Robotic active damping method, electronic device, medium, and system

Abstract

The application relates to the technical field of robots and provides a robot active damping method, electronic equipment, a medium and a system. The method comprises the following steps: collecting image data of a robot in the advancing direction of a motion route; predicting an overall change trend of the robot in a first dimension based on the image data; determining a current motion state of a first end; determining a reference motion state of the first end based on the overall change trend and the current motion state; and controlling an active damping unit associated with the first end to perform an active damping operation based on the reference motion state of the first end. In this way, active damping is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robots, and in particular to a robot active damping method, an electronic device, a medium and a system. BACKGROUND

[0002] The robot damping scheme in the prior art generally adopts passive damping, such as damping through a suspension device and a mechanical structure. However, the passive damping scheme is difficult to adapt to the increasingly complex application environment of robots, and it is difficult to maintain stability and avoid shaking in scenarios with high requirements for accurate control of robots. The robot active damping scheme in the prior art relies on the combination of a specially designed active damping mechanical structure and system control, so that special adaptation is often required when the hardware implementation structure of the robot changes, which is not conducive to meeting the demand for rapid iteration of robot technology. For example, a Chinese patent application with the application publication number CN116766851A discloses a mobile robot body stabilizing system, and the stabilizing mechanism includes a specially designed stabilizing damper and a sliding block and a sliding rail. When the ground clearance of the left and right wheels of the vehicle body is different, the sliding block is slid on the sliding rail to keep the ground clearance equal.

[0003] Therefore, the present application provides a robot active damping method, an electronic device, a medium and a system, which realizes a robot active damping scheme with good expansibility, can not only meet the requirements of stability and avoidance of shaking in the increasingly complex application environment of robots, but also can be conveniently popularized to robots with various hardware implementation structures. SUMMARY

[0004] In a first aspect, the present application provides a robot active damping method. The robot active damping method comprises: collecting, by a vision sensor of a robot, image data of the robot in a forward direction of a motion route, the image data at least indicating an overall change of the robot in a first dimension parallel to a gravity direction when the robot advances along the motion route; predicting, by a processor of the robot, an overall change trend of the robot in the first dimension based on the image data; determining, by a built-in acceleration displacement sensor of a first end of the robot, a current motion state of the first end; determining, by the processor, a reference motion state of the first end based on the overall change trend of the robot in the first dimension and the current motion state of the first end, wherein the reference motion state of the first end indicates that the change of the first end in the first dimension is less than a preset threshold; and controlling, by a controller of the robot, an active damping unit associated with the first end to perform an active damping operation based on the reference motion state of the first end, so that the motion state of the first end during the advancement of the robot along the motion route is maintained as the reference motion state of the first end.

[0005] By the first aspect of the present application, by using the separated information collection mode, the image data of the robot in the advancing direction of the motion route is collected through the visual sensor, and the current motion state of the first end of the robot is determined through the built-in acceleration displacement sensor of the first end of the robot, respectively; on the basis of the collected original data (i.e. the image data of the robot in the advancing direction of the motion route and the current motion state of the first end), the image data is processed, the overall change trend of the robot in the first dimension is predicted based on the image data, the current motion state of the first end is superimposed with the change of the robot in the first dimension in the overall sense, which can better offset the influence of various vibration phenomena that may be encountered in the motion route of the robot, improve the active damping effect, form a closed loop from information collection to active damping control, and the visual sensor can be matched with any first end to complete the separated information collection and finally focus on the active damping control of the first end itself, which means that various hardware implementation structure robots can be adapted, any part or joint on the robot can be selected as the first end, so as to construct the active damping scheme supporting the relative static state of the first end in the first dimension; it can be conveniently extended to simultaneously implement active damping on the ends of multiple robots, different preset threshold values can be set for the ends of different robots, which can better adapt to the fine control demand; without the combination of specially designed active damping structure and system control, the separated information collection and coordinated control mechanism can adapt to the active damping demand of the end of any robot.

[0006] In a possible implementation manner of the first aspect of the present application, the current motion state of the first end includes the current position of the first end, the current speed of the first end, and the current acceleration of the first end.

[0007] In a possible implementation manner of the first aspect of the present application, the reference motion state of the first end is determined by the processor based on the overall change trend of the robot in the first dimension and the current motion state of the first end, including: the overall motion state of the robot relative to the first end is determined by the processor based on the overall change trend of the robot in the first dimension, then the overall motion state of the robot relative to the first end is superimposed to the current motion state of the first end, so as to obtain the expected motion state of the first end; the reference motion state of the first end is determined based on the component of the expected motion state of the first end in the first dimension.

[0008] In one possible implementation of the first aspect of this application, the active damping operation performed by the active damping unit associated with the first end is used to counteract the change of the first end in the first dimension caused by the component of the expected motion state of the first end in the first dimension.

[0009] In one possible implementation of the first aspect of this application, the robot is a half-body humanoid robot, the robot including a head, an upper body, a waist and a movable base, the waist being fixed to the movable base, the first end being the upper body or a joint included in the upper body, the visual sensor being deployed on the head, and the active damping unit associated with the first end being deployed on the waist and the movable base.

[0010] In one possible implementation of the first aspect of this application, the robot is a fully humanoid robot, the robot including a head, an upper body, a waist and legs, the first end effector being the upper body or a joint included in the upper body, the visual sensor being deployed on the head, and the active damping unit associated with the first end effector being deployed on the waist and the legs.

[0011] In one possible implementation of the first aspect of this application, the image data also indicates terrain change information in the direction of travel of the movement route.

