figure 4 The diagram illustrates the multimodal perception and operational status fusion process. First, visual sensing data from visual sensors, force sensing data from force / torque sensors, and motion sensing data from motion sensors (encoders and IMUs) are acquired. Visual sensing data can include pose information such as the robotic arm pose, the pipe rack support pose, and the installation reference pose; force sensing data can include force contact state data and force slip state data; and motion sensing data can include motion / vibration data. Second, dynamic weight allocation can be applied to different action types (i.e., motion modes or modalities). For example, force sensing weight is increased in grasping mode, visual weight is increased in positioning mode, and the combined weight of force and vision is increased in installation mode. Finally, based on the dynamically allocated weights, the visual, force, and motion sensing data are weighted and fused using Kalman filtering to obtain a unified operational status. This unified operational status includes pose estimation and disturbance observation. Pose estimation can include the current pose and the target pose for different action types. Disturbance observations may include load weight disturbance data and / or vibration disturbance data.

[0088] For example, in scenarios with limited space, insufficient lighting, or dust obstruction, the quality of visual sensing data can be determined by identifying confidence levels. When the confidence level falls below a preset threshold, the robotic arm or sensor perspective is adjusted for resampling; if multiple samplings still fail to meet the confidence level requirements, the current installation cycle is paused and a prompt for manual confirmation or recalibration is output.

[0089] For example, force / torque sensors can continuously collect force data at the end effector to determine whether the gripper is in contact with the support, whether the support is stably clamped, whether there is a tendency to slip, and whether contact, pressure, or jamming occurs during installation. Encoders can be used to acquire joint angles, joint velocities, and joint accelerations of the robotic arm, and combine them with a kinematic model to calculate the current pose of the end effector. IMUs can be used to collect changes in vibration acceleration and angular velocity of the end effector or robotic arm.

[0090] In some embodiments, the load weight disturbance data includes a weight deviation estimate; correspondingly, when the pipe rack support is stably held and kept stationary by the robotic arm gripper, the actual weight of the pipe rack support can be determined based on the force / torque data or joint torque collected by the force / torque sensor, and the actual weight can be compared with the nominal weight corresponding to the support model of the pipe rack support to obtain a weight deviation estimate.

[0091] For example, the weight deviation estimate can be expressed as:

[0092]

[0093] in, This represents the weight deviation estimate obtained during the k-th control cycle. This represents the equivalent vertical load obtained by converting force / torque data or joint torque. Indicates the nominal mass corresponding to the bracket model; It represents the acceleration due to gravity.

[0094] Among them, when A value greater than zero indicates that the actual weight of the support is higher than the nominal weight, making the robotic arm's end effector more prone to sinking during handling and positioning; when When the value is less than zero, it indicates that the actual weight of the support is lower than the nominal weight. The gravity compensation amount should be reduced to avoid overcompensation that could cause the end to rise.

[0095] S302. Determine the target control command based on the disturbance data and motion control strategy.

[0096] In some embodiments, the target motion trajectory in the motion control strategy can be adjusted based on the load weight disturbance data, and the target control command can be determined based on the adjusted motion control strategy.

[0097] Referring to the foregoing, the motion control strategy includes the target motion trajectory; correspondingly, the low-frequency compensation amount can be determined based on the load weight disturbance data; and the target motion trajectory can be adjusted based on the low-frequency compensation amount.

[0098] Optionally, the low-frequency compensation may include vertical position compensation, pitch attitude compensation, and joint torque feedforward compensation. Vertical position compensation is used to raise or correct the end-effector pose of the robotic arm in advance for the target motion trajectory; pitch attitude compensation is used to counteract the drooping caused by the shift in the support's center of gravity; and joint torque feedforward compensation is used to increase the corresponding joint output in advance in the drive commands, ensuring the robotic arm maintains trajectory stability even when the load changes.

[0099] For example, vertical position compensation amount It can be represented as:

[0100]

[0101] in, This indicates the vertical compensation amount caused by weight deviation; This represents the compensation coefficient between weight deviation and end settlement. This coefficient can be obtained through calibration experiments or automatically updated based on historical errors during repeated installations. This represents the current estimated weight deviation.

