A force sensing and control system for coupled control rope-driven exoskeleton

Abstract

The present invention discloses a force sensing and control system for use in a coupled control rope-driven exoskeleton, comprising a force sensing mechanism and a control system module; the force sensing mechanism comprises a force sensing module and an energy storage mechanism module, the force sensing module is used to transmit and feedback human-machine interaction force information, and the energy storage mechanism module is used to store elastic potential energy. The control system module comprises a nonlinear control framework module and a safety protection strategy module. The nonlinear control framework module comprises a nonlinear framework module based on an energy storage mechanism, an adaptive PD controller module based on a force sensing mechanism, and a decoupling motion controller module, and is used to control the movement of the rope-driven exoskeleton and reduce the control impact caused by nonlinearity. The safety protection strategy module is used to detect interaction force information and switch the mode of the exoskeleton to protect human-machine safety.

Description

Technical Field

[0001] The present invention relates to the technical field of force sensing mechanisms and control systems for rope-driven exoskeletons, and in particular to a force sensing and control system applied to a coupled control rope-driven exoskeleton. Background Art

[0002] In the field of medical rehabilitation, the effectiveness of traditional rehabilitation training depends on the therapist's personal experience. Currently popular rehabilitation robots also have drawbacks, such as difficulty in wearing, moving, and transporting, becoming a burden rather than a helper for patients and psychologists. Therefore, the medical rehabilitation field urgently needs a small, portable, efficient, and safe rehabilitation robot. Bowden cable-driven upper limb exoskeletons offer a promising solution, offering significant advantages in terms of light weight and low power consumption. However, they lack a suitable force sensing mechanism. The acquisition of force information is crucial for these exoskeleton robots to perform rehabilitation training. Effective force information collection ensures precise control and the safety of both the user and the machine.

[0003] For the Bowden cable-driven upper limb exoskeleton mentioned above, the following requirements are met when collecting force information: (1) The sensor will not be disturbed by the exoskeleton or user's movements and will not be restricted by the joint structure; (2) It can collect force information of the entire robot; (3) After meeting the above two requirements, the collection process is efficient and accurate.

[0004] Previous research on force information collection in human-computer interaction environments can be divided into the following two directions:

[0005] (1) Regarding the acquisition of force information at the joints, the paper “Pseudo-Sensorless High-Performance Bilateral Teleoperation by Sliding-Mode Control and FPGA” (IEEE / ASME Transactions on Mechatronics, vol. 19, no. 1, pp. 384-393, 2014) uses Hall sensors to obtain force information. The paper “Design and Analysis of Force-Sensor-Less Power-Assist Control” (IEEE Transactions on Industrial Electronics, vol. 61, no. 2, pp. 985-993, 2014) only uses motor encoders to obtain force information. The paper “An Elbow Exoskeleton for Upper Limb Rehabilitation With Series Elastic Actuator and Cable-Driven Differential” (IEEE Transactions on Robotics, vol. 35, no. 6, pp. 1464-1474, 2019) uses elastic actuators to obtain force information.

[0006] (2) Regarding the collection of force information at the fixed connection between the human and the machine, the document “Preliminary Assessment of a Postural Synergy-Based Exoskeleton for Post-Stroke Upper Limb Rehabilitation” (IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 29, pp. 1795-1805, 2021) measures the force information using three six-axis force / torque sensors installed on the upper arm, lower arm, and wrist. The document “Preliminary Assessment of a Postural Synergy-Based Exoskeleton for Post-Stroke Upper Limb Rehabilitation” (IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 29, pp. 1795-1805, 2021) installs six-axis force / torque sensors on the forearm cuff and upper arm to estimate the patient’s movement intention. The paper "A Passivity-Based Nonlinear Admittance Control With Application to Powered Upper-Limb Control Under Unknown Environmental Interactions" (IEEE / ASME Transactions on Mechatronics, vol. 24, no. 4, pp. 1473-1484, 2019) installs a six-axis force / torque sensor at the end to measure the human-machine interaction force;

