A motion primitive-based humanoid motion planning method, device and medium

Abstract

The application discloses a kind of action primitive-based anthropomorphic motion planning method, device and medium, comprising: the task to be executed is divided into subtask, and the execution order of the subtask is planned according to task demand;Capture the motion data of human arm when each subtask is executed, and the action primitive under each task is extracted and fitted quantization, to construct the action primitive library of each subtask;The action primitive library of each subtask is used as carrier to transfer the motion skill of human arm to anthropomorphic arm, to realize the anthropomorphic motion planning of redundant mechanical arm.The application can greatly reduce the data volume while improving the adaptability of robot to task.

Description

Technical Field

[0001] This invention belongs to the field of redundant robotic arm motion planning technology, and particularly relates to an anthropomorphic motion planning method, device and medium based on motion primitives. Background Technology

[0002] With the improvement of economic and technological levels, the interaction scenarios between robots and humans are increasing. Humanization (HRI) refers to endowing non-human individuals with human characteristics, enabling robots to move in anthropomorphic motion, thus providing humans with a more realistic and comfortable interactive experience. The anthropomorphic arm, as a key component of interactive robots performing complex tasks, has the closest interaction with humans. To ensure the efficiency and safety of the interaction process, improving its intelligence level and anthropomorphic motion capabilities is a crucial issue that urgently needs to be addressed. Human-like motion planning is the core technology for generating anthropomorphic motion in robots.

[0003] Currently, anthropomorphic motion planning methods based on performance metrics and human arm motion reproduction suffer from numerous drawbacks, including difficulty in solving problems, a wide variety of metrics, and poor general applicability to different tasks. Therefore, to overcome these shortcomings of traditional anthropomorphic motion planning methods, reduce data volume, and improve the robot's adaptability to tasks, designing a computationally simple, redundant anthropomorphic motion planning method for robotic arms that can effectively adapt to tasks of varying difficulty is urgently needed and of great significance. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a humanoid motion planning method, device and medium based on motion primitives, which addresses the shortcomings of the prior art.

[0005] To achieve the above technical objectives, the technical solution adopted by this invention is: a humanoid motion planning method based on action primitives, comprising the following steps:

[0006] (1) Divide the task to be performed into subtasks and plan the execution order of the subtasks according to the task requirements;

[0007] (2) Capture the motion data of the human arm when performing each sub-task, and construct the motion primitive library for each sub-task by extracting and fitting the motion primitives under each task;

[0008] (3) Use the motion primitive library of each sub-task as a carrier to transfer the motion skills of the human arm to the anthropomorphic arm.

[0009] Furthermore, the capture of arm motion data during the execution of each sub-task specifically involves using motion capture equipment to obtain continuous motion data of the arm during the execution of each sub-task.

[0010] The motion capture device uses sensors to detect human movement.

[0011] Furthermore, the extraction of the action primitives specifically involves:

[0012] (2.1) Calculate the position of each frame of motion using the motion data of the human arm. ,attitude and elbow rotation angle ;

[0013] (2.2) For the position of motion in each frame ,attitude and elbow rotation angle Smoothing is performed using a filtering algorithm;

[0014] (2.3) Valid motion segments are selected by judging the value of the human hand speed; if the speed is non-zero, it is a valid motion.

[0015] (2.4) Identify the motion primitives of each frame of valid motion, wherein the motion primitives include the rate of change of position, the rate of change of posture, and the rate of change of elbow rotation angle;

[0016] (2.5) Merge adjacent frames with the same type of motion primitives to obtain all motion primitives for the entire effective motion segment;

[0017] The term "consistent motion primitives" refers to motion primitives that contain identical elements; these elements include the rate of change of position, the rate of change of posture, and the rate of change of elbow rotation angle.

[0018] Furthermore, the elbow rotation angle The angle between the reference plane and the arm plane, where the arm plane is composed of the upper arm and the forearm, and the reference plane is the plane formed by the two axes when the axis of joint 2 of the anthropomorphic serial robotic arm is parallel to the axis of joint 4;

[0019] The elbow rotation angle The calculation method is as follows:

[0020]

[0021] in, This represents the vector formed from the shoulder to the elbow; This represents the vector formed from the shoulder to the wrist; This represents the vector formed from the elbow to the wrist.