[0012] In one possible implementation of the first aspect of this application, the robot's processor is configured to input the image data into a trained artificial intelligence model to predict the robot's overall trend of change in the first dimension.

[0013] In one possible implementation of the first aspect of this application, the current motion state of the first end effector is determined based on the robot's internal three-dimensional spatial coordinate system, and the active damping operation performed by the active damping unit associated with the first end effector includes the coordination between multiple joints of the robot.

[0014] Secondly, embodiments of this application also provide a computer device, the computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a method according to any of the above-mentioned implementations.

[0015] Thirdly, embodiments of this application also provide a computer-readable storage medium storing computer instructions that, when executed on a computer device, cause the computer device to perform a method according to any of the above-described implementations.

[0016] Fourthly, embodiments of this application also provide a computer program product, the computer program product including instructions stored on a computer-readable storage medium, which, when executed on a computer device, cause the computer device to perform a method according to any of the above-described aspects.

[0017] Fifthly, this application also provides a robot active vibration damping system. The robot active vibration damping system includes: a vision sensor deployed on the robot's head, used to acquire image data of the robot in the forward direction of its movement path, the image data indicating at least the overall change of the robot in a first dimension parallel to the direction of gravity as it moves along the movement path; a processor deployed on the robot, used to predict the overall change trend of the robot in the first dimension based on the image data; a built-in acceleration-displacement sensor deployed on the robot's first end, used to determine the current motion state of the first end, wherein the processor is used to determine a reference motion state of the first end based on the overall change trend of the robot in the first dimension and the current motion state of the first end, wherein the reference motion state of the first end indicates that the change of the first end in the first dimension is less than a preset threshold; and a controller deployed on the robot, used to control an active vibration damping unit associated with the first end to perform active vibration damping operations based on the reference motion state of the first end, thereby maintaining the motion state of the first end as the reference motion state of the first end during the robot's movement along the movement path.

[0018] Through the fifth aspect of this application, a separate information acquisition method is used. A visual sensor acquires image data of the robot in the forward direction of its movement path, and a built-in acceleration-displacement sensor at the robot's first end effector determines the current motion state of the first end effector. Based on the acquired raw data (i.e., image data of the robot in the forward direction of its movement path and the current motion state of the first end effector), the image data is processed to predict the overall trend of the robot's change in the first dimension. By superimposing the current motion state of the first end effector with the overall change of the robot in the first dimension, the effects of various vibrations that may be encountered along the robot's movement path can be better offset, thus improving active vibration damping. The system achieves the following results: It forms a closed loop from information acquisition to active vibration damping control, and allows vision sensors to be paired with any first end effector to complete separate information acquisition and finally focus on the active vibration damping control of the first end effector itself. This means it can be adapted to robots with various hardware implementation structures, and any part or joint on the robot can be selected as the first end effector to construct an active vibration damping scheme that supports the first end effector to remain relatively stationary in the first dimension. It can be easily extended to implement active vibration damping on the end effectors of multiple robots simultaneously, and different preset thresholds can be set for the end effectors of different robots, thus better adapting to the needs of fine control. It does not require a specially designed active vibration damping structure combined with system control, but rather utilizes a separate information acquisition and coordinated control mechanism to adapt to the active vibration damping needs of any robot end effector. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating an active vibration reduction method for a robot provided in this application embodiment;

[0021] Figure 2 A flowchart illustrating a method for determining a reference motion state of a first end-effector, provided in an embodiment of this application;

[0022] Figure 3 This is a schematic diagram of an active vibration damping system for a robot provided in an embodiment of this application;

[0023] Figure 4 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application. Detailed Implementation

[0024] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0025] It should be understood that in the description of this application, "at least one" means one or more, and "multiple" means two or more. In addition, the words "first," "second," etc., unless otherwise stated, are used only for the purpose of distinguishing descriptions and should not be construed as indicating or implying relative importance or order.

[0026] Figure 1 This is a schematic flowchart illustrating a robot active vibration reduction method provided in an embodiment of this application. Figure 1 As shown, the robot active vibration reduction method includes the following steps.

[0027] Step S101: Using the robot's vision sensor, acquire image data of the robot in the forward direction of the movement path. The image data at least indicates the overall change of the robot in a first dimension parallel to the direction of gravity as it moves along the movement path.

[0028] Step S103: Based on the image data, the robot's processor predicts the overall change trend of the robot in the first dimension.

[0029] Step S105: Determine the current motion state of the first end effector by using the built-in acceleration displacement sensor of the first end effector of the robot.

[0030] Step S107: The processor determines a reference motion state of the first end effector based on the overall change trend of the robot in the first dimension and the current motion state of the first end effector, wherein the reference motion state of the first end effector indicates that the change of the first end effector in the first dimension is less than a preset threshold.

[0031] Step S109: The robot's controller controls the active damping unit associated with the first end effector to perform active damping operation based on the reference motion state of the first end effector, so that the motion state of the first end effector during the robot's movement along the motion path remains the reference motion state of the first end effector.

