A variable stiffness gear five-bar joint mechanism and a control system and control method thereof

Abstract

The application discloses a variable stiffness gear five-bar joint mechanism, which comprises a five-bar linkage mechanism, a ball screw mechanism I, a ball screw mechanism II, a gear pair, a main driving motor and a variable stiffness motor. The variable stiffness gear five-bar joint mechanism can not only realize the movement of the five-bar linkage mechanism through the main driving motor and the gear, and realize the movement characteristics of the joint, but also can realize the change of the spring mechanism position through the variable stiffness motor, and further change the mechanical characteristics of the joint mechanism. The application further discloses a control system and a control method of the variable stiffness joint. The control system comprises an angle sensor, a gait recognition algorithm module, a position selection module and a motor control algorithm module. The control method comprises the following steps: s1. establishing a BP neural network gait recognition algorithm; s2. establishing a position selection algorithm; and s3. PID controller design. The application can effectively solve the contradiction between the PID overshoot and the gait inconsistency in the variable stiffness joint system, and has high control precision and bionic performance.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of bionics, and particularly relates to a variable stiffness gear five-bar joint mechanism, a control system and a control method of the variable stiffness joint based on a BP neural network. BACKGROUND

[0002] Prosthetic joints are important devices for lower limb rehabilitation, which can provide standing support and swing driving for the disabled, and greatly improve their lives. Prostheses are important tools for motion function replacement for amputees, and are artificial limbs manufactured and assembled to compensate for limb defects and compensate for limb function. Although the current joint mechanism can realize basic rotation in functionality, the rotation characteristics do not conform to the motion characteristics of the human body, and the rigid joint also does not conform to the physiological characteristics of the human body. Therefore, the bionics of the prosthetic joint is an urgent task to solve the basic needs of amputees.

[0003] On the other hand, the five-bar joint mechanism system proposed by the application is a high-precision position servo system, which is an extremely important component of upper limb prostheses, lower limb prostheses, and exoskeleton robots, and is an important actuator of the control system. The upper computer system sends the joint expected gait command to the joint controller, and the joint resolver drives the main drive motor and the variable stiffness motor to rotate, thereby realizing the rotation of the joint under good mechanical characteristics. However, due to the uncertainty of human motion, the control of the joint becomes very complex. Another purpose of the application is to propose a high-precision, high-robustness, and high-reliability bionic joint control system that can predict gait and realize variable stiffness, so that the joint motion is more stable and more in line with the motion characteristics of the human body. SUMMARY

[0004] One of the purposes of the application is to solve the problems of poor bionic performance and non-adjustable stiffness of the existing joint mechanism, and to provide a new scheme to realize the bionic characteristics of high-performance joints.

[0005] The second purpose of the application is to propose a high-precision, high-robustness, and high-reliability bionic joint control system that can predict gait and realize variable stiffness, so that the joint motion is more stable and more in line with the motion characteristics of the human body, has high control precision and safety performance, and has important practical significance and application value.

[0006] The variable stiffness gear five-bar joint mechanism of the application comprises a five-bar linkage, a ball screw mechanism I, a ball screw mechanism II, a gear pair, a main drive motor and a variable stiffness motor; the gear pair is used to make the degree of freedom of the five-bar linkage be 1, and comprises gear I and gear II which are arranged on two links of the five-bar linkage respectively and mesh with each other; the ball screw mechanism I is used to drive the relative rotation of the two links of the five-bar linkage under the drive of the main drive motor so that the five-bar linkage is deformed; the ball screw mechanism II is used to pull the spring connected with a link of the five-bar linkage to adjust the deformation stiffness of the five-bar linkage under the drive of the variable stiffness motor.

[0007] Further, the five-bar linkage comprises a rack bar, a link I, a link II, a link III and a link IV which are hingedly connected in sequence; the gear I and the gear II of the gear pair are arranged on the rack bar and the link III respectively.

[0008] Further, the main drive motor and the ball screw mechanism I are arranged on the rack bar; the ball screw mechanism I comprises a ball screw I arranged on the rack bar and a nut block I matched with the ball screw I; a pull rod is hingedly connected between the nut block I and the link I; the main drive motor drives the ball screw I to rotate through a synchronous belt transmission mechanism;

[0009] Further, the ball screw mechanism II comprises a ball screw II supported on the rack bar and a nut block II matched with the ball screw II; the spring is connected between the nut block II and the link II; the variable stiffness motor drives the ball screw II to rotate through a gear transmission mechanism;

[0010] The application further discloses a control system of the variable stiffness joint based on a BP neural network, which comprises an angle sensor, a gait recognition algorithm module, a position selection module and a motor control algorithm module; the angle sensor is signal-connected to a joint output end and transmits obtained angle information to the gait recognition algorithm module; the gait recognition algorithm module is used for processing the obtained angle signal to obtain a control amount of the main drive motor and processing the obtained angle signal to obtain a walking step speed; the position selection module is used for processing the walking step speed signal from the gait recognition algorithm module to obtain a control amount of the variable stiffness motor; the motor control algorithm is used for controlling the motor to reach the expected position through a PID algorithm.