[0022] Furthermore, for the task to be performed, BVH data, i.e., motion data, is obtained through sensors on the body parts requiring data collection; the BVH data includes rotational data of the human skeleton and limb joints; the position and posture of the right / left hand in the shoulder coordinate system, i.e., the homogeneous transformation matrix, are calculated through the logical chain of the skeletal tree in the BVH data.

[0023]

[0024] in, and These represent the attitude and position of each physical node in the shoulder coordinate system, i.e., the base coordinate system; where This represents the number of physical nodes between the hand coordinate system and the shoulder coordinate system. It passes through four physical nodes: hand, forearm, upper arm, and shoulder. The value is 4; This represents the current physical node.

[0025] Furthermore, the optimized Fourier series function is used to fit and quantize the action primitives in the effective motion segments and establish an action primitive library corresponding to the sub-task.

[0026] The optimized Fourier series function is as follows:

[0027]

[0028]

[0029]

[0030]

[0031] in, , and All are coefficients of the Fourier series; This indicates the degree of the Fourier series expansion; Indicates rotational speed; The fitted function The cycle.

[0032] Furthermore, step (3) includes the following sub-steps:

[0033] (3.1) Transform each element of the human arm motion primitive into state quantities describing the human arm motion, namely, the end position of the human arm, the end posture of the human arm, and the elbow rotation angle.

[0034] (3.2) Solve the spatial position of each physical node of the human arm or redundant robotic arm relative to the shoulder coordinate system based on the state variables of the motion. The physical nodes include the end of the arm, the wrist node and the elbow node.

[0035] (3.3) Based on the spatial position of the physical nodes, the motion joint angles of the redundant robotic arm are analyzed as follows:

[0036]

[0037] in , , These are the rate of change of position, the rate of change of posture, and the rate of change of elbow rotation angle, respectively. Representing joints, in numerical order The following are represented in sequence: base, shoulder 1, shoulder 2, elbow 1, elbow 2, elbow 3, and wrist.

[0038] Furthermore, the process of transforming the elements of the human arm motion primitive into state variables describing the human arm motion specifically involves:

[0039]

[0040] in, , , These are the position of the human arm's end, the posture of the human arm's end, and the elbow rotation angle, respectively. The initial position of the human arm's end. The initial attitude angle of the human arm end. This is the initial elbow rotation angle.

[0041] The present invention also provides an anthropomorphic motion planning device based on action primitives, including one or more processors for implementing the above-described anthropomorphic motion planning method based on action primitives.

[0042] The present invention also provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, is used to implement the above-described anthropomorphic motion planning method based on action primitives.

[0043] This invention offers the following advantages: It provides a computationally simple method for anthropomorphic motion planning of redundant robotic arms that can adapt well to tasks of varying difficulty. It employs an optimized Fourier series function to fit and quantize basic motion primitives and establishes a motion primitive library corresponding to sub-tasks, reducing data volume while maintaining fitting accuracy. Taking single-arm reach-point motion as an example, the anthropomorphic motion planning method based on motion primitives analyzes the reach-point motion task, forms a motion primitive library for single-arm reach-point motion, and maps it to the joint angles of the anthropomorphic arm, thereby completing the anthropomorphic motion of the redundant robotic arm. This invention can significantly reduce data volume while improving the robot's adaptability to tasks. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a schematic diagram of the basic human arm motion state quantities, i.e., the basic human arm motion elements, as described in this invention.

[0046] Figure 2 This is a schematic diagram of the elbow rotation angle ESA described in this invention;

[0047] Figure 3 This is a schematic diagram of the anthropomorphic motion planning method based on action primitives described in this invention;

[0048] Figure 4 This is a schematic diagram related to the mapping of human arm motion primitives to joint layers as described in this invention;

[0049] Figure 5 This is a schematic diagram of the single-arm reaching point motion captured by the motion capture device in an embodiment of the present invention;

[0050] Figure 6 The image shows the simulation results of the basic action primitive—position P—fitted using an optimized Fourier function, obtained in an embodiment of the present invention.

[0051] Figure 7 The simulation results for the basic motion primitive—pose O—fitted using an optimized Fourier function in this embodiment of the invention are shown.

