Mechanical arm-exoskeleton collaborative exercise training robot system and method

Abstract

The invention provides a mechanical arm-exoskeleton collaborative exercise training robot system and method, and belongs to the technical field of intelligent exercise training equipment. The exoskeleton can give real-time feedback and correction to a user in the movement process, the dependence on vision is reduced, and the movement training efficiency is improved; the six-axis mechanical arm is used for tracking the motion state of the shoulder joint in real time, the pulling feeling is avoided, the gravity of the exoskeleton part is borne, and the influence on human motion is reduced; according to the exoskeleton forearm structure, the motion requirement of the arm is fully considered, and compared with a disclosed two-stage parallel four-bar mechanism, the designed exoskeleton forearm structure can meet the motion within a larger angle range and is simpler and more portable compared with a traditional arc-shaped guide rail; a highly-integrated man-machine interaction system is adopted, motion data of a human body can be captured in real time, and more effective guidance is provided for a wearer through an exoskeleton system.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent sports training equipment, and in particular to a sports training robot system and method for teaching and correcting sports postures and evaluating sports training effects using a manipulator-exoskeleton collaboration. Background Art

[0002] Traditional sports training primarily relies on observing and imitating the instructor's movements in the early stages of learning. The instructor observes the standardization of the movements and determines whether the exercise is acceptable. However, this method relies on the instructor's experience and is often too subjective. Furthermore, visual observation of exercise practice is often inaccurate and time-consuming.

[0003] Exoskeletons, as wearable robotic devices, can be attached to the human body's moving parts, driving muscle and joint movements. Their proximity to internal proprioceptors also allows for more efficient training. Exoskeletons are currently being used in a variety of fields, including sports rehabilitation and exercise assistance. In sports rehabilitation, the robot often acts as the master, driving the human joints to perform corresponding rehabilitation movements. In exercise assistance, the human joints are the master, with the exoskeleton providing assistance.

[0004] Related research abroad began early, and numerous products featuring advanced design concepts and technological innovations have been developed. For example, the ARMin series of robots developed by ETH Zurich, with its high-precision sensor integration and adaptive compensation concepts, holds a significant position in the field of rehabilitation engineering. The RUPERT series of upper limb rehabilitation exoskeleton robots, developed by Arizona State University, utilizes pneumatic muscles as a power source to simulate natural muscle properties, providing effective equipment for rehabilitation training. Furthermore, the BONES device developed by the University of California, Irvine, and the CADEN-7 exoskeleton robot developed by the University of Washington both employ innovative structural designs and control strategies to accommodate diverse rehabilitation training needs.

[0005] Although domestic research started relatively late, it has developed rapidly and achieved remarkable results. The upper limb rehabilitation robot UECM, developed by Tsinghua University, utilizes a two-link mechanism and offers multimodal rehabilitation training capabilities. The upper limb exoskeleton developed by Harbin Institute of Technology incorporates gravity balancing and cable transmission methods to enhance the exoskeleton's performance and flexibility. The CURE-7 exoskeleton robot enables seven-degree-of-freedom upper limb training, and its wearable mechanism has been experimentally verified for its comfort and adaptability.

[0006] For upper limb exoskeletons used in sports rehabilitation, once worn by the patient, the exoskeleton system can assist the patient in training according to a set motion trajectory. However, when using this type of exoskeleton, the user's body often needs to remain in a fixed position, with only the upper limbs free to move. Furthermore, these exoskeletons are designed primarily for rehabilitation training movements, and their training strategies and methods differ from those of sports training. Therefore, they cannot be directly and simply transferred to sports training.

[0007] Meanwhile, regarding the prior art, Chinese invention patent application number 201711102641.8 provides a 4-DOF forearm for an upper-limb exoskeleton robot. This system utilizes a two-stage four-bar linkage to achieve internal and external rotation (DOF) for the forearm. However, this linkage configuration cannot achieve large rotation angles and cannot fully meet the movement requirements of the forearm. Chinese invention patent application number 202310989916.3 describes an adaptive, reconfigurable virtual exoskeleton rehabilitation robot system. While this system can utilize multiple robotic arms to coordinate and drive the human upper limb for rehabilitation exercises, this approach is not only costly and complex to control, but also involves interference between the robotic arms, limiting the user's range of motion. Chinese invention patent application number 202111222461.X provides a cable-driven upper-limb exoskeleton rehabilitation robot with passive shoulder joint tracking. While this system can passively track the shoulder joint's translation and align with human joint movement, its frame-like structure restricts human posture during use and cannot meet the needs of physical exercise. Chinese invention patent application number 202310969702.X provides a lightweight exoskeleton robot system for posture correction during exercise training. While the exoskeleton is carried on the body, it does increase the burden on the wearer. Furthermore, the shoulder joint's limited two degrees of freedom restricts joint movement, making it uncomfortable to wear. Summary of the Invention

[0008] The purpose of the present invention is to provide a manipulator-exoskeleton collaborative sports training robot system and method, which includes a control system, a control method, a sports training robot device, and a training method, so as to solve the technical problems in the above-mentioned background technology in the application of exoskeleton robots in the field of sports training.

[0009] In order to achieve the above object, the present invention adopts the following technical solutions:

[0010] In a first aspect, the present invention provides a mechanical system of a sports training robot with a robotic arm and an exoskeleton collaboration, comprising a console, a robotic arm, and an exoskeleton, wherein the robotic arm is fixed to the console; the robotic arm and the exoskeleton are connected to the control system; and the end of the robotic arm is connected to the exoskeleton.