[0102] Understandably, by introducing load weight disturbance data, the target motion trajectory in the motion control strategy can be adjusted, thereby superimposing the corresponding weight deviation compensation onto the target motion trajectory, and thus obtaining the compensated target control command. This means that, based on the theoretical target pose, the end-effector target is corrected in advance according to the sinking trend caused by the actual load, so that when the robotic arm reaches the installation position, the actual installation reference point of the support coincides with the target installation position.

[0103] In some embodiments, an initial control command may be determined based on a motion control strategy, a vibration suppression command may be determined based on vibration disturbance data, and a target control command may be determined based on the initial control command and the vibration suppression command.

[0104] For example, vibration disturbance data may include current vibration disturbance data for the current control cycle and / or predicted vibration disturbance data for the next or future control cycle. Vibration disturbance data may include at least one of IMU vibration acceleration, encoder residuals, robotic arm end-effector pose fluctuations, and contact force fluctuations. The predicted disturbance data characterizes disturbances that may occur in the next control cycle or a future time window, and can be predicted based on current and historical vibration disturbance data.

[0105] For example, vibration disturbance data can be extracted from motion sensing data acquired by the IMU and encoder residuals. The vibration disturbance data can include vibration frequency, vibration amplitude, and vibration direction. When the vibration frequency is within the preset responsive frequency band of the robotic arm control system, a reverse compensation amount can be generated at the velocity / acceleration command layer to cancel out the predicted vibration in the end-effector motion command. When the vibration frequency exceeds the active control bandwidth of the robotic arm, its influence on the end-effector pose can be suppressed by reducing the end-effector velocity, improving trajectory smoothness, limiting acceleration abrupt changes, and enabling notch filtering or low-pass filtering.

[0106] For example, the predicted perturbation velocity can be determined based on the current perturbation data. and predicted perturbation acceleration It also generates compensated speed and acceleration commands.

[0107] Optional, speed command It can be represented as:

[0108]

[0109] in, This indicates the speed command output to the robotic arm controller during the k-th control cycle; This represents the reference velocity obtained from trajectory planning; This represents the velocity correction amount generated by pose error feedback; This indicates the predicted disturbance velocity for the next control cycle; This indicates the velocity disturbance compensation gain.

[0110] Optional, acceleration command It can be represented as:

[0111]

[0112] in, This indicates the acceleration command output to the robotic arm controller; This represents the reference acceleration obtained from trajectory planning; This represents the acceleration correction amount generated by the pose error feedback; This represents the predicted disturbance acceleration for the next control cycle; This represents the acceleration disturbance compensation gain.

[0113] In some embodiments, the target motion trajectory in the motion control strategy can be adjusted based on the load weight disturbance data, and an initial control command can be determined based on the adjusted motion control strategy, a vibration suppression command can be determined based on the vibration disturbance data, and a target control command can be determined based on the initial control command and the vibration suppression command.

[0114] S303. Control the movement of the target robot according to the target control command.

[0115] In the above steps, the end-effector target is corrected in advance based on the theoretical target pose and the sinking trend caused by the actual load, so that when the robotic arm reaches the installation position, the actual installation reference point of the support coincides with the target installation position. At the same time, by introducing vibration disturbance data and performing vibration suppression in advance at the velocity and acceleration command layer, end-effector jitter, repeated overshoot, and micro-oscillations within the positioning window can be reduced.

[0116] Based on the technical solutions of the above embodiments, this application also provides an optional embodiment in which a preprocessing step is added.

[0117] In some embodiments, the preprocessing steps may include initialization and calibration, task parameterization generation, and installation bit sequence output.