[0052] Figure 8 In this embodiment of the invention, the basic motion primitive—elbow rotation angle—is fitted using an optimized Fourier function. Renderings;

[0053] Figure 9 This is a simulation diagram of the single-arm reaching point motion obtained in an embodiment of the present invention;

[0054] Figure 10 This is a hardware structure diagram provided for an embodiment of the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0056] Although the steps in this invention are arranged by reference numerals, this is not intended to limit the order of the steps. Unless the order of the steps is explicitly stated or the execution of a step requires other steps as a basis, the relative order of the steps can be adjusted. It is understood that the term "and / or" as used herein refers to and covers any and all possible combinations of one or more of the associated listed items.

[0057] The present invention will now be described in detail with reference to the accompanying drawings. Unless otherwise specified, the features of the following embodiments and implementations can be combined with each other.

[0058] This invention provides a humanoid motion planning method based on motion primitives, used to realize humanoid motion of redundant robotic arms performing different tasks, comprising the following steps:

[0059] (1) Divide the task to be performed into subtasks and plan the execution order of the subtasks according to the task requirements;

[0060] The main goal of subdividing a task into subtasks is to be able to autonomously break down complex tasks into simpler subtasks and plan the optimal execution order of these subtasks based on task requirements. For example, taking the action of a robotic arm playing a piano as an example, the complex playing motion task to be completed by the robotic arm will be broken down into several simple robotic arm reaching point motion subtasks; further, the execution order of the subtasks will be planned based on the time information contained in the sheet music.

[0061] (2) Capture the arm motion data when performing each sub-task, and construct the motion primitive library for each sub-task by extracting and fitting the motion primitives under each task;

[0062] This invention can use a motion capture system to capture arm motion data when performing various sub-tasks, extract motion primitives for each task according to the motion data logic of the skeletal tree, and finally use an optimized Fourier series function to fit and quantize the basic motion primitives to establish a motion primitive library corresponding to the sub-tasks, thereby reducing the amount of data while ensuring fitting accuracy.

[0063] In one embodiment, the motion capture system uses motion capture equipment to obtain continuous motion data of the human arm while performing various sub-tasks; the motion capture equipment uses sensors to sense the movement of the human body, specifically for the task situation, and obtains BVH data, which includes rotational data of the human skeleton and limb joints, through sensors of the human body parts for which data needs to be collected.

[0064] In one embodiment, the extraction of action primitives includes:

[0065] (2.1) Calculate the position of each frame of motion using the motion data of the human arm. ,attitude and elbow rotation angle ;

[0066] (2.2) For the position of motion in each frame ,attitude and elbow rotation angle Smoothing is performed using a filtering algorithm;

[0067] (2.3) Valid motion segments are selected by judging the value of the human hand speed; if the speed is non-zero, it is a valid motion.

[0068] (2.4) Identify the motion primitives of each frame of effective motion, which include three basic motion primitives: position change rate, posture change rate, and elbow rotation angle change rate.

[0069] (2.5) Merge adjacent frames with the same type of motion primitives to obtain all motion primitives for the entire effective motion segment;

[0070] The term "consistent motion primitives" refers to motion primitives that contain identical elements; these elements include the rate of change of position, the rate of change of posture, and the rate of change of elbow rotation angle.

[0071] If time If the corresponding rate of change of position, rate of change of attitude, and rate of change of elbow rotation angle are not zero, then... The action primitives at any given moment contain corresponding elements.

[0072] For the position of motion in each frame of the valid motion segment ,attitude and elbow rotation angle An optimized Fourier function is used to fit the data, ensuring that the fitting function passes through the key points of the entire effective motion segment during the fitting process. The optimized Fourier function is used to quantify the three basic motion primitives.

[0073] The mathematical expression for the optimized Fourier function is as follows:

[0074]

[0075]

[0076]

[0077]

[0078] Indicates rotational speed; The fitted function The cycle; , and All are coefficients of the Fourier series; This indicates the order of the Fourier series expansion; the higher the expansion order, the better the fitting effect.

[0079] like Figure 1 As shown, position P, orientation O, and elbow rotation angle These are motion state quantities, and the three can completely express the state of a human arm / anthropomorphic arm at a certain moment; however, to reflect whether the human arm has moved over a period of time, the rate of displacement change from time t-1 to time t will be used. Represented as The characterization of the rate of change of attitude angle Represented as The rate of change of the elbow rotation angle will be used to characterize the elbow rotation angle. Represented as Thus, three basic motion primitives were obtained. , , .