[0007] (3) There are also methods that use indirect force sensing. For example, the paper "Stability-Guaranteed Force-Sensorless Contact Force / Motion Control of Heavy-Duty Hydraulic Manipulators" (IEEE Transactions on Robotics, vol. 31, no. 4, pp. 918-935, 2015) uses the cylinder pressure data of the manipulator to estimate force information. The paper "Adaptive Impedance Control for an Upper Limb Robotic Exoskeleton Using Biological Signals" (IEEE Transactions on Industrial Electronics, vol. 64, no. 2, pp. 1664-1674, 2017) uses biological signals to obtain force information. The paper "Force from Shape—Estimating the Location and Magnitude of the External Force on Flexible Instruments" (IEEE Transactions on Robotics, vol. 37, no. 5, pp. 1826-1833, 2021) uses a model-based method that allows for the simultaneous acquisition of shape and force information. The paper "Whole-body contact force sensing from motion capture" (2016 IEEE / SICE International Symposium on System Integration (SII), pp. 58-63, 2016) uses only motion capture to measure force information across all joints of the human body. The paper "Vibration-Based Multi-Axis Force Sensing: Design, Characterization, and Modeling" (IEEE Robotics and Automation Letters, vol. 5, no. 2, pp. 3082-3089, 2020) uses vibration to measure force information using acceleration signals from the structure.The document "3-D Force Sensing Strategy of Laryngeal Continuum Surgical Robot Based on Fiber Bragg Gratings" (IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1-10, 2021) constructs a force sensing model through mechanical modeling and temperature decoupling algorithm, etc.

[0008] However, none of the above methods can meet the three requirements mentioned above: (1) If force information is collected at the joints, the joint design will be restricted and may collide with the user's body or interfere with the exoskeleton's range of motion, which does not meet the first requirement we mentioned; (2) Force sensing at the fixed connection between the human and the robot does not allow the collection of force information for the entire robot, which conflicts with the second requirement we mentioned and also leads to inflexible daily tasks; (3) Force sensing methods using indirect information are too slow and inaccurate, which does not meet the third requirement. In addition, the interaction force can be obtained indirectly from external devices, but this may be limited by the working environment.

[0009] The guiding principles for the exoskeleton prototype design were power conservation and weight reduction. As mentioned above, current medical rehabilitation exoskeletons urgently require a force sensing solution that meets these three requirements. Therefore, we developed a force sensing and control system for a coupled control cable-driven exoskeleton. Summary of the Invention

[0010] To address the shortcomings of existing technologies and provide a suitable and attractive force sensing solution for coupled-control rope-driven medical rehabilitation exoskeletons, the present invention provides a force sensing mechanism for use in coupled-control rope-driven exoskeletons. Furthermore, the force sensing mechanism described in this invention is applicable not only to coupled-control rope-driven exoskeletons but also to all coupled-control rope-driven devices. Furthermore, it offers the advantages of practicality, reliability, simple structure, lightweight, easy fabrication, simple assembly and disassembly steps, and low cost.

[0011] In order to achieve the purpose of the present invention, the present invention provides a force sensing mechanism and control system applied to coupled control rope drive, including a force sensing mechanism and a control system module.

[0012] The force sensing mechanism includes a force sensing module and an energy storage mechanism module, the force sensing module is used to transmit and feedback human-computer interaction force information, and the energy storage mechanism module is used to store elastic potential energy;

[0013] The control system module includes a nonlinear control framework module and a safety protection strategy module. The nonlinear control framework module includes a nonlinear framework module based on an energy storage mechanism, an adaptive PD controller module based on a force sensing mechanism, and a decoupling motion controller module, which are used to control the movement of the exoskeleton driven by the drive rope and reduce the control impact caused by nonlinearity.

[0014] The nonlinear frame module based on the energy storage mechanism is used to reduce the unidirectional force characteristics of the rope drive and the influence of nonlinearity;

[0015] The adaptive PD controller module based on the force sensing mechanism is used to compensate for the nonlinear driving efficiency of the elbow joint;

[0016] The decoupling motion controller module is used to solve the coupling control planning of the elbow joint and the wrist joint;

[0017] The safety protection strategy module is used to detect interaction force information and switch the exoskeleton mode to protect human-machine safety.