[0011] As a further limitation of the first aspect of the present invention, the exoskeleton includes an upper arm structure connected to the end of the robotic arm via a connecting flange, the six-dimensional force sensor being connected to the connecting flange; an elbow joint structure connected to the upper arm structure via an elbow joint motor, a forearm structure connected to the elbow joint structure via a forearm joint motor, and a wrist joint structure connected to the forearm structure via a wrist joint motor.

[0012] As a further limitation of the first aspect of the present invention, the boom structure includes a boom binding base fixed to the other end of the six-dimensional force sensor; the two sides of the boom binding base are rotatably connected to the boom binding side plates, and the other end of the boom binding base is installed with an adjustable boom guide rod, and a boom cam buckle is installed above the adjustable boom guide rod, and the adjustable boom guide rod is fixed to the boom binding base by rotating the cam buckle.

[0013] As a further limitation of the first aspect of the present invention, the elbow joint structure includes an elbow motor base, one end of which is connected to the upper arm adjustable guide rod, and the other end is connected to the elbow joint motor; one end of the elbow binding base is connected to the elbow joint motor, and the two sides are rotatably connected to the elbow binding side plates; the other end of the elbow binding base is installed with the elbow adjustable guide rod, and an elbow cam buckle is installed above the elbow adjustable guide rod, and the elbow adjustable guide rod is fixed to the elbow binding base by the elbow rotating cam buckle.

[0014] As a further limitation of the first aspect of the present invention, the forearm structure includes a forearm motor base, one end of which is fixedly connected to the elbow adjustable guide rod, and the other end is connected to the forearm joint motor; one end of the first connecting rod is connected to the forearm motor base, and the other end is connected to the fourth connecting rod; the second connecting rod is fixed on the forearm joint motor, and its two ends are respectively connected to the third connecting rod and the fourth connecting rod; the two ends of the fifth connecting rod are respectively connected to the ends of the third connecting rod and the fourth connecting rod; the forearm binding base is connected to the fifth connecting rod through an adjustment buckle, and the two sides of the forearm binding base are respectively rotatably connected with forearm binding side plates.

[0015] As a further limitation of the first aspect of the present invention, the wrist joint structure includes one end of the wrist motor base connected to the forearm binding base, and the other end fixedly connected to the wrist joint motor; the two ends of the wrist rotation base are respectively connected to the wrist joint motor and the forearm binding base; and the wrist guide rod is fixed to the wrist rotation base by a guide rod buckle.

[0016] In a second aspect, the present invention provides a control system for a manipulator-exoskeleton collaborative motion training robot, comprising:

[0017] An acquisition device, used to obtain the current posture data of the robotic arm and exoskeleton, the force applied to the end of the robotic arm and the actual torque of the joint motor, as well as human motion image data;

[0018] a processing device, connected to the acquisition device, for importing the posture data, the force on the end of the manipulator, and the actual torque of the joint motor into a dynamic model, calculating the expected force on the end of the manipulator and the expected torque of each joint motor of the exoskeleton; and comparing the expected force and expected torque with the actual force and actual torque, respectively, to obtain the user's limb movement trend, which is used as the input value for admittance control to issue a control signal;

[0019] The control device is connected to the processing device and is used to receive the control signal and control the movement of the robotic arm and the exoskeleton according to the control signal to achieve flexible control of the robotic arm and the exoskeleton.

[0020] As a further limitation of the second aspect of the present invention, the acquisition device includes a depth camera, an angle encoder, a force sensor, and a torque sensor; the depth camera is used to acquire human motion image data; the angle encoder is used to acquire posture data of the robotic arm and the exoskeleton; the force sensor is used to acquire the force applied to the end of the robotic arm; the torque sensor is used to acquire the actual torque of each joint motor; the control device includes a robotic arm controller and a motor controller; the robotic arm controller is used to control the movement of the robotic arm; and the motor controller is used to control the torque, speed, and angle of the motor.

[0021] In a third aspect, the present invention provides a control method for a manipulator-exoskeleton collaborative motion training robot, comprising:

[0022] Obtain the current posture data of the robotic arm and exoskeleton, human motion image data, as well as the force at the end of the robotic arm and the actual torque of the joint motor;

[0023] The posture data, the force on the end of the manipulator, and the actual torque of the joint motor are imported into the dynamic model to solve the expected force of the end of the manipulator and the motor of each joint of the exoskeleton; the expected force is compared with the actual force to obtain the user's limb movement trend, which is used as the input value of the admittance control to issue a control signal;

[0024] According to the control signal, compliant control of the robotic arm and the exoskeleton is achieved.

[0025] In a fourth aspect, the present invention provides a motion training method for a manipulator-exoskeleton coordinated motion training robot system, comprising: in different stages of motion learning and in different training modes, the motion control of the motion training robot adopts different control modes; wherein,

[0026] Training data acquisition mode: Using zero-torque control, the robot does not provide a preset motion trajectory. Instead, the acquisition device collects dynamic data from the user during movement. Six-dimensional force sensors and torque sensors are used to monitor the interaction force between the user and the robotic arm in real time. The motion control variable is calculated using an admittance control algorithm to track and simulate the position and posture of the human upper limb in space in real time.

[0027] Basic Motion Training Mode: The position of the robotic arm end position does not have a preset slow motion teaching trajectory. The robotic arm and exoskeleton positions follow the preset slow motion teaching trajectory. In this mode, the position of the robotic arm end position follows the movement of the user's shoulder joint in real time. At the same time, the posture of the robotic arm end position and the exoskeleton motor adopt non-compliant control, strictly following the preset motion teaching trajectory to assist the user in completing the movement.