[0080] like Figure 3 As shown, the anthropomorphic motion planning method based on action primitives in this embodiment of the invention is divided into three layers: a task layer, a human arm action primitive layer, and a joint layer. The task layer segments the motion task, dividing the entire task into sub-tasks. Then, the human arm action primitive layer performs its work in each sub-task, namely, extracting, sorting, and quantizing the action primitives. Finally, in the joint layer, the action primitives quantized into optimized Fourier functions are mapped and transformed into various joint angles.

[0081] like Figure 4 As shown, the mapping approach from the human arm motion primitive layer to the joint layer described in this embodiment of the invention mainly includes three levels: the quantized human arm motion primitive is converted into motion state quantities and then into physical node positions, and finally into anthropomorphic arm joint angles.

[0082] like Figure 2 As shown, the elbow rotation angle (ESA) can be used to completely express the arm pose of a human arm. In the field of robotics, it is sometimes called the arm angle, and it can be used to decouple the redundancy of the human arm. Specifically, it means that while fixing the position of the wrist, the elbow can still rotate freely around the virtual axis from the shoulder to the wrist. Therefore, at the same target point, the human arm has countless poses, and each pose at that target point can be uniquely determined using the elbow rotation angle (ESA). The elbow rotation angle is defined as the angle between the reference plane and the arm plane. As shown in the figure, the reference plane consists of the virtual axis from the shoulder to the wrist and a straight line perpendicularly downwards from the center of the shoulder joint. The arm plane consists of the upper arm and the forearm. The elbow rotation angle is used to define the elbow rotation angle. The reference plane is defined as the plane formed by the two axes when the axis of joint 2 of the anthropomorphic serial manipulator with SRS configuration is parallel to the axis of joint 4.

[0083] In one embodiment, the position of motion in each frame is calculated using BVH data. ,attitude and elbow rotation angle Method (taking the right arm as an example):

[0084] The position and pose of the right hand in the shoulder coordinate system, i.e., the homogeneous transformation matrix, are calculated using the logical chain of the skeletal tree in the BVH data. .

[0085] The order of the nodes from the shoulder to the end of the arm is RightUpArm—RightLowArm—RightForceArm—RightHand—EndSite.

[0086]

[0087] in, and These represent the attitude and position of each physical node in the shoulder coordinate system, i.e., the base coordinate system.

[0088] In one embodiment, with attitude Related state variables The solution can be obtained using the general formula of the YXZ rotation method, as follows:

[0089]

[0090] in, Represents the rotation matrix; atan2 represents the arctangent function; , and These represent the Euler angles in the x, y, and z directions of the Cartesian coordinate system.

[0091] In one embodiment, the elbow rotation angle The solution formula is as follows:

[0092]

[0093] in This represents the vector formed from the shoulder to the elbow; This represents the vector formed from the shoulder to the wrist; This represents the vector formed from the elbow to the wrist;

[0094] In one embodiment, the type filtering of action primitives is performed by determining the action primitives at a certain time t (frame). , and Is the value 0? If it is not 0, then The primitive of time contains this element.

[0095] (3) The motion primitive library of each sub-task is used as a carrier to transfer the motion skills of the human arm to the anthropomorphic arm, so as to realize the anthropomorphic motion planning of the redundant robotic arm.

[0096] like Figure 2 As shown, the joint layer uses the motion primitives of each subtask constructed by the human arm motion primitive module as a carrier to transfer the motion skills of the human arm to the anthropomorphic arm. The entire mapping process is to convert the basic elements of the motion primitives into the motion angles of each joint of the robot, thereby realizing the anthropomorphic motion planning of the redundant robotic arm.

[0097] In one embodiment, mapping motion primitives to joint angles involves converting the three basic elements in the motion primitives into the motion angles of each joint of the robot, as expressed by the following formula:

[0098]

[0099] In one embodiment, step (3) includes the following sub-steps:

[0100] (3.1) The mapping approach first transforms each element of the human arm motion primitive into state quantities describing the human arm motion, namely the end position of the human arm, the end posture of the human arm, and the elbow rotation angle.

[0101] Specifically, the idea of ​​converting the human arm motion primitives into motion state quantities in step (3.1) can be expressed by the following formula:

[0102]

[0103] in, The initial position of the human arm's end. The initial attitude angle of the human arm end. The initial elbow rotation angle is given, and these initial values ​​are all represented in the shoulder coordinate system.

[0104] (3.2) Solve the spatial position of each physical node of the human arm or redundant arm relative to the shoulder coordinate system based on the state variables of the human arm motion, including the end of the arm, the wrist node and the elbow node.