[0018] Furthermore, the force sensing module is arranged between the human forearm connecting piece and the human upper arm connecting piece. The force sensing module includes two symmetrically arranged force sensor units. Each force sensor unit includes a driving rope, a pulley group, and a force sensor. The driving rope, the pulley group, and the force sensor constitute a fixed and movable pulley combination, which serves as the driving source of the force sensing module and the exoskeleton. One end of the driving rope is connected to the human forearm connecting piece through the pulley group, and the other end is connected to the motor.

[0019] Taking the example of a single drive rope, when the elbow joint rotates at a constant speed and tension is applied, the drive rope must provide tension to balance the total weight on the forearm. Since the tension in different parts of the same drive rope is equal, equation (1) is obtained, indicating that the mechanism conserves 2 / 3 of the force. The tension in the drive rope is 1 / 2 of the force detected by the force sensor.

[0020] f p =(1 / 3)f all =(1 / 2)f fs (1)

[0021] Regarding the right drive rope, since the calculation principle is the same for both sides, the same principle also forms the force sensing mechanism of the pulley system. This allows the Bowden cable to be driven and coupled through the invented force sensing mechanism. By incorporating the pulley system, this force sensing mechanism eliminates the need for joints and interferes with the coupling action of the rope drive mechanism, while also saving two-thirds of the force.

[0022] Furthermore, the energy storage mechanism module includes multiple extension springs, which require less spring deformation to achieve the same rotation compared to a single torsion spring, making it more durable. The geometric relationship is determined by formula (2).

[0023]

[0024] The energy storage mechanism module is not only used for collision buffering, but also provides a sufficiently large reverse torque to constrain the motion control of the elbow joint. In addition, this also enables it to have the ability to perform resistance training.

[0025] Furthermore, the nonlinear control framework module includes a nonlinear framework module based on an energy storage mechanism, an adaptive PD controller module based on a force sensing mechanism, and a decoupling motion controller module.

[0026] Furthermore, the nonlinear frame module based on the energy storage mechanism is designed as formula (3).

[0027]

[0028] Among them J f ,B f ,L f and u 1-d (t) are the moment of inertia, damping, linkage length and input torque of the elbow joint respectively. m f is the linkage mass. g is the acceleration due to gravity, which can be obtained from formula (2). It is the upper limit of the disturbance. e1(t)=θ 1-a -θ 1-d ,θ 1-d is the target angle of the elbow joint. Terminal function To reduce jitter, a saturation function is used, such as formula (4), where δ sats is the boundary layer thickness.

[0029]

[0030] Furthermore, in the adaptive PD controller module based on the force sensing mechanism, the elbow joint is driven with nonlinear efficiency during the training task. Due to different postures, different lengths of Bowden cables need to be released / recovered to rotate the same angle. In order to track the target torque u 1-d (t), which can be obtained by formula (3). The observer is designed to observe the driving efficiency in real time, and the adaptive PD controller module is designed according to the driving efficiency to further solve the problem of control hysteresis.

[0031] The target velocities of the left and right sides of the driving rope required by the elbow joint are shown in formulas (6) and (7), where u l-d =u r-d=0.5·u 1-d ,e u-l =u l-d -u l-a ,e u-r =u r-d -u r-a ,u l-d and u r-d are the target torques for the left and right drive ropes, respectively.

[0032]

[0033]

[0034] Furthermore, the decoupling motion controller module proposes a decoupling control method based on the differential drive method of the Mecanum wheel vehicle, as shown in formulas (8) and (9), where v l is the speed of the left driving rope, v r is the speed of the right drive rope.