[0118] Optionally, before starting the bracket installation, the system first performs an initialization and calibration process to ensure that millimeter-level control is not limited by coordinate errors, sensor bias, or time asynchrony. This process may include coordinate system establishment, tool end-effector calibration, sensor zero-point calibration, and timing synchronization. During implementation, the system establishes a base coordinate system with the robotic arm base as a reference, and a tool coordinate system with the robotic arm gripper end-effector as a reference. The coordinate system of the vision sensor (or depth coordinate system) is then associated with the base coordinate system through extrinsic parameter calibration, allowing the visually recognized hole positions, edges, or baselines to be converted into target poses that the robotic arm can execute. Simultaneously, the force / torque sensor completes zero-point calibration under no external load conditions, and if the gripper clamping force is estimated through motor current or strain gauge, it is calibrated on a standard component, enabling the correlation between the clamping force setpoint and the actual clamping force to be used for closed-loop control. Finally, the system timestamps the visual frames, force sensor samples, and encoder samples to avoid fusion errors caused by time delays, thereby providing stable input for subsequent feedforward compensation and mode switching. The resolution of the visual sensor can be no less than 640×480, and the resolution of the force / torque sensor can be no less than 0.1N or equivalent torque accuracy to meet the requirements of subsequent micro-motion and contact control.

[0119] Optionally, after system initialization, the system enters the task parameterization phase. During implementation, the system receives utility tunnel support layout parameters, support model parameters, installation reference definitions, and allowable installation deviation thresholds. The utility tunnel support layout parameters may include installation position number, designed installation spacing, installation height, installation surface normal, reference line direction, and installation sequence; the support model parameters may include support dimensions, nominal support weight, installation hole spacing, support load-bearing capacity, clamping area, and allowable clamping force range; the installation reference definition may include using the utility tunnel wall, embedded parts, installation guide rails, laser reference lines, or visual recognition reference lines as installation references.

[0120] Referring to the foregoing, the task data may include the theoretical installation positions of the pipe rack supports. If there are multiple pipe rack supports, the task data may include a sequence of theoretical installation positions. A sequence of theoretical installation positions can be understood as the set of theoretical installation poses of the multiple pipe rack supports (also called pipe rack cable supports) to be installed within the pipe rack. The theoretical installation position sequence can be represented as: .in, Indicates the theoretical mounting bit sequence; For the first The theoretical installation position corresponding to each bracket to be installed; This indicates the total number of brackets required for this task. Each theoretical installation position... This may include: mounting position number Theoretical installation location coordinates Theoretical installation posture Corresponding bracket model Mounting surface normal Installation baseline direction Allowable positional deviation Permissible attitude deviation and installation action template .

[0121] For example, theoretical installation location coordinates Used to describe the three-dimensional position of the bracket installation reference point in the base coordinate system; theoretical installation posture. Used to describe the orientation of the bracket relative to the mounting surface; bracket model Used to match bracket size, hole spacing, weight, and clamping parameters; mounting surface normal. Used to determine the direction of the bracket near the mounting surface; installation reference line direction. Used to determine the arrangement of multiple supports along the pipe rack direction; allowable positional deviation and allowable attitude deviation Used for subsequent installation completion determination; installation action template This sequence is used to define the basic action type corresponding to each mounting position. Therefore, the mounting position sequence can be understood as a task-oriented data set specifying "where brackets 1 to n need to be installed, in what orientation, what bracket model should be used, in what order, and what tolerance level is allowed." After each bracket installation cycle is completed, the system retrieves the current theoretical mounting position from the mounting position sequence. It combines sensor data to determine the target pose for this round of grasping, positioning, and installation control, and automatically switches to the next theoretical installation position after installation is completed. By introducing a theoretical installation position sequence, when the bracket model, installation spacing, installation height, or pipe gallery installation benchmark changes, the system only needs to update the corresponding parameters in the installation position sequence to adapt to different engineering scenarios, without needing to redesign the control process or change the entire control strategy.

[0122] Based on the technical solutions of the above embodiments, this application also provides an optional embodiment in which the motion control strategies for different action types of the target robot are described in detail.

[0123] In some embodiments, the installation action state of the target robot may include grasping mode, positioning mode and installation mode, with different motion control strategies corresponding to different action types.

[0124] For ease of understanding, the motion control strategies for the grasping mode, positioning mode, and mounting mode are illustrated below.