[0028] Simulated Motion Training Mode: The end-arm position does not have a preset motion teaching trajectory. Instead, the end-arm posture and exoskeleton use admittance control to adjust the actual robot motion trajectory in real time based on the preset normal speed motion trajectory and the interaction force between the user and the arm, gently guiding the user to perform more precise and smooth movements. In this mode, the end-arm position follows the movement of the user's shoulder joint in real time. By adjusting the impedance control parameters, the end-arm and exoskeleton provide appropriate resistance and / or assistance when the user performs the movement, helping the user to experience and learn the correct movement force and rhythm.

[0029] Enhanced training mode: The robotic arm dynamically adjusts the auxiliary force according to the user's real-time movement status, strengthening the correction force at positions with large deviations from the teaching trajectory and weakening the correction force at positions with small trajectory deviations. This allows the user to gradually reduce their dependence on the robotic arm during training and improve their muscle memory and control ability.

[0030] The beneficial effects of the present invention include: providing professional movement instruction for users, while providing real-time feedback and corrections during exercise training, thereby improving exercise training efficiency; utilizing a six-axis robotic arm to track shoulder joint motion in real time, thereby avoiding a sense of strain, and balancing the weight of the exoskeleton to reduce the burden of wearing; fully considering the movement requirements of the arm, the exoskeleton is designed to ensure elbow flexion and extension, forearm internal rotation / external rotation, and wrist adduction / abduction and internal rotation / external rotation. Furthermore, the exoskeleton's arm structure, in achieving forearm internal and external rotation, can accommodate a wider range of motion angles than previously known two-stage parallelogram linkages; and is simpler and lighter than traditional curved guide rails. In terms of exercise training methods, a highly integrated human-computer interaction system is employed, which not only captures human motion data in real time but also provides direct guidance to the wearer through the exoskeleton system; and through real-time motion data collection and analysis, a personalized training plan can be provided for each user. This personalized approach helps improve training effectiveness, particularly in correcting incorrect movements and improving motor skills.

[0031] Additional advantages of the present invention will be more clearly given in the following description or learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 This is a three-dimensional structural diagram of the sports training robot device according to an embodiment of the present invention.

[0034] Figure 2 This is a structural diagram of the exoskeleton portion of the sports training robot device according to an embodiment of the present invention.

[0035] Figure 3 This is a structural diagram of the upper arm according to an embodiment of the present invention.

[0036] Figure 4 This is a structural diagram of the elbow joint according to an embodiment of the present invention.

[0037] Figure 5 This is a structural diagram of the forearm according to an embodiment of the present invention.

[0038] Figure 6 This is a structural diagram of the wrist joint according to an embodiment of the present invention.

[0039] Figure 7This is a functional principle block diagram of the control system of the sports training robot device according to an embodiment of the present invention.

[0040] Figure 8 This is a flow chart of the motion training method of the motion training robot device according to an embodiment of the present invention.

[0041] Among them: 101-robotic arm, 102-exoskeleton, 103-depth camera, 104-robot console, 201-wrist joint structure, 202-forearm structure, 203-elbow joint structure, 204-upper arm structure, 301-connecting flange, 302-six-dimensional force sensor, 303-upper arm binding base, 304-upper arm binding side plate, 305-upper arm cam buckle, 306-upper arm adjustable guide rod, 401-elbow motor base, 402-elbow joint motor, 403-elbow cam buckle, 404 -Elbow adjustable guide rod, 405-Elbow binding side plate, 406-Elbow binding base, 501-Forearm motor base, 502-Forearm joint motor, 503-First connecting rod, 504-Second connecting rod, 505-Third connecting rod, 506-Fourth connecting rod, 507-Fifth connecting rod, 508-Adjustment buckle, 509-Forearm binding side plate, 510-Forearm binding base, 601-Wrist guide rod, 602-Guide rod buckle, 603-Wrist joint motor, 604-Wrist motor base, 605-Wrist rotation base. DETAILED DESCRIPTION

[0042] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and are not to be construed as limiting the present invention.

[0043] Those skilled in the art will understand that unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs.

[0044] It should also be understood that terms, such as those defined in commonly used dictionaries, should be understood to have a meaning consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless as defined herein.

[0045] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a," "an," "said," and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the stated features, integers, steps, operations, elements, and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, and / or groups thereof.

[0046] In the description of this specification, reference to the terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Those skilled in the art may combine and integrate different embodiments or examples described in this specification, as well as features of different embodiments or examples, unless otherwise contradictory.

[0047] To facilitate understanding of the present invention, the present invention is further explained below with reference to specific embodiments in conjunction with the accompanying drawings. However, the specific embodiments do not constitute a limitation on the embodiments of the present invention.

[0048] Those skilled in the art should understand that the drawings are merely schematic diagrams of embodiments, and the components in the drawings are not necessarily necessary for implementing the present invention.

[0049] In response to the shortcomings of the aforementioned related technologies and the specific use cases for sports training, the system designed in this invention can replace sports coaches in providing sports training and corrections to users. Furthermore, based on the needs and characteristics of sports training, a robotic structure combining a mechanical arm and an exoskeleton has been designed. The robot can bear its own weight without adding any additional load to the human body, thus meeting the space requirements for shooting training. A control system has also been designed for the exoskeleton, along with various sports training modes. This system enables precise collection of human motion data through wearable technology, real-time motion assistance and guidance, and assessment of training effectiveness, covering the entire training process from start to finish.

