A rope force control method for time-varying configuration

Abstract

The present invention provides a rope force control method that can be used for time-varying configuration, including: friction model establishment and parameter calibration; real-time parameter calculation and force control: parameter identification based on the friction model to obtain the parameters of the auxiliary rope Bowden system and the power Bowden system: the friction coefficient μ of the auxiliary rope Bowden system a , friction coefficient of power rope Bowden system μ p ; Calculate the auxiliary rope Bowden system θ in real time based on the model and sensor force value a , and as a powered rope Bowden system θ p The rope bending angle; According to the friction model, the inverse control formula F can be obtained in =F out ·e ‑uλθ , the power rope Bowden system θ p The inverse control formula is introduced as a feedforward controller to achieve real-time force control. This method solves the force control difficulties caused by the change of rope Bowden configuration in the wearable rope robot, and effectively improves the force control accuracy of the rope Bowden system under variable configuration.

Description

Technical Field

[0001] The invention belongs to the field of robot technology, relates to wearable exoskeleton robot technology, and specifically relates to a rope force control method based on a force sensor that can be used for time-varying configuration. Background Art

[0002] A rope drive system is a mechanical system that uses ropes and Bowden cables to transmit torque. Generally speaking, a rope drive system consists of a hollow outer sheath and an inner rope. Due to its light weight, low inertia and good flexibility, the rope drive system is more attractive for wearable exoskeleton robots than other solutions based on drive shafts or gears. However, the nonlinear relationship between the proximal force and the distal force deteriorates the control performance of the system.

[0003] In order to reduce the nonlinear effect, the transmission characteristic model has been extensively studied by scholars. In terms of control methods, the research can be divided into the following three categories: feedforward control, feedback control, and composite control combining feedforward and feedback control. Feedback control uses force sensors or changes in spring torque to indirectly complete feedback control, but the end sensor limits the displacement of the rope, thereby affecting flexibility. Combining the force feedback information at the far end with the feedforward is another way to improve the performance of the end effector. This combination can reduce the accuracy of the feedforward model, but the composite control requires the installation of a force sensor at the far end, which limits the flexibility of the end device such as a surgical robot. Considering the practicality and complexity of the control strategy, feedforward control has been widely studied. However, existing research is based on the modeling of the fixed-position rope system, ignoring the situation where the actual robot motion affects the shape of the rope system, so the force control of the time-varying configuration is a challenging problem. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a rope force control method that can be used for time-varying configuration based on a magnetorheological damper and a force sensor. The present invention is based on a self-built experimental platform, and uses a force sensor and a self-made magnetorheological damper to achieve. First, the friction coefficient and other model parameters are calibrated before using the rope system, and then during the use of the rope system, the angle is estimated according to the algorithm provided by the present invention, so as to achieve real-time force control.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A rope force control method applicable to time-varying configuration comprises the following steps:

[0007] Step 1: Friction model establishment and parameter calibration

[0008] The quasi-static equilibrium equation of the (s+ds) segment in the rope is expressed as:

[0009]

[0010] where F(s,t), F(s+ds,t), N(s,t), f(s,t) and κ(s,t) represent the rope tension at position s, the rope tension at position s+ds, the normal pressure, the friction force and the curvature at time t, respectively. Now define λ=sign[F'(s)], and according to formula (1), we can get:

[0011]

[0012] λ represents the direction of rope friction, and then integrating over the interval (s+ds), we can get the following expression:

[0013]

[0014] Expanding the above expression to the total length L, we can get:

[0015] F(L,t)=F(0,t)·e μλθ (4)

[0016] Where μ and θ represent the friction coefficient and the total bending angle of the rope respectively. Now let F out =F(L,t),F in =F(0,t) Through formula (4), we can get:

[0017]

[0018] In order to reduce the error between the model and experimental data, the parameter μ is optimized according to the following optimization rule:

[0019]

[0020] where F out , and N represent the actual measured force value, the force value predicted by the model, and the number of sampling times, respectively;

[0021] Step 2: Real-time parameter calculation and force control, including the following sub-steps:

[0022] S1 performs parameter identification based on the above friction model to obtain the parameters of the auxiliary rope Bowden system and the power Bowden system: The friction coefficient of the auxiliary rope Bowden system μ a , friction coefficient of power rope Bowden system μ p ;

[0023] S2 calculates the auxiliary rope Bowden system θ in real time based on the model and sensor force value a , and as a powered rope Bowden system θ p The bending angle of the rope;

[0024] S3 According to the friction model, the inverse control formula F can be obtained in =F out ·e -uλθ , the power rope Bowden system θ p Substitute the inverse control formula into it and use it as a feedforward controller to achieve real-time force control.