[0125] In some embodiments, the goal of the grasping mode is to stably move the gripper of the robotic arm to a preset grasping area and reliably clamp the pipe rack support without causing deformation of the support. In grasping mode, the robotic arm of the target robot can be controlled to gradually approach the pipe rack support along the target motion trajectory (i.e., the grasping trajectory) corresponding to the grasping action type. During the approach, the pose acquired by the visual sensor is used as the basis for trajectory correction, and the change in force / torque at the end of the robotic arm is used as the basis for contact determination. Once the gripper of the robotic arm enters the preset grasping area, the gripper is controlled to close, and the clamping force closed-loop control and clamping force stabilization control are entered.

[0126] For example, the gripping force of the robotic arm can be stably controlled based on at least one of the following: the target pose corresponding to the gripping action type (i.e., the target pose of the robotic arm's end effector), the end effector velocity command, the end effector acceleration command, the gripper opening / closing displacement command, the gripper closing speed command, the target value of the gripping force, the upper limit of the gripping force, the lower limit of the gripping force, and force control or impedance control parameters. Specifically, the target pose of the robotic arm's end effector, the velocity command, and the acceleration command are used to control the robotic arm gripper to reach the preset gripping area; the gripper opening / closing displacement command is used to control the opening or closing of the robotic arm gripper; the gripper closing speed command is used to avoid impact caused by rapid gripping; the target value of the gripping force is used to control the robotic arm gripper to apply stable gripping to the pipe rack support; the upper limit and lower limit of the gripping force are used to prevent insufficient gripping leading to slippage or excessive gripping leading to support deformation; and the force control or impedance control parameters are used to provide compliant adjustment capability when the pipe rack support has slight deviations or uneven contact.

[0127] For example, the target clamping force value can be adaptively determined based on the support model and load weight disturbance data of the pipe gallery support. Among them, the target value of clamping force It can be represented as:

[0128]

[0129] in, This represents the target clamping force value for the k-th control cycle; This indicates the basic clamping force corresponding to the current bracket model; Indicators representing slippage risk; This represents the clamping force gain corresponding to the slippage risk; This indicates the clamping force gain corresponding to the load weight disturbance data; This represents the load weight disturbance data, also known as the weight deviation estimate.

[0130] For example, when the actual clamping force is lower than the lower limit of the clamping force, the closing displacement of the robotic arm gripper or the motor output can be increased; when the actual clamping force is higher than the upper limit of the clamping force, the closing displacement of the gripper or the motor output can be reduced; when the clamping force is within the set range and the fluctuation is less than the threshold, the clamping force can be determined to be stable.

[0131] For example, the action switching conditions corresponding to the grasping action type may include at least one of the following: the pose difference between the current pose and the target pose is less than the preset difference corresponding to the grasping action type; the gripping force is within a set range; the gripping force fluctuation is less than a preset threshold; there is no significant relative slippage between the pipe gallery support and the robotic arm gripper; and the force change at the end of the robotic arm conforms to the force characteristics of the already gripped support. If the action switching conditions are met, a successful grasping status can be output and the system can enter the positioning mode; if the grasping criteria are not met multiple times consecutively, the system will retreat to a safe distance, adjust the grasping posture, and attempt to grasp again. After exceeding the retry count, the system will proceed to manual confirmation or fault diagnosis.

[0132] In some embodiments, the objective of the positioning mode is to stably deliver the pipe rack support to the pre-alignment window in the presence of external disturbances; wherein the pre-alignment window is also the preset positioning area.

[0133] In positioning mode, the current pose can be used as the real-time feedback quantity, and the target pose can be used as the positioning target. The pose error between the current pose and the target pose is determined, and the target control command is determined by the pose error and feedforward compensation. In other words, the target pose and the current pose can be directly used for error calculation and closed-loop control in positioning mode, and the current pose and the target pose are compared in each control cycle, and the velocity, acceleration and compensation weights are dynamically adjusted according to the magnitude of the error.