[0050] Example 1

[0051] like Figures 1 to 6 As described above, in this embodiment, a robotic arm-exoskeleton collaborative motion training robot mechanical system is provided.

[0052] The robotic arm-exoskeleton collaborative motion training robot system described in this embodiment 1 is a robot for training upper limb motion postures. At the hardware structure level, the robot consists of four parts: a robotic arm 101, an exoskeleton 102, a depth camera 103, and a robotic console 104. The robotic arm 101 is a six-degree-of-freedom robotic arm, and the exoskeleton 102 includes three active degrees of freedom and one passive degree of freedom. The weight of the robotic exoskeleton 102 itself is transmitted to the ground via the robotic arm 101 and the robotic console 104 to prevent the weight from being applied to the human body and affecting normal movement. The end of the robotic arm 101 follows the movement of the human shoulder joint and drives the exoskeleton 102 installed at the end to move. In the robotic exoskeleton 102, a special connecting rod mechanism is designed for the internal and external rotation degrees of freedom of the human forearm. The mechanism can be split into two parallelogram mechanisms. When tied together with the human arm, the arm acts as a new rotation axis, which can add virtual constraints to the quadrilateral mechanism, thereby eliminating the singular position of the mechanism. The robot's wrist joint structure 201, in addition to being designed for flexion and extension of the wrist, is also equipped with a wrist guide rod 601, which fits the human palm through a strap. This method has a simple structure and can achieve relative fixation between the wrist joint mechanism and the palm while also satisfying the adduction and outward swing of the wrist joint.

[0053] Specifically, such as Figure 1 As shown, the robot for sports training with robot-exoskeleton cooperation includes: a six-degree-of-freedom robot 101 and a depth camera 103 respectively mounted on a robot console 104, and a robot exoskeleton 102 mounted at the end of the robot arm 101. Figure 2 As shown, the robot exoskeleton 102 includes a boom structure 204, an elbow joint structure 203, a forearm structure 202, and a wrist joint structure 201, which are connected in sequence. The boom structure 204 is connected to the end of the robotic arm 101 via a connecting flange 301, to which a six-dimensional force sensor 302 is connected. The elbow joint structure 203 is connected to the boom structure 204 via an elbow joint motor mount 401, the forearm structure 202 is connected to the elbow joint structure 203 via a forearm joint motor mount 501, and the wrist joint structure 201 is connected to the forearm structure 202 via a wrist joint motor mount 604.

[0054] The boom structure includes a boom binding base fixed to the other end of the six-dimensional force sensor; the two sides of the boom binding base are rotatably connected to the boom binding side plates, the other end of the boom binding base is installed with a boom adjustable guide rod, and a boom cam buckle is installed above the boom adjustable guide rod, and the boom adjustable guide rod is fixed to the boom binding base by rotating the cam buckle. Specifically, Figure 3As shown, one end of the boom structure 204 is provided with a connecting flange 301 for connection to the robotic arm; a six-dimensional force sensor 302 is bolted to the connecting flange 301, and a boom binding base 303 is also bolted to the other end of the six-dimensional force sensor 302. Two boom binding side plates 304 are pivotally connected on either side of the boom binding base 303, which can be tightened to the trainee's boom by tightening the bolts. An adjustable boom guide rod 306 is installed in a circular hole at the other end of the boom binding base 303. A boom cam buckle 305 is installed above the adjustable boom guide rod 306, and the adjustable boom guide rod 306 can be fixed to the boom binding base 303 by rotating the cam buckle 305.

[0055] In a specific embodiment, the elbow joint structure includes an elbow motor base, one end of which is connected to the upper arm adjustable guide rod, and the other end is connected to the elbow joint motor; one end of the elbow binding base is connected to the elbow joint motor, and the two sides are rotatably connected to the elbow binding side plates; the other end of the elbow binding base is installed with the elbow adjustable guide rod, and the elbow cam buckle is installed above the elbow adjustable guide rod, and the elbow adjustable guide rod is fixed to the elbow binding base by rotating the elbow cam buckle. Specifically, Figure 4 As shown, the robot's elbow joint structure 203 has an elbow motor base 401 with one end connected to the upper arm adjustable guide rod 306 and the other end connected to the elbow joint motor 402. One end of the elbow binding base 406 is connected to the elbow joint motor 402, and both sides of the elbow binding base 406 are also pivotally connected to the elbow binding side plates 405. The elbow binding base 406 has a circular hole mounted on the other end. An elbow cam buckle 403 is mounted above the elbow binding base 404. The elbow binding base 406 can be rotated to secure the elbow binding base 406 to the elbow binding base 406. The elbow motor 402 drives the cam buckle 403, the elbow adjustable guide rod 404, the elbow binding side plates 405, and the elbow binding base 406 to move.