[0032] See Figure 1The system acquires image data of the robot along its forward movement path using a visual sensor, and determines the current motion state of the robot's first end effector using a built-in accelerometer-displacement sensor. Then, based on the image data, it predicts the overall trend of the robot's movement in the first dimension, and further determines a reference motion state of the first end effector based on the overall trend of the robot's movement in the first dimension and the current motion state of the first end effector. Therefore, data acquisition is performed separately; that is, image data of the robot along its forward movement path is acquired separately using a visual sensor, and the current motion state of the first end effector is determined using a built-in accelerometer-displacement sensor. Generally, the visual sensor can be deployed on the robot's head or other locations, such as the chest, to acquire image data of the robot along its forward movement path as it moves along the path. The built-in accelerometer-displacement sensor at the robot's first end effector can be flexibly varied according to the definition of the first end effector. For example, when the first end effector refers to a joint in the robot's arm, such as the elbow, wrist, or shoulder joint, the corresponding built-in acceleration and displacement sensor is also located at that joint to detect its current motion state. Similarly, when the first end effector refers to a joint in the robot's leg, such as the knee joint, the corresponding built-in acceleration and displacement sensor is also located at that joint to detect its current motion state. The first end effector can also refer to the entire upper body of the robot, such as the main body of the upper body; in this case, the corresponding built-in acceleration and displacement sensor is used to detect the current motion state of the upper body, such as its movement speed and acceleration. It should be understood that the motion state of the first end effector can include not only velocity and acceleration in the sense of linear motion but also velocity and acceleration in the sense of rotational motion. Thus, through a separate information acquisition method, the robot's vision sensor can be paired with the built-in acceleration and displacement sensor of any of the robot's first end effectors to acquire raw data.

[0033] Continue reading Figure 1Based on the acquisition of raw data through a separate information acquisition method (the robot's vision sensor can be paired with the built-in acceleration and displacement sensors of any of the robot's first end caps), the robot's processor can be used to parse and process the raw data, extracting relevant information for subsequent processing. In some embodiments, considering the limited computing resources and processing capabilities of the robot's built-in processor, external computing resources, such as cloud computing and cloud service platforms, can be connected to the robot via wired or wireless means to provide additional processing capabilities. The image data at least indicates the overall change of the robot along the motion path in a first dimension parallel to the direction of gravity. Here, the direction of gravity can generally be understood as the vertical direction, and the robot's motion path is generally along the ground plane, i.e., the horizontal direction. Therefore, depending on the application environment of the robot, it may encounter various terrain changes as it moves along the motion path, such as slopes that cause a sudden increase in elevation at the landing point, pits that cause a sudden decrease in elevation at the landing point, and upward or downward steps. When a robot encounters similar terrain changes while moving forward, without active damping, it may sway during movement. Furthermore, focusing on the end effector as a focal point, this can cause localized instability, such as sudden or repeated changes in the end effector's position in the first dimension. Therefore, based on the collected raw data (i.e., image data of the robot's movement along its path and the current motion state of the end effector), the image data is processed to predict the robot's overall trend of change in the first dimension. Predicting this trend means anticipating the potential changes in the first dimension (i.e., the direction of gravity or the vertical direction) as the robot continues along its path. For example, by analyzing the image data, it can be predicted that the robot will encounter an uphill slope or step on a pothole at some point in the future (considering its current speed). This allows for anticipating the resulting changes in the first dimension, such as a sudden rise or fall of the robot's center of gravity, providing a basis for active damping and enabling better mitigation.

[0034] Continue reading Figure 1Based on the overall change trend of the robot in the first dimension and the current motion state of the first end effector, a reference motion state of the first end effector is determined. The reference motion state of the first end effector indicates that the change of the first end effector in the first dimension is less than a preset threshold. Here, the preset threshold limits the tolerable magnitude of change in the first dimension, i.e., sets an acceptable level of vibration. By setting a sufficiently small preset threshold, the change of the first end effector in the first dimension can be limited to a sufficiently small size, thus achieving a relatively static effect, i.e., canceling the vibration impact of the first end effector in the first dimension. Therefore, using the overall change trend of the robot in the first dimension obtained through the above analysis and the current motion state of the first end effector acquired by the built-in acceleration and displacement sensor, the reference motion state of the first end effector can be determined. Then, through the robot's controller, based on the reference motion state of the first end effector, the associated active damping unit of the first end effector is controlled to perform active damping operation, thereby ensuring that the motion state of the first end effector remains the reference motion state of the first end effector while the robot moves along the motion path. Thus, by acquiring raw data through a separate information acquisition method (the robot's vision sensor can be paired with the built-in acceleration and displacement sensors of any of the robot's first end effectors), the robot's processor can parse and process the raw data, extracting relevant information for further processing, and finally obtaining the reference motion state of the first end effector. This reference motion state of the first end effector is used to eliminate changes in the first end effector in the first dimension, thereby achieving the effect of keeping the first end effector relatively stationary in the first dimension, achieving the purpose of active vibration reduction. Furthermore, because the overall change trend of the robot in the first dimension and the current motion state of the first end effector are considered together, the current motion state of the first end effector is superimposed with the changes in the first dimension in the overall sense of the robot. This can better offset the impact of various vibration phenomena that may be encountered along the robot's movement path, improving the active vibration reduction effect.