[0134] For example, the end effector linear velocity in positioning mode can be 5 mm / s-30 mm / s. After the pipe rack support enters the pre-alignment window, the end effector linear velocity is reduced to 1 mm / s-5 mm / s, and the weighting of the vision sensor and force sensor is increased, thereby ensuring that the robotic arm end effector does not overshoot the target position due to inertia or disturbance.

[0135] For example, when switching from grasping mode to positioning mode, a smooth transition during the action switching process can be achieved through a continuous transition mechanism. Optionally, during mode switching, the upper limit of control speed, upper limit of acceleration, feedforward compensation weight, and force control parameters can all gradually change within a transition time window. For example, the upper limit of speed gradually decreases from the handling speed to the preset precision positioning speed, the feedforward compensation weight gradually increases from the first weight to the second weight in the positioning stage, and the control gain gradually switches from prioritizing grasping stability to prioritizing pose accuracy, with the first weight being less than the second weight. Optionally, the transition time window can be 50ms to 300ms, and the gradual change can be linear or S-shaped. By introducing a continuous transition mechanism, it is possible to avoid end effector jitter, attitude overshoot, or secondary oscillation of the support caused by sudden command changes.

[0136] For example, the action switching conditions corresponding to the positioning action type may include at least one of the following: the pose difference between the current pose and the target pose is less than the preset difference corresponding to the positioning action type; the installation reference point of the pipe gallery support enters the pre-alignment window; and there is no abnormal contact or lateral pressure at the end of the robotic arm. For example, when the action switching conditions corresponding to the positioning action type are met, the system switches to the installation mode; if the error cannot converge within a preset time, a backtracking positioning mechanism is triggered.

[0137] In some embodiments, the installation mode is a key step in achieving the final installation accuracy. In the installation mode, the pipe rack support can be controlled to gradually approach the installation surface, installation hole, or pre-embedded connection structure within the pre-alignment window, and the contact status can be monitored in real time by force / torque sensors.

[0138] In the installation mode, when the contact force exceeds the contact threshold or a preset abrupt change occurs in the contact force change rate, the system determines that the bracket has made contact with the installation structure. Subsequently, the control strategy transitions from position-priority to a force-position coupling compliant control method. Force-position coupling control refers to the system prioritizing control of the contact force in the normal direction of the installation surface to prevent the bracket from rigidly pressing against the installation surface, and prioritizing control of position errors in the tangential and attitude directions of the installation surface to enable the bracket to perform micro-alignment along the installation surface. For example, the contact force can be controlled to remain within a preset safe range in the normal direction of the installation surface; micro-displacement adjustments can be made in the tangential direction based on visual recognition or force feedback; and attitude fine-tuning can be performed in the rotation direction based on hole position deviation or mating surface deviation. Micro-adjustment can be understood as position and posture adjustment using preset micro-adjustment compensation, with a single micro-adjustment step size of 0.1mm-1.0mm, which can be refined to 100um-300um in high-precision scenarios.

[0139] For example, the current pose and target pose in the installation mode can be used as the basis for determining micro-motion alignment, and the end effector jitter of the robotic arm can be suppressed by perturbation data. The aforementioned clamping force closed-loop control mechanism enables the robotic arm gripper to stably clamp the pipe rack support. Thus, pose estimation, perturbation compensation, and clamping force control form a closed-loop connection during the installation phase.

[0140] For example, after each micro-motion, the position error, attitude error, and contact force state between the current pose and the target pose can be reassessed. When the error gradually decreases and the contact force curve stabilizes, the next micro-motion is performed; when both the position error and attitude error meet the installation completion threshold, the system determines that the installation is complete. If the number of micro-motions exceeds a preset threshold, the contact force remains abnormal, the support slips, or the target error cannot be further reduced, the installation anomaly handling process is initiated. Installation anomaly handling may include at least one of the following: reverting to the pre-aligned pose, re-identifying the installation reference, regenerating the micro-motion trajectory, reducing the speed and reinstalling, or prompting manual intervention.

[0141] Based on the technical solutions of the above embodiments, this application also provides an optional embodiment in which a process recording step is added.