[0011] Further, the gait recognition algorithm module is a trained BP neural network gait predictor; the BP neural network gait predictor adopts a BP neural network which has four layers in total, including an input layer with 15 input nodes, a double hidden layer with 10*7, and an output layer with two outputs; the input layer uses a time window of 5 samples as input, wherein one sample contains the input of thigh angle, calf angle and foot load; the two outputs of the output layer are predicted joint angle and walking speed.

[0012] Further, the position selection module has five modes in total, corresponding to extremely slow, slow, natural, medium and fast stiffness modes respectively, and the mode judgment is realized by a finite state machine.

[0013] The application further discloses a control method of the variable stiffness joint based on the BP neural network.

[0014] s1. establishing a model of the BP neural network;

[0015] s2. training the network by using different walking speed data and 14% error data composed of noise or accident error;

[0016] s3. establishing a position selection program model.

[0017] The application has the following beneficial effects:

[0018] (1) The variable stiffness gear five-bar joint mechanism improves the joint transmission characteristics and mechanical properties.

[0019] (2) The variable stiffness gear five-bar joint mechanism has low degrees of freedom and simple driving mode.

[0020] (3) The variable stiffness gear five-bar joint mechanism has simple structure, is convenient to disassemble and install, and is easy to maintain.

[0021] (4) The control system and control method of the variable stiffness joint can predict gait and realize variable stiffness, so that the joint movement is more stable and more in line with human motion characteristics. BRIEF DESCRIPTION OF DRAWINGS

[0022] The application will be further described below in combination with the drawings and examples.

[0023] Figure 1 It is a schematic diagram of the overall structure of the variable stiffness gear five-bar joint mechanism of the application.

[0024] Figure 2 It is a schematic diagram of the rack rod in the variable stiffness gear five-bar joint mechanism of the application.

[0025] Figure 3This is a schematic diagram of link III in the variable stiffness gear five-bar linkage mechanism of the present invention;

[0026] Figure 4 This is a schematic diagram of the variable stiffness joint control system of the present invention;

[0027] Figure 5 This is a diagram of the BP neural network structure of the present invention. Detailed Implementation

[0028] Example 1: Variable stiffness gear five-bar linkage mechanism

[0029] like Figures 1-3 As shown, the variable stiffness gear five-bar linkage mechanism of this embodiment includes a five-bar linkage, a ball screw mechanism I, a ball screw mechanism II, a gear pair, a main drive motor and a variable stiffness motor 7;

[0030] The five-bar linkage includes frame rod 5, link I 10, link II 11, link III 15 and link IV 17, which are hinged from end to end.

[0031] The gear pair is used to make the five-bar linkage have 1 degree of freedom, and includes gears I and II respectively disposed on the frame link 5 and the connecting link III 15 in the five-bar linkage and meshing with each other; such as Figure 2 As shown, the frame rod 5 has a T-shaped structure. The middle part is used to install the ball screw mechanism II, the left side is used to install the ball screw mechanism I, and the right side is used to install the main drive motor 14. The right end of the frame rod 5 has an integrally formed gear I, and one end of the connecting rod III 15 has an integrally formed gear II. Both gear I and gear II are incomplete gears.

[0032] Driven by the main drive motor 14, the ball screw mechanism I drives the frame rod 5 and the connecting rod I10 in the five-bar linkage to rotate relative to each other, causing the five-bar linkage to deform. The ball screw mechanism I and the main drive motor 14 are respectively located on the left and right sides of the frame rod 5. The ball screw mechanism I includes a ball screw I6 located on the frame rod 5 and a nut slider I8 that cooperates with the ball screw I6. A tie rod 9 is hinged between the nut slider I8 and the connecting rod I10. The main drive motor 14 drives the ball screw I6 to rotate through a synchronous belt transmission mechanism. A drive pulley is installed on the output shaft of the main drive motor 14, and a driven pulley 4 is installed at the lower end of the ball screw I6. The drive pulley 16 and the driven pulley are connected by a synchronous belt 1. Therefore, the main drive motor 14 can drive the ball screw I6 to rotate through the synchronous belt transmission mechanism, causing the nut slider I8 to move along the axis of the ball screw I6, and finally drive the connecting rod I10 to rotate relative to the frame rod 5 through the tie rod 9.

[0033] The ball screw mechanism II is driven by the variable stiffness motor 7 to pull the spring 13 connected with the connecting rod II 11 in the five connecting rod mechanism to adjust the deformation stiffness of the five connecting rod mechanism; the ball screw mechanism II includes a ball screw II 12 supported in the middle of the rack rod 5 and a nut block II matched with the ball screw II 12; the spring 13 is connected between the nut block II and the connecting rod II 11; the variable stiffness motor 7 drives the ball screw II 12 to rotate through a gear transmission mechanism (including gear III 2 and gear IV 3 meshing with each other), so as to drive the left end of the spring 13 to move through the nut block II, and the deformation stiffness of the whole five connecting rod mechanism can be adjusted by adjusting the stretching amount of the spring 13.