[0035] Continue reading Figure 1 Because a separate approach is used for data acquisition, the vision sensor can be paired with any end effector. Furthermore, the control system simultaneously considers the robot's overall change trend in the first dimension and the current motion state of the end effector. The generated control signal is then used to control the active damping unit associated with the end effector to perform active damping operations. The final result is that the motion state of the end effector remains the reference motion state of the end effector as the robot moves along the specified path. Therefore, Figure 1The robot active vibration reduction method shown forms a closed loop from information acquisition to active vibration reduction control. It allows a vision sensor to be paired with any first end effector to perform separate information acquisition and finally focus on the active vibration reduction control of the first end effector itself. This means it can be adapted to robots with various hardware implementations, such as half-humanoid robots or fully humanoid robots. Furthermore, any part or joint on the robot can be selected as the first end effector to construct an active vibration reduction scheme that supports the first end effector to remain relatively stationary in a first dimension. For example, when the robot's dexterous hand needs to grip or lift a target object, such as in laboratory applications where the robot retrieves test tubes or other experimental materials, this dexterous hand can be designated as the first end effector, ensuring that the dexterous hand remains relatively stationary in the first dimension during the robot's movement, preventing damage or loss of experimental materials. As another example, when a robot's mobile base is equipped with a tray holding a large number of workpieces, such as in automotive industrial automation applications where the robot transfers a large number of parts, this tray can be designated as the first end effector, ensuring that the tray remains relatively stationary in the first dimension during the robot's movement, preventing workpieces from falling. Additionally, both the first and second ends can be simultaneously designated as active damping targets, thus allowing for the utilization of... Figure 1 The robot active vibration damping method shown applies active vibration damping to both the first and second ends to maintain relative stillness in the first dimension. Using a robot's dexterous hand as an example, the robot's left hand can be designated as the first end effector, and the right hand as the second end effector, both referencing... Figure 1The illustrated active vibration damping method for a robot can be specifically described by the following steps for the active vibration damping method of the second end effector: The robot's vision sensor acquires image data of the robot's forward movement along its path, the image data indicating at least the overall change of the robot in a first dimension parallel to the direction of gravity as it moves along the path; the robot's processor predicts the overall trend of change of the robot in the first dimension based on the image data; the built-in acceleration-displacement sensor of the second end effector determines the current motion state of the second end effector; the processor determines a reference motion state of the second end effector based on the overall trend of change of the robot in the first dimension and the current motion state of the second end effector, wherein the reference motion state of the second end effector indicates that the change of the second end effector in the first dimension is less than a preset threshold; the robot's controller controls the associated active vibration damping unit of the second end effector to perform active vibration damping operations based on the reference motion state of the second end effector, thereby maintaining the motion state of the second end effector as the reference motion state of the second end effector during the robot's movement along the path. Thus, assuming active vibration damping operations need to be implemented on the end effectors of two or more robots... Figure 1 The illustrated active vibration damping method for robots can be easily extended to simultaneously apply active vibration damping to the end effectors of multiple robots. The final effect is that not only does the motion state of the first end effector remain as a reference motion state during the robot's movement along the motion path, but the motion state of the second end effector also remains as a reference motion state during the same period. Therefore, not only is the change in the first end effector in the first dimension less than a preset threshold, but the change in the second end effector in the first dimension is also less than the preset threshold. Furthermore, different preset thresholds can be set for different robot end effectors, which better adapts to fine-grained control requirements. For example, a first preset threshold can be set for the first end effector, and a second preset threshold can be set for the second end effector. The final effect is that the change in the first end effector in the first dimension is less than the first preset threshold, and the change in the second end effector in the first dimension is also less than the second preset threshold. It should be understood that the above-mentioned active vibration reduction method for robots does not require a specially designed active vibration reduction structure combined with system control. Instead, it utilizes a separate information acquisition and coordination control mechanism, which can adapt to the active vibration reduction needs of any robot end effector. It can also achieve the effect of upper body stability of the robot through the mutual coordination between robot joints, such as controlling the legs and waist through algorithms.

[0036] In short, Figure 1The illustrated active vibration reduction method for robots utilizes a separate information acquisition approach. A visual sensor acquires image data of the robot's forward movement path, and a built-in acceleration-displacement sensor at the robot's end effector determines the current motion state of the first end effector. Based on the acquired raw data (i.e., image data of the robot's forward movement path and the current motion state of the first end effector), the image data is processed to predict the overall trend of the robot's movement in the first dimension. By superimposing the current motion state of the first end effector with the overall change in the first dimension of the robot, the method can better offset the effects of various vibrations that may be encountered along the robot's movement path, thereby improving the active vibration reduction effect. The vibration reduction effect forms a closed loop from information acquisition to active vibration reduction control. Furthermore, it allows vision sensors to be paired with any end effector to perform separate information acquisition and finally focus on the active vibration reduction control of the end effector itself. This means it can be adapted to robots with various hardware implementations, allowing any part or joint on the robot to be selected as the end effector, thereby constructing an active vibration reduction scheme that supports the end effector to remain relatively stationary in the first dimension. It can be easily extended to simultaneously implement active vibration reduction on the end effectors of multiple robots, and different preset thresholds can be set for different robot end effectors, thus better adapting to refined control requirements. It does not require a specially designed active vibration reduction structure combined with system control; instead, it utilizes a separate information acquisition and coordinated control mechanism to adapt to the active vibration reduction needs of any robot end effector. Therefore, it not only meets the stability and anti-shaking requirements of increasingly complex robot application environments but also can be easily extended to robots with various hardware implementations.

[0037] Figure 2 This is a flowchart illustrating a method for determining a reference motion state of a first end-effector, as provided in an embodiment of this application. Figure 2 As shown, the method for determining the reference motion state of the first end includes the following steps.

[0038] Step S201: Collect image data of the robot's forward direction along the movement path using a vision sensor.

[0039] Step S203: Based on image data, predict the overall trend of the robot's change in the first dimension parallel to the direction of gravity.

[0040] Step S205: Based on the overall change trend of the robot in the first dimension, determine the overall motion state of the robot relative to the first end effector.

[0041] Step S210: Determine the current motion state of the first end by using the built-in acceleration displacement sensor at the first end.