[0056] In a specific embodiment, the forearm structure includes a forearm motor base, one end of which is fixedly connected to the elbow adjustable guide rod, and the other end is connected to the forearm joint motor; one end of the first connecting rod is connected to the forearm motor base, and the other end is connected to the fourth connecting rod; the second connecting rod is fixed to the forearm joint motor, and its two ends are respectively connected to the third connecting rod and the fourth connecting rod; the two ends of the fifth connecting rod are respectively connected to the ends of the third connecting rod and the fourth connecting rod; the forearm binding base is connected to the fifth connecting rod through an adjustment buckle, and the two sides of the forearm binding base are rotatably connected to the forearm binding side plates. Specifically, as Figure 5As shown, the robot arm structure 202 has an arm motor base 501 with one end fixed to the elbow adjustable guide rod 404 and the other end connected to the arm joint motor 502. One end of the first connecting rod 503 is connected to the arm motor base 501, and the other end is connected to the fourth connecting rod 506. The second connecting rod 504 is fixed to the arm joint motor 502, and its two ends are respectively connected to the third connecting rod 505 and the fourth connecting rod 506. The arm motor base 501, the arm joint motor 502, the first connecting rod 503, the second connecting rod 504 and the fourth connecting rod 506 are interconnected to form a parallelogram. The two ends of the fifth connecting rod 507 are respectively connected to the third connecting rod 505 and the end of the fourth connecting rod 506. The second connecting rod 504, the third connecting rod 505, the fourth connecting rod 506 and the fifth connecting rod 507 form a parallelogram. The forearm binding base 510 is connected to the fifth link 508 via an adjustment buckle 508. Its two sides are pivotally connected to the forearm binding side plates 509. The forearm joint motor 502 drives the two four-bar linkages, which in turn rotates the forearm binding base 510, thereby rotating the forearm bound to it.

[0057] In a specific embodiment, the wrist joint structure includes a wrist motor base with one end connected to the forearm binding base and the other end fixedly connected to the wrist joint motor; the two ends of the wrist rotation base are respectively connected to the wrist joint motor and the forearm binding base; the wrist guide rod is fixed to the wrist rotation base through a guide rod buckle. Figure 6 As shown, the robot wrist joint structure 201 has a wrist motor base 604 with one end connected to the forearm binding base 510 and the other end fixed to the wrist joint motor 603. The two ends of the wrist rotation base 605 are respectively connected to the wrist joint motor 603 and the forearm binding base 510. At the same time, the wrist guide rod 601 is fixed to the wrist rotation base 605 through the guide rod buckle 602. During use, the wrist rotation base 605, the guide rod buckle 602, and the wrist guide rod 601 are driven to rotate by the wrist joint motor 603, thereby driving the flexion and extension of the wrist. At the same time, because the wrist guide rod 601 has a passive degree of freedom, the wrist can freely perform inward and outward swing movements.

[0058] In this embodiment, as a further optimization of the robotic arm-exoskeleton collaborative motion training robot, arm length adjustment is also possible. For the upper arm structure 204, the fixed state of the upper arm adjustable guide rod 306 can be adjusted by rotating the cam buckle 403. Changing the installation position of the upper arm adjustable guide rod 306 can change the distance from the shoulder joint to the elbow joint of the exoskeleton. Similarly, for the elbow joint structure 203, the fixed assembly platform of the elbow adjustable guide rod 404 can be adjusted by rotating the cam buckle 403, thereby changing the distance from the elbow joint to the wrist joint of the exoskeleton.

[0059] On the other hand, in the exoskeleton forearm structure 202, the third link 505, fourth link 506, and fifth link 507 are designed to be curved to reduce friction with the human forearm. Furthermore, to accommodate different arm sizes, the tightness of the adjustment buckle 508 can be adjusted during use, thereby changing the fixed position of the forearm binding base 510.

[0060] Example 2

[0061] In this embodiment 2, a control system for a robot for sports training with a manipulator-exoskeleton collaboration is first provided. Figure 7 As shown, the core functions of the control system in this embodiment include: 1. Robot state detection: acquiring real-time measurement results from the sensor detection system through a data acquisition card; 2. Interaction data and target parameter calculation: calculating the human-machine interaction force and robot motion control parameters based on the collected data; 3. Robot motion control: the control program connects to the drive control system through the motion controller to achieve robot motion. The drive control system includes a motor driver and a manipulator driver, and the sensor detection system includes a depth camera, a force sensor, and an angle encoder.

[0062] Furthermore, a control system for a robot for sports training with a coordinated manipulator and exoskeleton is divided into three parts, including: an acquisition device for acquiring the current posture data of the manipulator and exoskeleton, human motion image data, and the force at the end of the manipulator and the actual torque of the joint motor; a processing device connected to the acquisition device, for importing the posture image data, the force at the end of the manipulator and the actual torque of the joint motor into a dynamic model, and solving the expected force of the current manipulator end and each joint motor of the exoskeleton; and comparing the expected force with the actual force to obtain the user's limb movement trend, and sending a control signal as the admittance control input value; a control device connected to the processing device, for receiving the control signal and realizing flexible control of the manipulator and exoskeleton according to the control signal.

[0063] The acquisition device includes a depth camera, an angle encoder, a force sensor and a torque sensor; the depth camera is used to collect human motion image data; the angle encoder reads the angle q of each joint of the robotic arm in real time. r ∈R 6 and the exoskeleton's motor angle θ m ∈R 3 The force sensor is used to collect the force F at the end of the robotic arm ext =[f x ,f y ,f z ,τ x ,τ y ,τ z ] TThe force sensor is a six-dimensional force sensor. The torque sensor is used to collect the actual torque τ of each joint motor of the exoskeleton. m ∈R 3 ; The control device includes a robotic arm controller and a motor controller; the robotic arm controller is used to control the movement of the robotic arm; the motor controller can be used to control the movement of the motor.