[0142] In some embodiments, after each bracket installation is completed, key data during the installation process can be recorded and evaluated. The recorded key data may include at least one of the following: current installation position number, target pose, actual completed pose, final position deviation, final attitude deviation, clamping force curve, end contact force curve, weight deviation estimate, vibration disturbance observation, disturbance prediction, feedforward compensation, velocity / acceleration command change, mode switching time, and whether anomaly handling was triggered.

[0143] For example, the final installation deviation can be calculated based on the actual completed pose and the target pose corresponding to the installation action type. When the final installation deviation meets the allowable positional deviation... and allowable attitude deviation When the installation is successful, the system determines that the installation is qualified and writes the installation result into the quality record; when the deviation exceeds the threshold, the system initiates the review or rework process.

[0144] For example, when statistics show that the repeatability dispersion of multiple installation positions is gradually increasing, the source of error can be automatically analyzed. The repeatability dispersion is used to characterize the installation error. If the error originates from a fixed directional offset, the coordinate calibration parameters or installation position correction parameters are adjusted; if the error originates from weight deviation, the weight compensation coefficient is updated. If the error originates from enhanced vibration, reduce the positioning speed, improve trajectory smoothness, or adjust the high-frequency disturbance compensation gain. , If the error stems from unstable clamping, then update the baseline clamping force value. or slip risk gain Through the aforementioned parameter self-adjustment mechanism, millimeter-level positioning and installation capabilities can be maintained continuously during long-term repetitive operations, avoiding a gradual decrease in accuracy due to mechanical backlash, sensor drift, load changes, or changes in the field environment.

[0145] Based on the technical solutions of the above embodiments, this application also provides an optional embodiment in which the control method of the robot is described in detail.

[0146] refer to Figure 5 The diagram shown is a flowchart of a robot control method in another embodiment, including the following steps:

[0147] S501. Acquire task data and sensor data of the target robot. The sensor data includes at least one of visual sensor data, force sensor data, and motion sensor data.

[0148] S502. Based on the task data and sensor data, determine the target pose and the current pose of the target robot corresponding to the current action type.

[0149] S503. Determine the target motion trajectory of the current action type based on the current pose and the target pose.

[0150] S504. Determine the motion control strategy for the target robot based on the current action type and the target motion trajectory corresponding to the current action type.

[0151] For example, when the current action type is a grasping action type, the motion control strategy is determined to be the first control strategy. The first control strategy includes controlling the robotic arm of the target robot to move along the target motion trajectory corresponding to the grasping action type, and controlling the robotic arm to grip the component to be installed when the robotic arm moves to the preset grasping area.

[0152] For example, when the current action type is a positioning action type, the motion control strategy is determined to be the second control strategy. The second control strategy includes controlling the robotic arm holding the component to be installed to move along the target motion trajectory corresponding to the positioning action type, and reducing the movement speed of the robotic arm when the robotic arm moves to the preset positioning area.

[0153] For example, when the current action type is the installation action type, the motion control strategy is determined to be the third control strategy. The third control strategy includes keeping the contact force at the end of the robotic arm within a preset contact force range and controlling the robotic arm holding the part to be installed to make micro-adjustments along the target motion trajectory corresponding to the installation action type.

[0154] S505. Obtain the disturbance data of the target robot, including load weight disturbance data and vibration disturbance data.

[0155] S506. Adjust the target motion trajectory in the motion control strategy based on the load weight disturbance data.

[0156] S507. Determine the initial control command based on the adjusted motion control strategy.

[0157] S508. Determine the vibration suppression command based on the vibration disturbance data.

[0158] S509. Determine the target control command based on the initial control command and the vibration suppression command.

[0159] S510: Control the movement of the target robot according to the target control command.