[0042] Step S220: Superimpose the overall motion state of the robot relative to the first end effector onto the current motion state of the first end effector to obtain the expected motion state of the first end effector.

[0043] Step S230: Determine the reference motion state of the first end based on the component of the expected motion state of the first end in the first dimension.

[0044] Figure 2 The method described herein for determining a reference motion state of a first end effector utilizes a separate information acquisition approach. A vision sensor is used to acquire image data of the robot in the forward direction of its movement path, and a built-in accelerometer / displacement sensor at the first end effector is used to determine the current motion state of the first end effector. Based on the acquired raw data (i.e., image data of the robot in the forward direction of its movement path and the current motion state of the first end effector), the image data is processed to predict the overall trend of the robot's change in the first dimension. Using the overall trend of the robot's change in the first dimension obtained through analytical processing, and the current motion state of the first end effector acquired by the built-in accelerometer / displacement sensor, the reference motion state of the first end effector can be determined. In this way, the current motion state of the first end effector is superimposed with the changes in the first dimension in the overall sense of the robot. This can better offset the impact of various vibration phenomena that may be encountered on the robot's motion path, improve the active vibration reduction effect, help to form a closed loop from information acquisition to active vibration reduction control, and allow vision sensors to be paired with any first end effector to complete separate information acquisition and finally focus on the active vibration reduction control of the first end effector itself. This means that robots with various hardware implementation structures can be adapted.

[0045] See Figure 1 and Figure 2In one possible implementation, the current motion state of the first end effector includes the current position, current velocity, and current acceleration of the first end effector. The built-in acceleration-displacement sensor of the robot's first end effector can be flexibly varied according to the definition of the first end effector. For example, when the first end effector refers to a joint in the robot's arm, such as the elbow, wrist, or shoulder joint, the corresponding built-in acceleration-displacement sensor is located at the corresponding joint to detect its current motion state. Similarly, when the first end effector refers to a joint in the robot's leg, such as the knee joint, the corresponding built-in acceleration-displacement sensor is located at the corresponding joint to detect its current motion state. The first end effector can also refer to the entire upper body of the robot, such as the main body of the upper body; in this case, the corresponding built-in acceleration-displacement sensor is also used to detect the current motion state of the upper body, such as its movement speed and acceleration. It should be understood that the motion state of the first end effector can include not only velocity and acceleration in the sense of linear motion but also velocity and acceleration in the sense of rotational motion. In this way, through a separate information acquisition method, the robot's vision sensor can be paired with the built-in acceleration and displacement sensor at any of the robot's first end caps to acquire raw data.

[0046] In some embodiments, the processor determines a reference motion state of the first end effector based on the overall change trend of the robot in the first dimension and the current motion state of the first end effector. This includes: determining the overall motion state of the robot relative to the first end effector based on the overall change trend of the robot in the first dimension; then, superimposing the overall motion state of the robot relative to the first end effector onto the current motion state of the first end effector to obtain the expected motion state of the first end effector; and determining the reference motion state of the first end effector based on the component of the expected motion state of the first end effector in the first dimension. Thus, the reference motion state of the first end effector is used to eliminate changes in the first end effector in the first dimension, thereby achieving the effect of maintaining the relative stillness of the first end effector in the first dimension and achieving active vibration reduction. Furthermore, by superimposing the current motion state of the first end effector onto the changes in the first dimension in the overall sense of the robot, the effects of various vibration phenomena that may be encountered along the robot's movement path can be better offset, improving the active vibration reduction effect.

[0047] In some embodiments, the active damping operation performed by the active damping unit associated with the first end effector is used to counteract the change of the first end effector in the first dimension caused by the component of the expected motion state of the first end effector in the first dimension. Thus, by superimposing the current motion state of the first end effector with the change in the first dimension in the overall sense of the robot, the effects of various vibrations that may be encountered along the robot's motion path can be better counteracted, improving the active damping effect. This constitutes a closed loop from information acquisition to active damping control, and allows a vision sensor to be paired with any first end effector to complete separate information acquisition and finally focus on the active damping control of the first end effector itself. This means that it can be adapted to robots with various hardware implementation structures, and any part or joint on the robot can be selected as the first end effector, thereby constructing an active damping scheme that supports the first end effector to remain relatively stationary in the first dimension.

[0048] In one possible implementation, the robot is a half-body humanoid robot, comprising a head, an upper body, a waist, and a movable base. The waist is fixed to the movable base. The first end effector is the upper body or a joint included in the upper body. A vision sensor is deployed on the head, and an active damping unit associated with the first end effector is deployed on the waist and the movable base. This forms a closed loop from information acquisition to active damping control, and allows the vision sensor to be paired with any first end effector to perform separate information acquisition and finally focus on the active damping control of the first end effector itself. This means that robots with various hardware implementations can be adapted.

[0049] In one possible implementation, the robot is a fully humanoid robot comprising a head, upper body, waist, and legs. The first end effector is the upper body or a joint included in the upper body. The vision sensor is deployed on the head, and the active damping unit associated with the first end effector is deployed on the waist and legs. This forms a closed loop from information acquisition to active damping control, and allows the vision sensor to be paired with any first end effector to perform separate information acquisition and ultimately focus on the active damping control of the first end effector itself. This means that robots with various hardware implementations can be adapted.