[0105] (3.3) Finally, based on the spatial position of each physical node, the motion joint angles of the redundant robotic arm are analyzed:

[0106]

[0107] in , , These are the rate of change of position, the rate of change of posture, and the rate of change of elbow rotation angle, respectively. Representing joints, in numerical order The following are represented in sequence: base, shoulder 1, shoulder 2, elbow 1, elbow 2, elbow 3, and wrist.

[0108] In one embodiment, the motion of a single arm reaching a point (taking the right arm as an example) is taken as the object of study, and the simulation results obtained are as follows: Figure 5-9 As shown;

[0109] like Figure 5 As shown, the experimenter's right arm made a point movement, that is, moving from one point to another in space from left to right, using the Neuten motion capture system.

[0110] like Figure 6-8 As shown, Figure 6-8 (b) in the middle are respectively Figure 6-8 The enlarged view in (a) shows the local area. In the simulation, the optimized Fourier function is used to fit the basic motion primitive. It can be seen that the motion process is relatively complex for the basic motion primitive. The fitted function ensures the fitting accuracy at specific key points while reducing the amount of data.

[0111] like Figure 9 As shown, the results of building an anthropomorphic arm using the Matlab Robotics Toolbox and reproducing anthropomorphic motion using the method based on the present invention (referring to the trajectory of the anthropomorphic arm in space) show that the method reproduces the single-arm reaching-point motion well, realizing the reaching-point motion from one point to another from left to right.

[0112] The present invention also provides an anthropomorphic motion planning device based on action primitives, comprising:

[0113] The task planning module is used to divide the task to be executed into subtasks and plan the execution order of the subtasks according to the task requirements.

[0114] The motion primitive library construction module is used to capture the motion data of the human arm when performing various sub-tasks, and to construct the motion primitive library for each sub-task by extracting, fitting and quantizing the motion primitives under each task.

[0115] The anthropomorphic motion module is used to transfer the motion techniques of the human arm to the anthropomorphic arm by using the motion primitive library of each subtask as a carrier.

[0116] It should be noted that the device embodiment shown in this embodiment matches the content of the above method embodiment, and the content of the above method embodiment can be referred to, and will not be repeated here.

[0117] Corresponding to the aforementioned embodiment of an anthropomorphic motion planning method based on action primitives, the present invention also provides an embodiment of an anthropomorphic motion planning device based on action primitives.

[0118] See Figure 10The present invention provides an anthropomorphic motion planning device based on action primitives, comprising one or more processors for implementing an anthropomorphic motion planning method based on action primitives as described in the above embodiments.

[0119] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0120] An embodiment of the anthropomorphic motion planning device based on action primitives of the present invention can be applied to any device with data processing capabilities, such as a computer. The device embodiment can be implemented in software, hardware, or a combination of both. Taking software implementation as an example, as a logical device, it is formed by the processor of any data processing device loading the corresponding computer program instructions from non-volatile memory into memory for execution. From a hardware perspective, such as... Figure 10 The diagram shown is a hardware structure diagram of any device with data processing capabilities, including the anthropomorphic motion planning device based on action primitives according to the present invention. (Except for...) Figure 10 In addition to the processor, memory, network interface, and non-volatile memory shown, any data processing device in the embodiment may also include other hardware depending on the actual function of the data processing device, which will not be described in detail here.

[0121] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0122] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the present invention according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0123] This invention also provides a computer-readable storage medium storing a program thereon, which, when executed by a processor, implements a humanoid motion planning method based on action primitives as described in the above embodiments.

[0124] The computer-readable storage medium can be an internal storage unit of any data processing device described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium can also be any data processing device, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc., equipped on the device. Furthermore, the computer-readable storage medium can include both internal storage units of any data processing device and external storage devices. The computer-readable storage medium is used to store the computer program and other programs and data required by the data processing device, and can also be used to temporarily store data that has been output or will be output.