[0160] In some embodiments, when installing supports in integrated utility tunnels or power tunnels, the system can first complete initial calibration and load task data such as support layout parameters, support models, and installation reference information, and generate an installation position sequence along the baseline. Then, a single installation cycle is entered. The system identifies the support and the installation reference, and fuses force and motion information to obtain reliable pose and disturbance data. During the gripping phase, stable support gripping is achieved through a clamping force closed loop, and weight deviation is estimated to generate feedforward compensation. During the positioning phase, smooth end-effector convergence is achieved under the superposition of feedforward compensation, and a smooth switching mechanism is used to enter the installation phase. During the installation phase, contact detection triggers force-position coupling control and performs micro-motion alignment until the final deviation meets the ±5mm requirement. After completion, the system records key curves and deviation results for quality traceability and parameter optimization, thereby achieving high precision, high stability, and universal adaptability to different support layout spacings, installation heights, and installation references in complex environments. Furthermore, it automatically enters the corresponding fault-tolerant processing flow in case of gripping failure, positioning over-limit, or installation abnormalities. The pose state can include the current pose of the target robot and the target pose corresponding to different action types; the disturbance data includes load weight disturbance data and vibration disturbance data.

[0161] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0162] Based on the same inventive concept, this application also provides a robot control device for implementing the robot control method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more robot control device embodiments provided below can be found in the limitations of the robot control method described above, and will not be repeated here.

[0163] In one exemplary embodiment, such as Figure 6 As shown, a robot control device is provided, including: a first acquisition module 601, a first determination module 602, a control module 603, and a switching module 604, wherein:

[0164] The first acquisition module 601 is used to acquire the current pose, current action type, and target pose corresponding to the current action type of the target robot.

[0165] The first determining module 602 is used to determine the motion control strategy of the target robot based on the current pose, the current action type and the target pose.

[0166] The control module 603 is used to control the movement of the target robot according to the motion control strategy.

[0167] The switching module 604 is used to switch the current action type when the action switching conditions corresponding to the current action type are met.

[0168] In some embodiments, the first determining module 602 includes: a first determining unit, configured to determine the target motion trajectory of the current action type based on the current pose and the target pose; and a second determining unit, configured to determine the motion control strategy of the target robot based on the current action type and the target motion trajectory corresponding to the current action type.

[0169] In some embodiments, the second determining unit includes: a first determining subunit, configured to determine a motion control strategy as a first control strategy when the current action type is a grasping action type, the first control strategy including controlling the robotic arm of the target robot to move along a target motion trajectory corresponding to the grasping action type, and controlling the robotic arm to clamp the component to be installed when the robotic arm moves to a preset grasping area; a second determining subunit, configured to determine a motion control strategy as a second control strategy when the current action type is a positioning action type, the second control strategy including controlling the robotic arm clamping the component to be installed to move along a target motion trajectory corresponding to the positioning action type, and reducing the movement speed of the robotic arm when the robotic arm moves to a preset positioning area; and a third determining subunit, configured to determine a motion control strategy as a third control strategy when the current action type is an installation action type, the third control strategy including maintaining the contact force at the end of the robotic arm within a preset contact force range, and controlling the robotic arm clamping the component to be installed to make micro-adjustments along the target motion trajectory corresponding to the installation action type.

[0170] In some embodiments, the control module 603 includes: an acquisition unit for acquiring disturbance data of the target robot; a third determination unit for determining a target control command based on the disturbance data and a motion control strategy; and a control unit for controlling the movement of the target robot according to the target control command.

[0171] In some embodiments, the disturbance data includes load weight disturbance data and / or vibration disturbance data; the third determining unit includes any one of the following: a fourth determining subunit, configured to adjust the target motion trajectory in the motion control strategy based on the load weight disturbance data, and determine a target control command based on the adjusted motion control strategy; a fifth determining subunit, configured to determine an initial control command based on the motion control strategy, determine a vibration suppression command based on the vibration disturbance data, and determine a target control command based on the initial control command and the vibration suppression command; a sixth determining subunit, configured to adjust the target motion trajectory in the motion control strategy based on the load weight disturbance data, and determine an initial control command based on the adjusted motion control strategy, determine a vibration suppression command based on the vibration disturbance data, and determine a target control command based on the initial control command and the vibration suppression command.