[0064] In this embodiment, the above control system can be used to control a sports training robot. The control method includes: first, processing the data acquired by the acquisition device to extract the human-machine interaction force. Based on the SolidWorks model, the mass m of each link of the robot is derived. i , center of mass position r i and moment of inertia I i , construct the quality matrix M(q)∈R n×n (n = 9 degrees of freedom), and establish the Denavit-Hartenberg (DH) parameter model. According to the real-time robot arm joint angle q r and motor angle θ m , calculate the gravity term using the Newton-Euler recursive algorithm in is the linear velocity Jacobian matrix of link i, g = [0, 0, -9.81] T m / s 2 is the gravity acceleration vector. Combined with the joint angular velocity and angular acceleration Calculating Coriolis force terms and inertia force The exoskeleton dynamics are mapped to the end of the manipulator and the rotational joints of the exoskeleton through the Jacobian matrix J e (q) Realize the force conversion from joint space to end Cartesian space: The actual collected F ext and τ m Subtract F dir Get the human-machine interaction force F at the end of the robotic arm human , the human-machine interaction torque τ at the exoskeleton joint. This is used as the admittance control input value, combined with a specific training method to derive the control signals for the manipulator and exoskeleton motors. Finally, based on these control signals, compliant control of the manipulator and exoskeleton is achieved.

[0065] For example, in this embodiment, the control system is used in a robot-exoskeleton collaborative motion training device described in Example 1. The robot-exoskeleton motion training device includes a robot arm and an exoskeleton. The robot arm is a six-degree-of-freedom robot arm that can achieve full-range free motion. The exoskeleton is connected to the end of the robot arm. The trainee's upper limbs are tied to the exoskeleton. A six-dimensional force sensor is provided between the exoskeleton and the robot arm. A torque sensor is provided on each joint motor to respectively collect the force at the end of the robot arm and the torque of each joint motor. At the same time, an angle encoder is provided at each joint to collect the joint angles of the robot arm and the exoskeleton. In addition, an image acquisition device such as a depth camera is provided to collect state image data of the robot arm, the exoskeleton and the trainee. The shooting posture training robot device of this embodiment includes a robot arm controller, a motor controller, a depth camera, an angle encoder, a six-dimensional force sensor, and a torque sensor at the hardware level. In this embodiment, a modular design is adopted in the series structure with 9 active degrees of freedom formed by the connection of the robot arm and the exoskeleton, and the motion paths of the six-degree-of-freedom robot arm and the exoskeleton are planned separately. For the admittance control of the robot arm, the calculated interaction force F at the end of the robot arm is calculated. human Satisfies the formula, Among them, K d Stiffness matrix, B d Damping matrix, M d Inertia matrix, ΔX = X d -X is the position deviation, is the admittance control speed, The following steps are taken to update the end position of the robotic arm at the next moment and solve the inverse kinematics to obtain the robotic arm joint angles at the next moment. The exoskeleton motor control method is similar. The human-machine interaction torque τ and the angle deviation Δθ at the exoskeleton joint are substituted into the formula to calculate the desired joint angles. This ultimately achieves compliant control of the robotic arm and exoskeleton.

[0066] Example 3

[0067] In this embodiment, a motion training method based on the motion training robot device described in Embodiment 1 and Embodiment 2 is provided. Specifically, Figure 8 As shown, this embodiment sets different modes according to different stages of the training process, including training data collection mode, basic sports training mode, simulated sports training mode, and intensive training mode.

[0068] (1) In the training data acquisition mode, the robot is in zero torque control state to accurately record the user's movements. The robotic arm and exoskeleton do not impose any preset motion trajectory, but instead monitor the interaction force between the user and the robot in real time through six-dimensional force sensors and torque sensors. By modifying the admittance control algorithm, the stiffness matrix K is shielded. d ,Right now Calculate the necessary motion control amount to achieve tracking of human upper limb joint motion.

[0069] (2) In the basic motion training mode, the position of the end of the robot arm does not have a preset slow motion teaching trajectory, and the posture of the end of the robot arm and the position of the exoskeleton follow the preset slow motion teaching trajectory. In this mode, the end position of the robot arm follows the movement of the user's shoulder joint in real time. At the same time, the posture of the end of the robot arm and the exoskeleton motor adopt non-compliant control, strictly following the preset motion teaching trajectory to assist the user to complete the action.

[0070] (3) In the simulated motion training mode, the robot uses the impedance control mode under trajectory tracking. The robot guides the user to perform more accurate and smooth shooting movements according to the preset standard motion trajectory. By adjusting the impedance control parameters, the robot can provide appropriate resistance feedback when the user performs the action, thereby helping the user experience and learn the correct movement posture and rhythm.

[0071] (4) Enhanced Training Mode: For users who are already proficient in performing sports movements, the enhanced training mode provides more advanced training. In this mode, the robot dynamically adjusts the auxiliary force, strengthening the correction force at positions with large deviations from the teaching trajectory and weakening the correction force at positions with small trajectory deviations. In addition, the system can analyze the user's movement data through intelligent algorithms and provide personalized feedback and suggestions to help users further improve their sports skills.

[0072] Example 4

[0073] In this embodiment, an auxiliary training strategy and effect evaluation method for a manipulator-exoskeleton collaborative sports training robot are provided.

[0074] The overall plan is as follows: First, a machine learning algorithm analyzes standard motion data to obtain a basis for evaluating exercise training effectiveness. Next, combined with sensor feedback data, an evaluation method based on human motion similarity is designed. By determining the similarity between the subject's training motions and those of experts, the training effectiveness is evaluated, and the exoskeleton control parameters are adjusted in real time to optimize the training strategy. Finally, through long-term follow-up experiments, training data is collected and analyzed to evaluate the effectiveness of exoskeleton training and identify areas for improvement, ensuring the scientific and effective nature of the training.