[0050] In one possible implementation, the image data also indicates terrain change information in the direction of travel along the movement path. Depending on the application environment of the robot, various terrain changes may be encountered as the robot moves along the movement path, such as slopes that cause a sudden increase in elevation at the landing point, pits that cause a sudden decrease in elevation at the landing point, and steps that go up or down. When the robot encounters similar terrain changes while moving forward, without active shock absorption, it may cause the robot to sway during movement, and if focusing on a local area centered on the first end effector, it may also cause local instability, such as sudden changes or repeated changes in the first end effector in the first dimension. Therefore, based on the collected raw data (i.e., image data of the robot in the direction of travel along the movement path and the current motion state of the first end effector), the image data is processed to predict the overall change trend of the robot in the first dimension based on the image data. Here, predicting the overall change trend of the robot in the first dimension means judging in advance the overall change trend that may occur in the first dimension (i.e., the direction of gravity or the vertical direction) if the robot continues to move along the movement path. For example, by using image data to discover that the robot will encounter an uphill slope or step on a pothole at some point in the future (which can be combined with the robot's current moving speed), we can anticipate the changes that will occur in the first dimension of the robot, such as causing the robot's center of gravity to rise or fall suddenly. This provides a basis for active shock absorption operations, allowing for better countermeasures.

[0051] In one possible implementation, the robot's processor is used to input the image data into a trained artificial intelligence model to predict the robot's overall trend of change in the first dimension. Thus, by utilizing artificial intelligence models and algorithms, such as reinforcement learning, a closed loop from information acquisition to active vibration damping control is achieved. Furthermore, a visual sensor can be paired with any first end effector to perform separate information acquisition and ultimately focus on the active vibration damping control of the first end effector itself. This means that robots with various hardware implementations can be adapted, and any part or joint on the robot can be selected as the first end effector to construct an active vibration damping scheme that supports the first end effector in maintaining relative stillness in the first dimension.

[0052] In one possible implementation, the current motion state of the first end effector is determined based on the robot's internal three-dimensional spatial coordinate system, and the active damping operation performed by the active damping unit associated with the first end effector includes the coordination between multiple joints of the robot. Thus, instead of requiring a specially designed active damping structure integrated with system control, a separate information acquisition and coordination control mechanism can be used to adapt to the active damping needs of any robot end effector. Furthermore, by coordinating the joints of the robot, such as through algorithms to control the legs and waist, the robot's upper body stability can be achieved.

[0053] Figure 3 This is a schematic diagram of a robot active shock absorption system provided in an embodiment of this application. Figure 3 As shown, the robot active vibration reduction system includes: a vision sensor 310 deployed on the robot's head, a processor A312 deployed on the robot, a built-in acceleration-displacement sensor 320 deployed on the robot's first end effector, and a controller 314 deployed on the robot. The vision sensor 310 deployed on the robot's head is used to acquire image data of the robot in the forward direction of its movement path. The image data at least indicates the overall change of the robot in a first dimension parallel to the direction of gravity as it moves along the movement path. The processor A312 deployed on the robot is used to predict the overall change trend of the robot in the first dimension based on the image data. The built-in acceleration-displacement sensor 320 deployed on the robot's first end effector is used to determine the current motion state of the first end effector. The processor A312 is used to determine a reference motion state of the first end effector based on the overall change trend of the robot in the first dimension and the current motion state of the first end effector. The reference motion state of the first end effector indicates that the change of the first end effector in the first dimension is less than a preset threshold. A controller 314 deployed on the robot is configured to control an active damping unit associated with the first end-effector to perform active damping operations based on a reference motion state of the first end-effector, thereby ensuring that the motion state of the first end-effector remains the reference motion state of the first end-effector as the robot moves along the motion path.

[0054] In short, Figure 3The robot active vibration reduction system shown utilizes a separate information acquisition method. A vision sensor 310 acquires image data of the robot's forward movement path, and a built-in acceleration-displacement sensor 320 at the robot's first end effector determines the current motion state of the first end effector. Based on the acquired raw data (i.e., image data of the robot's forward movement path and the current motion state of the first end effector), the image data is processed to predict the overall trend of the robot's movement in the first dimension. By superimposing the current motion state of the first end effector with the overall change in the first dimension of the robot, the system can better counteract the effects of various vibrations that may be encountered along the robot's movement path, thus improving the active vibration reduction system. The vibration reduction effect forms a closed loop from information acquisition to active vibration reduction control. Furthermore, the vision sensor 310 can be paired with any first end effector to complete separate information acquisition and finally focus on the active vibration reduction control of the first end effector itself. This means it can be adapted to robots with various hardware implementation structures. Any part or joint on the robot can be selected as the first end effector, thereby constructing an active vibration reduction scheme that supports the first end effector to remain relatively stationary in the first dimension. It can be easily extended to simultaneously implement active vibration reduction on the ends effectors of multiple robots, and different preset thresholds can be set for different robot ends effectors, thus better adapting to the needs of refined control. It does not require a specially designed active vibration reduction structure combined with system control; instead, it utilizes a separate information acquisition and coordinated control mechanism to adapt to the active vibration reduction needs of any robot end effector. Therefore, it not only meets the stability and anti-shaking requirements of increasingly complex robot application environments but also can be easily extended to robots with various hardware implementation structures.