[0172] In some embodiments, the apparatus further includes: a second acquisition module, configured to acquire task data and sensing data of the target robot, wherein the sensing data includes at least one of visual sensing data, force sensing data, and motion sensing data; and a second determination module, configured to determine the target pose and the current pose of the target robot based on the task data and the sensing data.

[0173] In some embodiments, the action switching condition includes the pose difference between the current pose and the target pose being less than a preset difference corresponding to the current action type.

[0174] The various modules in the control device of the robot described above can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0175] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a robot control method. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0176] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0177] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0178] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0179] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0180] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0181] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0182] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for controlling a robot, characterized in that, The method includes: Obtain the current pose, current action type, and target pose corresponding to the current action type of the target robot; Based on the current pose, current action type, and target pose, determine the motion control strategy for the target robot; The target robot is controlled to move according to the motion control strategy described above; If the action switching conditions corresponding to the current action type are met, switch the current action type.

2. The method according to claim 1, characterized in that, The step of determining the motion control strategy for the target robot based on the current pose, the current action type, and the target pose includes: Based on the current pose and the target pose, determine the target motion trajectory of the current action type; Based on the current action type and the target motion trajectory corresponding to the current action type, the motion control strategy of the target robot is determined.

3. The method according to claim 2, characterized in that, The step of determining the motion control strategy of the target robot based on the current action type and the target motion trajectory corresponding to the current action type includes: When the current action type is a grasping action type, the motion control strategy is determined to be a first control strategy. The first control strategy includes controlling the robotic arm of the target robot to move along the target motion trajectory corresponding to the grasping action type, and controlling the robotic arm to clamp the component to be installed when the robotic arm moves to the preset grasping area. When the current action type is a positioning action type, the motion control strategy is determined to be a second control strategy. The second control strategy includes controlling the robotic arm holding the component to be installed to move along the target motion trajectory corresponding to the positioning action type, and reducing the movement speed of the robotic arm when the robotic arm moves to the preset positioning area. When the current action type is the installation action type, the motion control strategy is determined to be the third control strategy. The third control strategy includes maintaining the contact force at the end of the robotic arm within a preset contact force range and controlling the robotic arm holding the component to be installed to make micro-adjustments along the target motion trajectory corresponding to the installation action type.

4. The method according to claim 1, characterized in that, Controlling the target robot's movement according to the motion control strategy includes: Obtain the disturbance data of the target robot; Based on the disturbance data and the motion control strategy, the target control command is determined; The target robot is controlled to move according to the target control command.

5. The method according to claim 4, characterized in that, The disturbance data includes load weight disturbance data and / or vibration disturbance data; the determination of the target control command based on the disturbance data and the motion control strategy includes any one of the following: Based on the load weight disturbance data, the target motion trajectory in the motion control strategy is adjusted, and the target control command is determined based on the adjusted motion control strategy. Based on the motion control strategy, an initial control command is determined; based on the vibration disturbance data, a vibration suppression command is determined; and based on the initial control command and the vibration suppression command, a target control command is determined. Based on the load weight disturbance data, the target motion trajectory in the motion control strategy is adjusted, and based on the adjusted motion control strategy, an initial control command is determined. Based on the vibration disturbance data, a vibration suppression command is determined, and based on the initial control command and the vibration suppression command, a target control command is determined.

6. The method according to any one of claims 1-5, characterized in that, The method further includes: Acquire task data and sensor data of the target robot, wherein the sensor data includes at least one of visual sensor data, force sensor data and motion sensor data; Based on the task data and the sensor data, the target pose and the current pose of the target robot are determined.

7. The method according to any one of claims 1-5, characterized in that, The action switching condition includes the pose difference between the current pose and the target pose being less than a preset difference corresponding to the current action type.

8. A control device for a robot, characterized in that, The device includes: The first acquisition module is used to acquire the current pose, current action type, and target pose corresponding to the current action type of the target robot. The determination module is used to determine the motion control strategy of the target robot based on the current pose, the current action type, and the target pose. A control module is used to control the movement of the target robot according to the motion control strategy; The switching module is used to switch the current action type when the action switching conditions corresponding to the current action type are met.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-7.

Images