[0075] Taking basketball training as an example, the motion trajectory, shooting trajectory and hit rate are important indicators for evaluating the training effect. In actual sports training, users need to be guided and corrected based on the overlap of the two trajectories. Therefore, the steps for establishing a sports training model are as follows: ① Use the motion data acquisition mode of the present invention to collect the upper limb joint motion data of professionals when shooting, and use a filtering algorithm to reduce noise on the data to obtain the joint motion trajectory. ② Use a depth camera to shoot the basketball motion trajectory when shooting, and use a deep learning detection algorithm, such as YOLO, to perform real-time segmentation processing on the point cloud and extract the key points of the basketball. Then use the RANSAC spherical fitting method to obtain the center position of the ball, and use spline fitting to obtain the basketball trajectory. ③ Use the joint motion trajectory and the basketball trajectory as model training samples to obtain a standard motion data model.

[0076] To capture the trajectory of the basketball and calculate the hit rate, the robot console is equipped with a depth camera that can capture images and point clouds. At the same time, a deep convolutional neural network object detection algorithm, such as YOLOv8 (You Only Look Once version 8), is used in conjunction with a basketball dataset to train a basketball recognition neural network. The recognition steps are as follows: ① The depth camera captures images and point cloud data during the movement; ② The image and point cloud data are imported into the basketball recognition neural network to identify the basketball's position in the image within the range b = [x min ,y min ,x max ,y max ] T , extract the corresponding depth point cloud according to b; ③ Use the RANSAC algorithm to fit the spherical model and calculate the coordinates of the sphere center p ball =[x c ,y c ,z c ] T . For consecutive frames p ball Perform cubic spline interpolation to generate the basketball trajectory T ball (t), and calculate the hit rate η in combination with the basket position:

[0077] The steps of the sports training effect evaluation method are as follows: ① Using the zero torque mode, let the user shoot freely according to personal experience, and record the user's joint movement data and basketball movement trajectory X user (t). ②Use dynamic time warping (DTW) to transform the user trajectory X user (t) and standard motion data model X expert (t) Compare and obtain the degree of trajectory deviation Where T is the number of time steps, S∈[0,1], the closer it is to 1, the higher the overlap. ③ If the trajectory overlap reaches a certain standard, the training is considered qualified.

[0078] Exoskeleton-assisted training method for sports training: ① Based on the results of training effectiveness evaluation, the exoskeleton's compliance control parameters are set during training. If the deviation is low, indicating good training results, the exoskeleton's impedance control stiffness is reduced during subsequent training to reduce user intervention. If the deviation is high, the exoskeleton's impedance control stiffness is increased during subsequent training to strengthen user intervention. ② During assisted training, real-time motion data is compared with a standard motion data model, and the deviation values of each joint are used as admittance control inputs to adjust the corrective force of each exoskeleton joint in real time.

[0079] In summary, the control system, control method, and robotic device for sports training described in the embodiments of the present invention, as well as the training method, are more direct, standardized, and efficient in training compared to traditional training methods that rely on coaching. The exoskeleton of the present invention can provide users with professional sports movement instruction while providing real-time feedback and correction during exercise, reducing reliance on vision and improving the efficiency of sports training. Compared to traditional upper limb rehabilitation training exoskeletons, the present invention can better meet the human body's motion space during shooting and is more comfortable to wear. The present invention utilizes a six-axis robotic arm to track the movement of the shoulder joint in real time, avoiding a sense of pulling, and the gravity of the exoskeleton is borne by the exoskeleton, reducing the impact on human movement. At the same time, the exoskeleton fully considers the movement requirements of the arm, and the designed forearm connecting rod rotation device can accommodate a wider range of motion than existing dual four-bar linkage technology. The forearm rotation device designed by the present invention has a compact structure and a large rotation range compared to traditional curved guide rails and dual four-bar linkages, which better meets the forearm's motion space requirements. This invention offers a more efficient sports training method than traditional methods. It utilizes a highly integrated human-computer interaction system that not only captures human motion data in real time but also provides hands-on guidance to the wearer through the exoskeleton system. Furthermore, through real-time motion data collection and analysis, the invention provides personalized training plans for each user. This personalized approach helps improve training outcomes, particularly in correcting incorrect movements and enhancing athletic skills.

[0080] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0081] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts 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, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0082] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0083] These computer program instructions can also be loaded onto a computer or other programmable data processing device, and a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide the functions for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0084] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solutions disclosed in the present invention without the need for creative work should be included in the scope of protection of the present invention.

Claims

1. A robot arm-exoskeleton collaborative motion training robot device, characterized in that: include: Control system, including acquisition device, processing device and control device; The acquisition device includes a depth camera, an angle encoder, a six-dimensional force sensor and a torque sensor; the depth camera is used to collect human motion image data; the angle encoder is used to collect the position data of the robotic arm and the exoskeleton; the six-dimensional force sensor is used to collect the force on the end of the robotic arm; The torque sensor is used to collect the actual torque of each joint motor; the processing device is used to import the posture data, the force on the end of the manipulator and the actual torque of the joint motor into the dynamic model, solve the expected force / torque and generate a control signal; the control device is used to achieve compliant control according to the control signal; The mechanical system includes a console, a robotic arm and an exoskeleton; the robotic arm is fixed to the end of the console, and the exoskeleton is connected to the end of the robotic arm via a connecting flange; the exoskeleton includes an upper arm structure, an elbow joint structure, a lower arm structure and a wrist joint structure, and each joint is driven by a motor and connected to the control system.