[0055] Figure 4This is a schematic diagram of a computing device 400 provided in an embodiment of this application. The computing device 400 includes one or more processors B410, a communication interface 420, and a memory 430. The processors B410, the communication interface 420, and the memory 430 are interconnected via a bus 440. Optionally, the computing device 400 may further include an input / output interface 450, which is connected to input / output devices for receiving user-set parameters, etc. The computing device 400 can be used to implement some or all of the functions of the device embodiment or system embodiment in the above-described embodiments of this application; the processor B410 can also be used to implement some or all of the operation steps of the method embodiment in the above-described embodiments of this application. For example, the specific implementation of various operations performed by the computing device 400 can be referred to the specific details in the above embodiments, such as the processor B410 being used to execute some or all of the steps or operations in the above-described method embodiments. For example, in the embodiments of this application, the computing device 400 can be used to implement some or all of the functions of one or more components in the above-described device embodiments. In addition, the communication interface 420 can be used for communication functions necessary to implement the functions of these devices and components, and the processor B410 can be used for processing functions necessary to implement the functions of these devices and components.

[0056] It should be understood that, Figure 4 The computing device 400 may include one or more processors B410, and the multiple processors B410 may collaboratively provide processing power in a parallel connection, a serial connection, a serial-parallel connection, or an arbitrary connection manner; or the multiple processors B410 may form a processor sequence or a processor array; or the multiple processors B410 may be divided into a main processor and an auxiliary processor; or the multiple processors B410 may have different architectures, such as adopting a heterogeneous computing architecture. Furthermore, Figure 4 The structural and functional descriptions of the computing device 400 shown are exemplary and non-limiting. In some exemplary embodiments, the computing device 400 may include... Figure 4 The diagram shows more or fewer components, or combinations of some components, or splitting of some components, or different arrangements of components.

[0057] The processor B410 can have various specific implementations. For example, the processor B410 may include one or more combinations of a central processing unit (CPU), a graphics processing unit (GPU), a neural network processing unit (NPU), a tensor processing unit (TPU), or a data processing unit (DPU), etc. This application does not impose specific limitations on these embodiments. The processor B410 can also be a single-core processor or a multi-core processor. The processor B410 can be a combination of a CPU and hardware chips. The aforementioned hardware chips can be application-specific integrated circuits (ASICs), programmable logic devices (PLDs), or combinations thereof. The aforementioned PLDs can be complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), generic array logic (GALs), or any combination thereof. The processor B410 can also be implemented solely using logic devices with built-in processing logic, such as FPGAs or digital signal processors (DSPs). The communication interface 420 can be a wired interface or a wireless interface, used to communicate with other modules or devices. The wired interface can be an Ethernet interface, a local interconnect network (LIN), etc., and the wireless interface can be a cellular network interface or a wireless LAN interface, etc.

[0058] Memory 430 may be non-volatile memory, such as 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. Memory 430 may also be volatile memory, which may be random access memory (RAM) used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM). The memory 430 can also be used to store program code and data, so that the processor B410 can call the program code stored in the memory 430 to execute some or all of the operation steps in the above method embodiments, or to execute the corresponding functions in the above device embodiments. Furthermore, the computing device 400 may include, compared to... Figure 4 The number of components displayed may be more or less, or there may be different component configurations.

[0059] Bus 440 can be a Peripheral Component Interconnect Express (PCIe) bus, or an Extended Industry Standard Architecture (EISA) bus, a Unified Bus (Ubus or UB), a Compute Express Link (CXL) bus, a Cache Coherent Interconnect for Accelerators (CCIX) bus, etc. Bus 440 can be divided into address bus, data bus, control bus, etc. In addition to the data bus, bus 440 can also include a power bus, control bus, and status signal bus. However, for clarity,Figure 1 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.

[0060] The methods and devices provided in this application are based on the same inventive concept. Since the principles by which the methods and devices solve problems are similar, the embodiments, implementation methods, examples, or methods of implementation of the methods and devices can be referred to each other, and repeated details will not be repeated. This application also provides a system comprising multiple computing devices, the structure of each computing device of which can refer to the structure of the computing devices described above. The functions or operations achievable by this system can refer to the specific implementation steps in the above method embodiments and / or the specific functions described in the above device embodiments, and will not be repeated here.

[0061] This application also provides a computer-readable storage medium storing computer instructions. When these computer instructions are executed on a computer device (such as one or more processors), they can implement the method steps described in the above method embodiments. The specific implementation of the above method steps by the processor of the computer-readable storage medium can refer to the specific operations described in the above method embodiments and / or the specific functions described in the above device embodiments, and will not be repeated here.

[0062] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. This application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Embodiments of this application can be implemented wholly or partially by software, hardware, firmware, or any other combination. When implemented in software, the above embodiments can be implemented wholly or partially as a computer program product. This application can take the form of a computer program product embodied on one or more computer-usable storage media containing computer-usable program code. The computer program product includes one or more computer instructions. When the computer program instructions are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application 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, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. Computer-readable storage media can be any available medium that a computer can access, or a data storage device such as a server or data center that contains one or more sets of available media. Available media can be magnetic media (such as floppy disks, hard disks, and magnetic tapes), optical media, or semiconductor media. Semiconductor media can be solid-state drives, random access memory, flash memory, read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, or any other suitable form of storage medium.

[0063] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. Each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes ​ The steps of the function specified in one or more boxes.

[0064] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not described in detail in a certain embodiment can be referred to in the relevant descriptions of other embodiments. Obviously, those skilled in the art can make various modifications and variations to the embodiments of this application without departing from the spirit and scope of the embodiments of this application. The steps in the methods of the embodiments of this application can be adjusted in order, combined, or deleted according to actual needs; the modules in the systems of the embodiments of this application can be divided, combined, or deleted according to actual needs. If these modifications and variations of the embodiments of this application fall within the scope of the claims of this application and their equivalents, then this application also intends to include these modifications and variations.