[0035]

[0036]

[0037] Furthermore, in order to consider the safety of the exoskeleton during rehabilitation training, a safety protection strategy is proposed, which integrates the tension of the drive rope and the actual posture to evaluate whether an abnormality has occurred. Then, the exoskeleton will switch to free mode to ensure safety. Abnormal situations will be divided into two cases: 1) Spasm is a spontaneous bending of the elbow joint, which may cause muscle strain. Therefore, the Bowden cable may be loose, and the judgment condition is shown in formula (10), where u 1-a =u l-a +u r-a 2) In unstructured environments, collisions caused by obstacles need to be considered, which may put the exoskeleton and the user in danger. The occurrence of obstacles will increase the tension and the target posture cannot be tracked, so the judgment condition is shown in formula (11), where δ u is the torque margin. When (10) or (11) is satisfied, the exoskeleton will immediately switch to free mode. In order to offset the reverse torque and gravity, the target torque is calculated by formula (12).

[0038] u 1-a <u min (10)

[0039] u l-a ≥u max =m f gL f sinθ 1-a +τe (θ 1-a )+δ u (11)

[0040] u 1-d =m f gL f sinθ 1-a +τ e (θ 1-a ) (12)

[0041] Compared with the prior art, the present invention has the following advantages and technical effects:

[0042] (1) This invention utilizes the principle of a pulley system to design a force sensing and control system for a coupled control rope-driven exoskeleton. This system can reliably measure the tension of the drive rope, i.e., the human-machine interaction force. Furthermore, it can save 2 / 3 of the force and does not interfere with the normal operation of the coupling device.

[0043] (2) The present application does not require force sensing at the joints, but rather performs force sensing externally; the invention can collect force information of the entire robot; the measurement accuracy of the invention is the same as that of the direct measurement method. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a schematic diagram of the force sensing mechanism structure used in the coupled control rope-driven exoskeleton (right perspective).

[0045] Figure 2 This is a schematic diagram of the force sensing mechanism structure used in the coupled control rope-driven exoskeleton (left side perspective).

[0046] Figure 3 This is a schematic diagram of the control system used in the coupled control rope-driven exoskeleton.

[0047] Figure 4 It is a mathematical model diagram of the energy storage mechanism module.

[0048] Figure 5 It is a schematic diagram of the driving efficiency characteristics based on the force sensing mechanism.

[0049] Figure 6 It is a schematic diagram of a verification platform for comparing the effects of direct force sensing and the invented force sensing mechanism.

[0050] Figure 7 is the comparison result of tension measurement.

[0051] Figure 8 It is a schematic diagram of coupled motion trajectory tracking of an embodiment equipped with the invented force sensing mechanism. DETAILED DESCRIPTION

[0052] The present invention will be further described below with reference to specific embodiments, but is not limited thereto.

[0053] The present invention provides a force sensing and control system for coupling control rope driven exoskeleton, such as Figure 1 、 2 As shown in 3 , it includes a force sensing mechanism and a control system module. The force sensing mechanism includes a force sensing module and an energy storage mechanism module. The control system module includes a nonlinear control framework module and a safety protection strategy module.

[0054] The force sensing module is used to transmit and feedback human-computer interaction force information;

[0055] The energy storage mechanism module is used to store elastic potential energy;

[0056] The nonlinear control framework module is used to control the movement of the rope-driven exoskeleton and reduce the control impact caused by nonlinearity.

[0057] The safety protection strategy module is used to detect interaction force information and switch the exoskeleton mode to protect human-machine safety.

[0058] In some embodiments of the present invention, the force sensing module is positioned between the forearm connector 5 and the upper arm connector 6. The force sensing module includes two force sensor units, each with identical left and right sides and symmetrical about the central axis. Each force sensor unit comprises a drive rope 1, a pulley assembly 2, and a force sensor 3. The drive rope 1, pulley assembly 2, and force sensor 3 form a fixed and movable pulley assembly that serves as the drive source for the force sensing module and the exoskeleton. One end of the drive rope 1 is connected to the forearm connector 5 via the pulley assembly 2, and the other end is connected to a motor to indirectly drive the exoskeleton.