[0025] Furthermore, in step S1, using the tension sensors on the auxiliary rope Bowden system and the power rope Bowden system, the proximal force is defined as the input force, and the distal force is defined as the output force. The parameter with the minimum error is defined as μ by formula (6). a and μ p .

[0026] Furthermore, in step S2, the auxiliary rope Bowden system sets the expected sinusoidal input, and the auxiliary rope Bowden system θ is calculated using formula (5). a .

[0027] Furthermore, in step S3, the desired force value at the far end of the power rope Bowden system is set, and the inverse control model F in =F out ·e -uλθ The corresponding expected input value can be obtained, and the angle value of the power rope Bowden system can be calculated in real time by the auxiliary rope Bowden system.

[0028] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0029] This method can achieve real-time force control effect, solve the force control difficulty caused by the change of rope Bowden configuration in the wearable rope robot, and effectively improve the force control accuracy of the rope Bowden system under variable configuration. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a flow chart of a rope force control method that can be used for time-varying configuration in an embodiment of the present invention;

[0031] Figure 2 It is a schematic diagram of parameters derived from the friction model of the present invention;

[0032] Figure 3 It is a module diagram of the self-made magnetorheological damper described in the present invention;

[0033] Figure 4 is a schematic diagram of a verification platform for the force control method of the present invention;

[0034] Description of reference numerals:

[0035] 1-motor fixing seat, 2-small pulley, 3-belt, 4-motor, 5-winding shaft, 6-slip ring, 7-slip ring fixing seat, 8-magnetorheological damper end fixing seat, 9-large pulley, 10-magnetorheological damper, 11-magnetorheological damper head end fixing seat, 12-module base, 13-reel shaft fixing seat, 14-magnetorheological damper module, 15-electric push rod, 16-tension sensor, 17-rope guide seat, 18-auxiliary rope Bowden system, 19-power rope Bowden system, 20-rope fixing seat, 21-spring. DETAILED DESCRIPTION

[0036] The technical solution provided by the present invention will be described in detail below in conjunction with specific embodiments. It should be understood that the following specific implementation methods are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0037] The rope force control method provided by the present invention can be used for time-varying configuration, and its process is as follows: Figure 1 As shown, it includes the following steps:

[0038] Step 1: Friction model establishment and parameter calibration

[0039] Figure 1 This is a schematic diagram of the friction model derivation parameters, where the quasi-static equilibrium equation of the (s+ds) segment is expressed as:

[0040]

[0041] Where F(s,t), F(s+ds,t), N(s,t), f(s,t) and κ(s,t) represent the rope tension at position s, the rope tension at position s+ds, the normal pressure, the friction force and the curvature at time t, respectively. Now define λ=sign[F'(s)], and according to formula (1), we can get:

[0042]

[0043] λ represents the direction of rope friction. If it is greater than zero, it means that the friction force is consistent with the movement trend of the rope, and if it is less than zero, it means that it is opposite to the movement trend of the rope. Then integrate on the interval (s+ds), we can get the following expression:

[0044]

[0045] Expanding the above expression to the total length L, we can get:

[0046] F(L,t)=F(0,t)·e μλθ (4)

[0047] Where μ and θ represent the friction coefficient and the total bending angle of the rope respectively. Now let F out=F(L,t),F in =F(0,t) Through formula (4), we can get:

[0048]

[0049] In order to reduce the error between the model and experimental data, the parameter μ is optimized according to the following optimization rule:

[0050]

[0051] where F out , and N represent the actual measured force value, the force value predicted by the model, and the number of sampling times, respectively.

[0052] Step 2: Real-time parameter calculation and force control, including the following sub-steps:

[0053] S1 performs parameter identification based on the above friction model to obtain the parameters of the auxiliary rope Bowden system and the power Bowden system: The friction coefficient of the auxiliary rope Bowden system μ a , friction coefficient of power rope Bowden system μ p .

[0054] Specifically, the motor 4 transmits torque to the magnetorheological damper 10 through the synchronous wheel 2 and the belt 3. The actual output torque of the magnetorheological damper is controlled by controlling the current of the slip ring 6. The torque generates tension on the rope winding shaft 5 fixed thereon. Using the tension sensors on the auxiliary rope Bowden system 18 and the power rope Bowden system 19, the proximal force is defined as the input force, and the distal force is defined as the output force. The parameter with the minimum error is defined as μ by calculating the formula (6). a and μ p .