[0125] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0126] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A humanoid motion planning method based on action primitives, characterized in that, Includes the following steps: (1) Divide the task to be performed into subtasks and plan the execution order of the subtasks according to the task requirements; (2) Capture the motion data of the human arm when performing each sub-task, and construct the motion primitive library for each sub-task by extracting and fitting the motion primitives under each task; The extraction of the action primitives specifically involves: (2.1) Calculate the position of each frame of motion using the motion data of the human arm. ,attitude and elbow rotation angle ; (2.2) For the position of motion in each frame ,attitude and elbow rotation angle Smoothing is performed using a filtering algorithm; (2.3) Valid motion segments are selected by judging the value of the human hand speed; if the speed is non-zero, it is a valid motion. (2.4) Identify the motion primitives of each frame of valid motion, wherein the motion primitives include the rate of change of position, the rate of change of posture, and the rate of change of elbow rotation angle; (2.5) Merge adjacent frames with the same type of motion primitives to obtain all motion primitives for the entire effective motion segment; The term "consistent motion primitives" refers to motion primitives that contain identical elements; these elements include the rate of change of position, the rate of change of posture, and the rate of change of elbow rotation angle. (3) Use the motion primitive library of each sub-task as a carrier to transfer the motion skills of the human arm to the anthropomorphic arm.

2. The method according to claim 1, characterized in that, The specific method for capturing arm motion data during the execution of each sub-task is to use motion capture equipment to obtain continuous motion data of the arm during the execution of each sub-task. The motion capture device uses sensors to detect human movement.

3. The method according to claim 1, characterized in that, The elbow rotation angle The angle between the reference plane and the arm plane is defined by the upper arm and the forearm. The reference plane is the plane formed by the two axes when the axis of joint 2 and the axis of joint 4 of the anthropomorphic serial robotic arm are parallel. The joints are numbered sequentially as follows: base, shoulder 1, shoulder 2, elbow 1, elbow 2, elbow 3, and wrist. The elbow rotation angle The calculation method is as follows: in, This represents the vector formed from the shoulder to the elbow; This represents the vector formed from the shoulder to the wrist; This represents the vector formed from the elbow to the wrist.

4. The method according to claim 1, characterized in that, For the task to be performed, BVH data, i.e., motion data, is obtained through sensors on the body parts requiring data collection. The BVH data includes rotational data of the human skeleton and limb joints. The position and orientation of the right / left hand in the shoulder coordinate system, i.e., the homogeneous transformation matrix, are calculated using the logical chain of the skeletal tree in the BVH data. in, and These represent the attitude and position of each physical node in the shoulder coordinate system, i.e., the base coordinate system; where This represents the number of physical nodes between the hand coordinate system and the shoulder coordinate system. It passes through four physical nodes: hand, forearm, upper arm, and shoulder. The value is 4; This represents the current physical node.

5. The method according to claim 1, characterized in that, An optimized Fourier series function is used to fit and quantize the motion primitives in the effective motion segments, and a motion primitive library corresponding to the sub-task is established. The optimized Fourier series function is as follows: in, , and All are coefficients of the Fourier series; This indicates the order of the Fourier series expansion; Indicates rotational speed; The fitted function The cycle.

6. The method according to claim 1, characterized in that, Step (3) includes the following sub-steps: (3.1) Transform each element of the human arm motion primitive into state quantities describing the human arm motion, namely, the end position of the human arm, the end posture of the human arm, and the elbow rotation angle. (3.2) Solve the spatial position of each physical node of the human arm or redundant robotic arm relative to the shoulder coordinate system based on the state variables of the motion. The physical nodes include the end of the arm, the wrist node and the elbow node. (3.3) Based on the spatial position of the physical nodes, the motion joint angles of the redundant robotic arm are analyzed as follows: in , , These are the rate of change of position, the rate of change of posture, and the rate of change of elbow rotation angle, respectively. Representing joints, in numerical order The following are represented in sequence: base, shoulder 1, shoulder 2, elbow 1, elbow 2, elbow 3, and wrist.

7. The method according to claim 6, characterized in that, The process of transforming the elements of the human arm motion primitive into state variables describing the human arm motion specifically involves: in, , , These are the position of the human arm's end, the posture of the human arm's end, and the elbow rotation angle, respectively. The initial position of the human arm's end. The initial attitude angle of the human arm end. This is the initial elbow rotation angle.

8. A humanoid motion planning device based on action primitives, characterized in that, It includes one or more processors for implementing the anthropomorphic motion planning method based on action primitives as described in any one of claims 1-7.

9. A computer-readable storage medium having a program stored thereon, characterized in that, When executed by the processor, the program is used to implement the anthropomorphic motion planning method based on action primitives as described in any one of claims 1-7.

Images