Claims

1. A robot active damping method, characterized by, The robot active damping method comprises: acquiring, by a vision sensor of the robot, image data of the robot in a forward direction of a motion route, the image data at least indicating an overall change of the robot in a first dimension parallel to a gravity direction when the robot moves along the motion route; predicting, by a processor of the robot, a change trend of the robot in the first dimension based on the image data; determining, by an acceleration displacement sensor built in a first end of the robot, a current motion state of the first end; determining, by the processor, a reference motion state of the first end based on the change trend of the robot in the first dimension and the current motion state of the first end, wherein the reference motion state of the first end indicates that a change of the first end in the first dimension is less than a preset threshold value; controlling, by a controller of the robot, an active damping unit associated with the first end to perform an active damping operation based on the reference motion state of the first end, so that a motion state of the first end during the robot moving along the motion route is maintained as the reference motion state of the first end, wherein the robot active damping method further comprises: determining, by an acceleration displacement sensor built in a second end of the robot, a current motion state of the second end; determining, by the processor, a reference motion state of the second end based on the change trend of the robot in the first dimension and the current motion state of the second end, wherein the reference motion state of the second end indicates that a change of the second end in the first dimension is less than a second preset threshold value, which is different from the preset threshold value; controlling, by the controller, an active damping unit associated with the second end to perform an active damping operation based on the reference motion state of the second end, so that a motion state of the second end during the robot moving along the motion route is maintained as the reference motion state of the second end, wherein the first end is a left hand of the robot, the second end is a right hand of the robot, and the active damping operation performed by the active damping unit associated with the first end comprises mutual coordination between a plurality of joints of the robot.

2. The robot active damping method of claim 1, wherein, The current motion state of the first end comprises a current position of the first end, a current speed of the first end, and a current acceleration of the first end.

3. The robot active damping method of claim 2, wherein, Determining, by the processor, the reference motion state of the first end based on the change trend of the robot in the first dimension and the current motion state of the first end comprises: determining, by the processor, an overall motion state of the robot relative to the first end based on the change trend of the robot in the first dimension, and then superimposing the overall motion state of the robot relative to the first end to the current motion state of the first end to obtain an expected motion state of the first end. determine a reference motion state of the first end based on a component of the expected motion state of the first end in the first dimension.

4. The robot active damping method of claim 3, wherein, an active damping operation performed by an active damping unit associated with the first end is configured to counteract a change of the first end in the first dimension caused by the component of the expected motion state of the first end in the first dimension.

5. The robot active damping method of claim 1, wherein, the robot is a half-body humanoid robot, the robot comprises a head, an upper body, a waist and a mobile base, the waist is fixed on the mobile base, the first end is the upper body or a joint comprised by the upper body, the visual sensor is disposed on the head, and the active damping unit associated with the first end is disposed on the waist and the mobile base.

6. The robot active damping method of claim 1, wherein, the robot is a full-body humanoid robot, the robot comprises a head, an upper body, a waist and legs, the first end is the upper body or a joint comprised by the upper body, the visual sensor is disposed on the head, and the active damping unit associated with the first end is disposed on the waist and the legs.

7. The robot active damping method of claim 1, wherein, the image data further indicates terrain change information in a forward direction of the motion route.

8. The robot active damping method of claim 1, wherein, the processor of the robot is configured to input the image data into a trained artificial intelligence model, thereby predicting an overall change trend of the robot in the first dimension.

9. The robot active damping method of claim 1, wherein, the current motion state of the first end is determined based on an internal three-dimensional coordinate system of the robot.

10. An electronic device, comprising: the electronic device comprises a memory, a processor and a computer program stored on the memory and executable on the processor, and the processor implements the method according to any one of claims 1 to 9 when executing the computer program.

11. A computer readable storage medium, characterized in that, the computer readable storage medium stores computer instructions, and when the computer instructions are executed on a computer device, the computer device executes the method according to any one of claims 1 to 9.

12. A robot active damping system, characterized by the robot active damping system comprises: a visual sensor disposed on a head of a robot, configured to collect image data of the robot in a forward direction of a motion route, the image data at least indicating an overall change of the robot in a first dimension parallel to a direction of gravity when the robot advances along the motion route; a processor disposed on the robot, configured to predict an overall change trend of the robot in the first dimension based on the image data; an internal acceleration displacement sensor disposed on a first end of the robot, configured to determine a current motion state of the first end, wherein the processor is configured to determine a reference motion state of the first end based on the overall change trend of the robot in the first dimension and the current motion state of the first end, wherein the reference motion state of the first end indicates that a change of the first end in the first dimension is less than a preset threshold value; a controller disposed on the robot, configured to control the active damping unit associated with the first end to perform an active damping operation based on the reference motion state of the first end, so that the motion state of the first end during the robot advancing along the motion route is maintained as the reference motion state of the first end, wherein the robot active damping system further comprises an in-built acceleration displacement sensor disposed on the second end of the robot, configured to determine a current motion state of the second end, the processor is further configured to determine a reference motion state of the second end based on the overall change trend of the robot in the first dimension and the current motion state of the second end, wherein the reference motion state of the second end indicates that the change of the second end in the first dimension is less than a second preset threshold, and the second preset threshold is different from the preset threshold; the controller is further configured to control the active damping unit associated with the second end to perform an active damping operation based on the reference motion state of the second end, so that the motion state of the second end during the robot advancing along the motion route is maintained as the reference motion state of the second end, wherein the first end is a left hand of the robot, and the second end is a right hand of the robot, and the active damping operation performed by the active damping unit associated with the first end comprises mutual coordination between a plurality of joints of the robot.

Images