[0034] The working principle of the variable stiffness gear five-bar joint mechanism of the embodiment is as follows: the main drive motor 14 drives the ball screw I 6 to rotate through the synchronous belt transmission mechanism, and then drives the pull rod to move, the pull rod drives the connecting rod I 10 to rotate through the hinge, the movement of the connecting rod I 10 drives the connecting rod II 11, the connecting rod III 15 and the connecting rod IV 17 to move through the hinge, and the gear on the connecting rod III 15 meshes with the gear on the rack rod 5 to reduce the degree of freedom of movement, so that the degree of freedom of the five-bar joint mechanism is 1, and then when the main drive motor 14 is driven, the connecting rod I 10 (connected with the prosthetic thigh mechanism) and the connecting rod IV 17 (connected with the prosthetic lower leg mechanism) can rotate relative to the expected instantaneous center. When the variable stiffness motor 7 drives the ball screw II 12 to rotate through the gear transmission mechanism, the nut block II reciprocates up and down, and then drives one end of the spring to move, so as to change the mechanical properties of the whole joint and realize the variable stiffness and bionic characteristics.

[0035] Embodiment two: control system and control method of variable stiffness joint

[0036] The control system of the variable stiffness joint based on the BP neural network of the embodiment includes an angle sensor, a gait recognition algorithm module, a position selection module and a motor control algorithm module; the angle sensor is signal connected to the joint output end and transmits the obtained angle information to the gait recognition algorithm module; the gait recognition algorithm module is used to process the obtained angle signal to obtain the control amount (joint angle) of the main drive motor and the walking step speed obtained by processing the obtained angle signal; the position selection module is used to process the walking step speed signal from the gait recognition algorithm module to obtain the control amount of the variable stiffness motor; the motor control algorithm is used to control the motor to reach the expected position through the PID algorithm.

[0037] The gait recognition algorithm module is a trained BP neural network gait predictor; the BP neural network gait predictor adopts a BP neural network having four layers in total, including an input layer having 15 input nodes, a double hidden layer of 10x7, and an output layer having two outputs; the input layer uses a time window of 5 samples as input, wherein one sample contains the input of thigh angle, calf angle and foot load; the two outputs of the output layer are predicted joint angle and walking speed.

[0038] The position selection module has five modes (λ1, λ2, λ3, λ4 and λ5) in total, corresponding to extremely slow, slow, natural, medium and fast stiffness modes respectively, and the mode determination is realized by a finite state machine.

[0039] The flow of controlling the variable stiffness joint movement by the control system of the embodiment is as follows: the host computer sends an expected gait joint controller, the joint controller inputs the current step speed and the expected output angle of the main drive motor according to the feedback value of the angle sensor into the gait recognition algorithm module to predict; the position selection program selects the appropriate stiffness according to the current step speed and calculates the output angle of the variable stiffness motor according to the angle feedback information; the PID position controller controls the motor to reach the specified position according to the expected angle. The variable stiffness motor can change the joint stiffness, thereby realizing better bionic characteristics and mechanical properties; the main drive motor drives the joint to rotate.

[0040] The control method of the variable stiffness joint based on the BP neural network of the embodiment includes the following steps:

[0041] s1. establishing a model of the BP neural network;

[0042] s2. training the network by using different walking speed data and adding 14% error data composed of noise or accident error;

[0043] s3. establishing a position selection program model.

[0044] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit it. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced equivalently without departing from the purpose and scope of the present application, and they should be covered in the scope of the claims of the present application.

Claims

1. A variable stiffness gear five-bar linkage mechanism, characterized in that: It includes a five-bar linkage, ball screw mechanism I, ball screw mechanism II, gear pairs, main drive motor, and variable stiffness motor; The five-bar linkage includes a frame rod, link I, link II, link III, and link IV that are hinged together from end to end; gear I and gear II of the gear pair are respectively mounted on the frame rod and link III; The gear pair is used to make the degree of freedom of the five-bar linkage 1, and it includes gear I and gear II, which are respectively disposed on two links in the five-bar linkage and mesh with each other. The main drive motor and the ball screw mechanism I are both mounted on the frame rod. Under the drive of the main drive motor, the ball screw mechanism I is used to drive the frame rod and the connecting rod I in the five-bar linkage to rotate relative to each other, causing the five-bar linkage to deform. Driven by a variable stiffness motor, the ball screw mechanism II is used to pull a spring connected to the connecting rod II in the five-bar linkage to adjust the deformation stiffness of the five-bar linkage; the spring is connected between the nut slider II of the ball screw mechanism II and the connecting rod II; the ball screw mechanism II includes a ball screw II supported on the frame rod and a nut slider II that cooperates with the ball screw II.

2. The variable stiffness gear five-bar linkage mechanism according to claim 1, characterized in that: The ball screw mechanism I includes a ball screw I mounted on the frame rod and a nut slider I that cooperates with the ball screw I; a tie rod is hinged between the nut slider I and the connecting rod I; the main drive motor drives the ball screw I to rotate through a synchronous belt transmission mechanism.

3. The variable stiffness gear five-bar linkage mechanism according to claim 1, characterized in that: The variable stiffness motor drives the ball screw II to rotate through a gear transmission mechanism.

Images