[0059] Taking drive rope 1 on one side as an example, when the elbow joint rotates at a constant speed and tension is applied, drive rope 1 needs to provide tension to balance the total weight on the forearm. Since the tension in different parts of the same drive rope 1 is equal, formula (1) is obtained. Formula (1) shows that the mechanism conserves 2 / 3 of the force. The tension in drive rope 1 is 1 / 2 of the force obtained by force sensor 3.

[0060]

[0061] Where, f p represents the tension of a single rope segment, f all represents the total force on the system, f fs Represents the perception force (interaction force) of the force sensing module.

[0062] Regarding the right drive rope, since the calculation principle is the same for both sides, the same principle also forms the force sensing mechanism of the pulley system. This allows the Bowden cable to be driven and coupled via the invented force sensing mechanism. By incorporating the pulley system 2, this force sensing mechanism eliminates the need for joints, preventing interference with the coupling action of the rope drive mechanism and saving two-thirds of the force.

[0063] In some embodiments of the present invention, the energy storage mechanism module includes multiple tension springs. Compared with a single torsion spring, multiple tension springs require less spring deformation to achieve the same rotation, making the energy storage mechanism module more durable. The structure of the energy storage mechanism module 7 is as follows: Figure 1 and 4 As shown, it includes multiple tension springs 4, and the multiple tension springs 4 equally divide a circumference, and the tension spring layout is as follows Figure 4 As shown, a tension spring 4 is mounted within the energy storage module 7. One end of the tension spring 4 is connected via a hook on the upper rod, while the other end is connected via a hook on the energy storage module 7. The energy storage module 7 is fixedly connected to the lower rod. During operation, as the elbow joint moves, the tension spring 4 is stretched, storing elastic potential energy, which provides reverse kinetic energy for returning to its original position.

[0064] Its geometric relationship is determined by formula (2).

[0065]

[0066] Where, τ e Represents the reverse torque generated by the energy storage mechanism module, r e Represents the inner diameter of the energy storage mechanism module, R e Represents the outer diameter of the energy storage mechanism module, θ 1-a represents the rotation angle of the elbow joint (the rotation angle of the energy storage mechanism module), k s Represents the spring stiffness of the energy storage mechanism module, L s0 Represents the original length of the spring of the energy storage mechanism module. Among them, the posture sensor is used as the sensor to feedback the current posture, including the actual rotation angle θ of the elbow joint 1-a and the wrist rotation angle θ 2-a .

[0067] The proposed energy storage mechanism module not only serves to cushion collisions but also provides sufficiently large reverse torque to constrain the motion control of the elbow joint, which in addition enables it to perform resistance training.

[0068] In some embodiments of the present invention, the nonlinear control framework module includes a nonlinear framework module based on an energy storage mechanism, an adaptive PD controller module based on a force sensing mechanism, and a decoupling motion controller module.

[0069] In some embodiments of the present invention, the nonlinear frame module based on the energy storage mechanism is designed as formula (3).

[0070]

[0071] Among them J f ,B f ,L f and u 1-d (t) are the moment of inertia, damping, linkage length and input torque of the elbow joint, m f is the linkage mass. g is the acceleration due to gravity, τ e (θ 1-a ) can be obtained from formula (2), δ d Is the upper bound of the system disturbance. e1(t)=θ 1-a -θ 1-d ,θ 1-d is the target angle of the elbow joint. Terminal function It can be obtained by a quintic polynomial (quintic polynomial is a commonly used algorithm in local path planning, such as https: / / blog.csdn.net / weixin_43794327 / article / details / 123335112 for a detailed introduction to quintic polynomial). In order to reduce jitter, the sigmoid function in formula (4) is used, where δ sats is the boundary layer thickness. d represents the upper bound of the system disturbance, s represents the expression after Laplace transform, c is an adjustable performance parameter, subscript a represents the actual attribute, f represents the overall system, and d has no specific reference.

[0072]

[0073] In some embodiments of the present invention, the adaptive PD controller module based on the force sensing mechanism drives the elbow joint with nonlinear efficiency during the training task. Due to different postures, different lengths of Bowden cables need to be released / recovered to rotate the same angle. In order to track the target torque u 1-d (t), the design of the observer is shown in formula (5), and the design of the observer is to observe the driving efficiency in real time. Figure 5 shown.