[0055] S2 calculates the auxiliary rope Bowden system θ in real time based on the model and sensor force value a , and as a powered rope Bowden system θ p The bending angle of the rope;

[0056] The auxiliary rope Bowden system sets the expected sinusoidal input and uses formula (5) to calculate the auxiliary rope Bowden system θ a , since the auxiliary rope Bowden system and the power rope Bowden system are fixed side by side, the bending angles of the two systems can be almost equal, that is, θ p =θ a .

[0057] S3 According to the friction model, the inverse control formula F can be obtained in =F out ·e -uλθ , the power rope Bowden system θ pSubstitute the inverse control formula into it and use it as a feedforward controller to achieve real-time force control.

[0058] Simulation test platform such as Figure 4 As shown, due to the real-time force control method proposed in this paper, in order not to affect the flexibility of the end, the auxiliary rope Bowden system adopts the method of fixing the end and being still. It is difficult for traditional motors to maintain force control under long-term stall conditions. Therefore, a magnetorheological damper module 14 is made by ourselves. The specific parts of the module are as follows Figure 3 In the test platform, the spring 21 is used as an external load simulation actuator to set the desired force value at the far end of the power rope Bowden system. in =F out ·e -uλθ , the corresponding expected input value can be obtained, and the angle value of the power rope Bowden system can be calculated in real time by the auxiliary rope Bowden system, so real-time force control effect can be achieved.

[0059] The technical means disclosed in the scheme of the present invention are not limited to the technical means disclosed in the above-mentioned implementation mode, but also include technical schemes composed of any combination of the above-mentioned technical features. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications are also regarded as the protection scope of the present invention.

Claims

1. A rope force control method applicable to time-varying configuration, characterized in that: The steps include: Step 1: Friction model establishment and parameter calibration The quasi-static equilibrium equation of the (s+ds) segment in the rope is expressed as: where F(s,t), F(s+ds,t), N(s,t), f(s,t) and κ(s,t) represent the rope tension at position s, the rope tension at position s+ds, the normal pressure, the friction force and the curvature at time t, respectively. Now define λ=sign[F'(s)], and according to formula (1), we can get: λ represents the direction of rope friction, and then integrating over the interval (s+ds), we can get the following expression: Expanding the above expression to the total length L, we can get: F(L,t)=F(0,t)·e μλθ (4) Where μ and θ represent the friction coefficient and the total bending angle of the rope respectively. Now let F out =F(L,t),F in =F(0,t) Through formula (4), we can get: In order to reduce the error between the model and experimental data, the parameter μ is optimized according to the following optimization rule: Among them, F out , and N represent the actual measured force value, the force value predicted by the model, and the number of sampling times, respectively; Step 2: Real-time parameter calculation and force control, including the following sub-steps: S1 performs parameter identification based on the above friction model to obtain the parameters of the auxiliary rope Bowden system and the power Bowden system: The friction coefficient of the auxiliary rope Bowden system μ a , friction coefficient of power rope Bowden system μ p ; S2 calculates the auxiliary rope Bowden system θ in real time based on the model and sensor force value a , and as a powered rope Bowden system θ p The rope bending angle; S3 According to the friction model, the inverse control formula F can be obtained in =F out ·e -uλθ , the power rope Bowden system θ p Substitute the inverse control formula into it and use it as a feedforward controller to achieve real-time force control.

2. The rope force control method applicable to time-varying configuration according to claim 1, characterized in that: In step S1, the tension sensors on the auxiliary rope Bowden system and the power rope Bowden system are used, the proximal force is defined as the input force, and the distal force is defined as the output force. The parameter with the minimum error is defined as μ by formula (6). a and μ p .

3. The rope force control method applicable to time-varying configuration according to claim 1, characterized in that: In step S2, the auxiliary rope Bowden system sets the expected sinusoidal input, and the auxiliary rope Bowden system θ is calculated using formula (5). a .

4. The rope force control method applicable to time-varying configuration according to claim 1, characterized in that: In step S3, the desired force value at the far end of the power rope Bowden system is set, and the inverse control model F in =F out ·e -uλθ The corresponding expected input value can be obtained, and the angle value of the power rope Bowden system can be calculated in real time by the auxiliary rope Bowden system.

Images