2. The robot device for motion training with a robot arm and exoskeleton collaboration according to claim 1, characterized in that: The exoskeleton includes an upper arm structure connected to the end of the robotic arm via a connecting flange, the six-dimensional force sensor is connected to the connecting flange; an elbow joint structure connected to the upper arm structure via an elbow joint motor, a forearm structure connected to the elbow joint structure via a forearm joint motor, and a wrist joint structure connected to the forearm structure via a wrist joint motor.

3. The robot device for motion training with a robot arm and exoskeleton collaboration according to claim 2, characterized in that: The said big arm structure comprises a big arm binding base fixed to the other end of the six-dimensional force sensor; The two sides of the boom binding base are rotatably connected to the boom binding side plates, and the other end of the boom binding base is installed with an boom adjustable guide rod, and a boom cam buckle is installed above the boom adjustable guide rod. The boom adjustable guide rod is fixed to the boom binding base by rotating the cam buckle.

4. The robot arm-exoskeleton collaborative motion training device according to claim 3, characterized in that: The elbow joint structure includes an elbow motor base, one end of which is connected to the upper arm adjustable guide rod, and the other end is connected to the elbow joint motor; one end of the elbow binding base is connected to the elbow joint motor, and the two sides are rotatably connected to the elbow binding side plates; the other end of the elbow binding base is installed with the elbow adjustable guide rod, and an elbow cam buckle is installed above the elbow adjustable guide rod, and the elbow adjustable guide rod is fixed to the elbow binding base by rotating the elbow cam buckle.

5. The robot arm-exoskeleton coordinated motion training device according to claim 4, characterized in that: The forearm structure includes a forearm motor base, one end of which is fixedly connected to the elbow adjustable guide rod, and the other end is connected to the forearm joint motor; one end of the first connecting rod is connected to the forearm motor base, and the other end is connected to the fourth connecting rod; the second connecting rod is fixed to the forearm joint motor, and its two ends are respectively connected to the third connecting rod and the fourth connecting rod; the two ends of the fifth connecting rod are respectively connected to the ends of the third connecting rod and the fourth connecting rod; the forearm binding base is connected to the fifth connecting rod through an adjustment buckle, and the two sides of the forearm binding base are respectively rotatably connected with forearm binding side plates.

6. The robot device for motion training of a manipulator-exoskeleton collaboration according to claim 5, characterized in that: The wrist joint structure includes one end of the wrist motor base connected to the forearm binding base, and the other end fixedly connected to the wrist joint motor; the two ends of the wrist rotation base are respectively connected to the wrist joint motor and the forearm binding base; the wrist guide rod is fixed to the wrist rotation base through a guide rod buckle.

7. The control system of the robot device for motion training with a manipulator-exoskeleton collaboration according to claim 1, characterized in that: include: The acquisition device is used to obtain the current posture data of the robotic arm and exoskeleton, human motion image data, as well as the force at the end of the robotic arm and the actual torque of the joint motor; a processing device connected to the acquisition device, configured to import the posture data, the force on the end of the manipulator, and the actual torque of the joint motor into a dynamic model, and calculate the current expected force on the end of the manipulator and the expected torque of each joint motor of the exoskeleton; The expected force and expected torque are compared with the actual force and actual torque respectively to obtain the user's limb movement trend, which is used as the admittance control input value to issue a control signal; The control device is connected to the processing device and is used to receive the control signal and realize flexible control of the robotic arm and the exoskeleton according to the control signal.

8. The control system for a robot arm-exoskeleton coordinated sports training device according to claim 7, characterized in that: The acquisition device includes a depth camera, an angle encoder, a six-dimensional force sensor and a torque sensor; the depth camera is used to acquire human motion image data; the angle encoder is used to acquire posture data of the robotic arm and the exoskeleton; the six-dimensional force sensor is used to acquire the force applied to the end of the robotic arm; the torque sensor is used to acquire the actual torque of each joint motor; the control device includes a robotic arm controller and a motor controller; the robotic arm controller is used to control the movement of the robotic arm; and the motor controller is used to control the torque of the motor.

9. A control method for a robot device for sports training with a robot arm and an exoskeleton, characterized in that: include: Obtain the current posture data of the robotic arm and exoskeleton, human motion image data, the force on the end of the robotic arm, and the actual torque of the joint motor; Importing the posture image data, the force on the end of the manipulator, and the actual torque of the joint motor into the dynamic model to solve the current expected force on the end of the manipulator and the expected torque of each joint motor of the exoskeleton; The expected force and expected torque are compared with the actual force and actual torque respectively to obtain the user's limb movement trend, which is used as the admittance control input value to issue a control signal; According to the control signal, combined with different motion training modes, the robot's admittance control parameters are adjusted, and the motion paths of the robotic arm and exoskeleton are planned separately to achieve flexible control of the robotic arm and exoskeleton.

10. A motion training method based on a motion training robot device for manipulator-exoskeleton collaboration according to any one of claims 1 to 9, characterized in that: include: In different stages of motion learning and different training modes, the robot's motion control adopts different control modes; among them, Training data collection mode: The robot uses zero-torque control to collect user motion data and track the user's upper limb joint movements in real time through sensors; Basic motion training mode: The end position of the robotic arm follows the movement of the user's shoulder joint. The end posture of the robotic arm and the exoskeleton joint angle strictly follow the preset slow trajectory to assist the user. Simulated motion training mode: The robotic arm and exoskeleton use admittance control to guide the movement compliantly according to the preset normal speed trajectory and interaction force; Intensive training mode: The robot dynamically adjusts the auxiliary force, strengthening the correction force at positions with large deviations from the teaching trajectory and weakening the correction force at positions with small trajectory deviations.

Images