[0074] δ v =L um sin(π-θ1-θ u -θ f ) / (3L B ) (5)

[0075] Where, δv Represents the driving efficiency of the system, L um represents the upper arm driving length, θ1 represents the elbow joint rotation angle, θ u Represents the upper arm drive angle, θ f Represents the forearm driving angle, L B Represents the rope drive length.

[0076] Designing an adaptive PD controller module based on the driving efficiency can further solve the problem of control hysteresis.

[0077] The target velocities of the left and right sides of the drive rope required by the elbow joint are shown in equations (6) and (7), where u l-d =u r-d =0.5u 1-d ,e u-l =u l-d -u l-a ,e u-r =u r-d -u r-a , where u l-d and u r-d are the target torques for the left and right drive ropes, respectively.

[0078]

[0079]

[0080] Where, v 1-l (t) represents the velocity of the left driving rope of the elbow joint, δ v (θ 1-a ) represents the elbow joint driving efficiency, K 1-p represents the P parameter of the elbow joint PD controller, e u-l (t) represents the PD controller error input of the left drive rope of the elbow joint, K 1-d represents the D parameter of the elbow joint PD controller, v 1-r (t) represents the speed of the right driving rope of the elbow joint, e u-r (t) represents the error input of the PD controller of the right drive rope of the elbow joint, the subscript r represents right, u represents the control signal, p represents the parameter P of PD control, and v represents the velocity.

[0081] In some embodiments of the present invention, the decoupled motion controller module is designed as shown in formulas (8) and (9), where v l is the speed of the left driving rope, v r is the speed of the right drive rope.

[0082]

[0083]

[0084] represents the target velocity of the elbow joint, Represents the target velocity of the wrist joint.

[0085] In some embodiments of the present invention, the safety protection strategy module detects the interaction force information through a force sensor. In order to consider the safety of the exoskeleton during rehabilitation training, a safety protection strategy is proposed, which integrates the tension of the drive rope and the actual posture to evaluate whether an abnormality has occurred. Then, the exoskeleton will switch to free mode to ensure safety (free mode is also called zero gravity mode, which can be dragged freely). Abnormal situations will be divided into two cases: 1) Spasm is a spontaneous bending of the elbow joint, which may cause muscle strain. Therefore, the Bowden cable may be loose, and the judgment condition is shown in formula (10), where u 1-a =u l-a +u r-a , where u l-a and u r-a Obtained by the force sensing modules of the left and right drive ropes. 2) In unstructured environments, collisions caused by obstacles need to be considered, which may put the exoskeleton and the user in danger. The occurrence of obstacles will increase the tension and the target posture cannot be tracked. Therefore, the judgment condition is shown in formula (11), where δ u is the torque margin. When (10) or (11) is satisfied, the exoskeleton will immediately switch to free mode. In order to offset the reverse torque and gravity, the target torque is calculated by formula (12).

[0086] u 1-a <u m i n (10)

[0087] u l-a ≥u max =m f gL f sinθ 1-a +τ e (θ 1-a )+δ u (11)

[0088] u 1-d =m f gL f sinθ 1-a +τ e (θ 1-a ) (12)

[0089] Where u 1-a represents the actual torque of the elbow joint, u min Represents the lower limit of the torque of the elbow joint, u maxRepresents the upper limit of the torque of the elbow joint, u 1-d Represents the target torque of the elbow joint.

[0090] Figure 3 The schematic diagram of the control system of the coupled control rope driven exoskeleton is shown. First, the target angle θ of the elbow joint of the exoskeleton is input. 1-d ,speed and acceleration To the nonlinear frame module based on the energy storage mechanism, a control signal is obtained, which is adjusted by the feedback input of the energy storage mechanism module, and then the control signal is input into the elbow joint controller to obtain the control signal of the elbow joint. In addition, the target angle θ of the wrist joint is input 2-d The adaptive PD controller module based on the force sensing mechanism of the wrist joint obtains the control signal of the wrist joint. Then the control signal of the elbow joint and the control signal of the wrist joint are input into the decoupled motion controller module to drive the exoskeleton. Among them, the posture sensor is used as a sensor to feedback the current posture, including the actual rotation angle θ of the elbow joint. 1-a and the wrist rotation angle θ 2-a During the entire process, the safety protection strategy module will constantly detect the actual torque of the elbow joint to protect the exoskeleton.

[0091] Figure 6 A schematic diagram of a verification platform comparing the effects of direct force sensing and the force sensing mechanism of the present invention is shown. From the results, it can be seen that the present invention does not interfere with the movement of the exoskeleton, while the direct force sensing solution will interfere with the movement of the exoskeleton.

[0092] Figure 7 The comparison results of tension measurement are shown. It can be seen from the figure that the average measurement error of the present invention is 0.119N, and the correlation coefficient R 2 It is 0.996. Obviously, the proposed force sensing mechanism can accurately measure the tension of the Bowden cable.

[0093] Figure 8 It is shown that by using the force sensing mechanism of the present invention, the position tracking curves of the elbow joint and the wrist joint are smooth and without abrupt changes, which will not interfere with the normal coupling movement of the exoskeleton driven by the coupling control rope. The elbow joint inputs a sinusoidal signal and the wrist joint inputs a ramp signal, and the speed curves of the Bowden cables on both sides in the coupling movement can be seen.

[0094] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A force sensing and control system for a coupled control rope driven exoskeleton, characterized in that: Including force sensing mechanism and control system module, The force sensing mechanism includes a force sensing module and an energy storage mechanism module, the force sensing module is used to transmit and feedback human-computer interaction force information, and the energy storage mechanism module is used to store elastic potential energy; The control system module includes a nonlinear control framework module and a safety protection strategy module. The nonlinear control framework module includes a nonlinear framework module based on an energy storage mechanism, an adaptive PD controller module based on a force sensing mechanism, and a decoupling motion controller module, which are used to control the movement of the exoskeleton driven by the drive rope and reduce the control impact caused by nonlinearity. The nonlinear frame module based on the energy storage mechanism is used to reduce the unidirectional force characteristics of the rope drive and the influence of nonlinearity; The adaptive PD controller module based on the force sensing mechanism is used to compensate for the nonlinear driving efficiency of the elbow joint; The decoupling motion controller module is used to solve the coupling control planning of the elbow joint and the wrist joint; The safety protection strategy module is used to detect interaction force information and switch the exoskeleton mode to protect human-machine safety; The force sensing module is arranged between the human forearm connecting piece (5) and the human upper arm connecting piece (6), and the force sensing module includes two symmetrically arranged force sensor units, each force sensor unit includes a driving rope (1), a pulley group (2), and a force sensor (3), and the driving rope (1), the pulley group (2), and the force sensor (3) form a fixed and movable pulley combination, which serves as a driving source for the force sensing module and the exoskeleton, wherein one end of the driving rope (1) is connected to the human forearm connecting piece (5) through the pulley group (2), and the other end is connected to the motor; Taking the driving rope (1) of one of the force sensor units as an example, when the elbow joint is rotated at a constant speed and there is tension, the driving rope (1) needs to provide tension to balance the total gravity on the forearm. Since the tensions of different parts of the same driving rope (1) are equal, formula (1) is obtained. At the same time, the force sensing mechanism does not involve the structural form of the joint by introducing the pulley group (2), and does not interfere with the coupling action of the rope drive device. (1) Where, represents the tension of a single section of drive rope, represents the total force on the system, Represents the perception of the force sensing module; The energy storage mechanism module includes a plurality of tension springs (4), and the plurality of tension springs (4) equally divide a circumference. The geometric relationship of the energy storage mechanism module is determined by formula (2): (2) Where, Represents the reverse torque generated by the energy storage mechanism module, Represents the inner diameter of the energy storage mechanism module, Represents the outer diameter of the energy storage mechanism module, represents the rotation angle of the elbow joint, represents the spring stiffness of the energy storage mechanism module, Represents the original length of the spring of the energy storage mechanism module.

2. The force sensing and control system for a coupled control rope-driven exoskeleton according to claim 1, characterized in that: The nonlinear frame module based on the energy storage mechanism is designed as formula (3): (3) Where, J f , B f , L f and are the moment of inertia, damping, linkage length and input torque of the elbow joint, m f It is the linkage quality, g is the acceleration due to gravity, Represents the reverse torque generated by the energy storage mechanism module, represents the rotation angle of the elbow joint, is the upper bound of the system perturbation, , is the target angle of the elbow joint, 、 is the terminal function, is a saturation function, c is an adjustable performance parameter, subscript a Represents the actual attributes, f Represents the overall system, d No specific reference.

3. The force sensing and control system for a coupled control rope-driven exoskeleton according to claim 2, characterized in that: To reduce jitter, a saturation function is used, as shown in formula (4): (4) Where, is the boundary layer thickness, s represents the expression after Laplace transformation.

4. The force sensing and control system for a coupled control rope-driven exoskeleton according to claim 1, characterized in that: The design of the observer of the adaptive PD controller module based on the force sensing mechanism is shown in formula (5): (5) Where, Represents the driving efficiency of the system, represents the upper arm drive length, represents the elbow joint rotation angle, Represents the upper arm driving angle, represents the forearm driving angle, represents the rope drive length; The target velocities of the left and right sides of the drive rope required by the elbow joint are shown in equations (6) and (7): (6) (7) Where, represents the speed of the left driving rope of the elbow joint, represents the elbow joint driving efficiency, represents the P parameter of the elbow joint PD controller, represents the error input of the PD controller of the left drive rope of the elbow joint, represents the D parameter of the elbow joint PD controller, represents the speed of the right driving rope of the elbow joint, represents the error input of the PD controller of the right drive rope of the elbow joint, Represents the right, Represents the control signal, represents the parameter P of PD control, Represents speed, and are the target torques for the left and right drive ropes, respectively.

5. The force sensing and control system for a coupled control rope driven exoskeleton according to claim 1, characterized in that: The decoupled motion controller module is designed as shown in formulas (8) and (9): (8) (9) Where, represents the target velocity of the elbow joint, represents the target velocity of the wrist joint, is the speed of the left driving rope, is the speed of the right driving rope, Represents the driving efficiency of the system.

6. A force sensing and control system for a coupled control rope driven exoskeleton according to any one of claims 1 to 5, characterized in that: The safety protection strategy module integrates the tension of the drive rope and the actual posture to evaluate whether an abnormality occurs. Then, the exoskeleton will switch to free mode to ensure safety. Abnormal situations include two cases: 1) Spasm is a spontaneous bending of the elbow joint, which may cause muscle strain. Therefore, the Bowden cable may be loose. The judgment condition is shown in formula (10); 2) In an unstructured environment, collisions caused by obstacles need to be considered. It may put the exoskeleton and the user in danger. The occurrence of obstacles will increase the tension and the target posture cannot be tracked. Therefore, the judgment condition is shown in formula (11). When formula (10) or formula (11) is met, the exoskeleton will immediately switch to free mode: (10) (11) Where, represents the actual torque at the elbow joint, represents the lower limit of the torque of the elbow joint, represents the upper limit of the torque of the elbow joint, represents the target torque of the elbow joint, is the torque margin, m f It is the linkage quality, g is the acceleration due to gravity, L f is the linkage length of the elbow joint, represents the rotation angle of the elbow joint, Represents the reverse torque generated by the energy storage mechanism module.

7. The force sensing and control system for a coupled control rope driven exoskeleton according to claim 6, characterized in that: In order to counteract the reverse torque and gravity, the target torque is calculated by formula (12): (12) Where, represents the target torque of the elbow joint, represents the rotation angle of the elbow joint, Represents the reverse torque generated by the energy storage mechanism module.

Images