High-precision force control method for industrial body robot end gripper

Abstract

This invention discloses a high-precision force control method for the end effector gripper of an industrial omnidirectional robot, belonging to the field of industrial omnidirectional robot control technology. This method employs a dual closed-loop control strategy of velocity loop and current loop to achieve precise control of sawtooth torque and suppress torque pulsation. A pre-compensation algorithm is introduced to feedforward compensate for system errors, friction torque, and delay interference, improving torque control accuracy to ±0.01N. A stroke adaptive algorithm is designed, which dynamically adjusts control parameters by real-time acquisition of stroke parameters, enabling the same control board to adaptively adapt to end effectors with different strokes. This invention significantly improves force control accuracy and stability, enhances the versatility of the control board, reduces equipment costs, and is suitable for various high-precision force control scenarios for end effectors of industrial omnidirectional robots, possessing significant engineering application value.

Description

Technical Field

[0001] This invention relates to the field of industrial automaton control technology, specifically to a high-precision force control method for the end effector gripper of an industrial automaton. Background Technology

[0002] Industrial robotic arms are core equipment in the field of intelligent manufacturing, widely used in precision assembly, micro-parts machining, and material handling due to their flexible operation capabilities. The end effector, as the actuator that directly contacts the workpiece, directly determines the quality and reliability of the operation through its force control accuracy. In high-end applications such as precision assembly and micro-parts gripping, the torque control accuracy of the end effector typically needs to reach ±0.01N, while effectively suppressing torque pulsations caused by sawtooth torque to avoid damage to fragile workpieces.

[0003] Existing force control methods for the end effector grippers of industrial robots mostly employ a single closed-loop control strategy, which makes it difficult to achieve precise control of the sawtooth torque. The torque pulsation is large, and the force control accuracy is generally above ±0.1N, which cannot meet the requirements of high-end operations. Although some solutions introduce compensation algorithms, they mostly use post-compensation, which has the problem of compensation lag. This makes it difficult to effectively offset the effects of system errors and frictional torque, thus limiting the improvement of force control accuracy.

[0004] In addition, most existing end effector control boards are dedicated designs. One control board is only compatible with a single stroke end effector. When changing to a different stroke end effector, the control board must be changed simultaneously, which increases equipment costs and reduces operational flexibility. It is difficult to adapt to the universal operation requirements of multi-specification and multi-stroke end effectors, thus restricting the promotion and application of industrial robots.

[0005] In existing related technologies, some solutions only focus on improving force control accuracy, without addressing precise control of sawtooth torque, and also cannot solve the problem of the control board adapting to different strokes; some solutions adopt adaptive control, but only adapt to load changes, without designing for gripper stroke changes, resulting in poor versatility. Therefore, developing a force control method that combines high-precision force control, precise control of sawtooth torque, and adaptive control board stroke has become an urgent need in the current field of industrial embodied robots. Summary of the Invention

[0006] This invention aims to overcome the shortcomings of existing technologies, such as low force control accuracy of end effector grippers in industrial robots, difficulty in adjusting sawtooth torque, and inability of control boards to adapt to different strokes. It provides a high-precision force control method for end effector grippers in industrial robots, which achieves precise control of sawtooth torque, improves force control accuracy to ±0.01N, and enables the same control board to adapt to grippers with different strokes, thereby reducing equipment costs and improving operational flexibility.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A high-precision force control method for the end effector gripper of an industrial embodied robot includes the following steps:

[0009] Step 1: Build an end effector force control system for an industrial robot. The force control system includes an end effector, a drive motor, a force sensor, and a control board. The end effector is a mechanical actuator. The drive motor is a coreless brushless DC motor. The force sensor is used to collect the actual output torque of the end effector. The control board is used for motor driving, control algorithm execution, and control command output.

[0010] Step 2: A dual closed-loop control strategy of speed loop and current loop is adopted to construct a sawtooth torque precision control mechanism. The current loop takes the three-phase current of the drive motor as the control object, collects the three-phase current signal in real time and compares it with the current setpoint, and outputs the current control signal through PID regulation to achieve precise control of the motor current. The speed loop takes the speed of the drive motor as the control object, collects the speed signal in real time and compares it with the speed setpoint, and uses the adjustment result as the current setpoint of the current loop to achieve coordinated control of speed and current, precisely control the sawtooth torque of the end gripper and suppress torque pulsation.

[0011] Step 3: Introduce a pre-compensation algorithm to improve torque control accuracy; collect the actual output torque of the end gripper in real time through a force sensor, and calculate the torque control error by combining it with the preset target torque; construct a pre-compensation model based on the mechanical characteristics of the end gripper, the friction torque model, and the system delay parameters; calculate the compensation amount based on the torque control error and the pre-compensation model, and inject it into the dual closed-loop control loop to compensate for interference factors such as system error and friction torque in advance, so that the torque control accuracy of the end gripper reaches ±0.01N.

[0012] Step 4: Design an adaptive algorithm to achieve adaptive adaptation of the same control board to different strokes; collect the stroke parameters of the end effector gripper, including stroke range, stroke speed, and stroke position; the adaptive algorithm analyzes and processes the stroke parameters, identifies the stroke type of the end effector gripper, and dynamically adjusts the output parameters of the control board, the adjustment coefficient of the dual closed-loop control, and the compensation parameters of the pre-compensation algorithm; the control board outputs a control signal adapted to the current stroke based on the adjusted parameters, drives the motor to drive the end effector gripper to move, and realizes adaptive force control of the end effector gripper with different strokes by the same control board.

[0013] Furthermore, in step 2, the speed loop adjustment uses an incremental PID algorithm, and the adjustment formula is as follows:

[0014] Δu(k)=K p *[e(k)-e(k-1)]+K i *e(k)+K d *[e(k)-2e(k-1)+e(k-2)];

[0015] Where Δu(k) is the speed loop output increment, K p K is the proportionality coefficient. i K is the integral coefficient. d Here are the differential coefficients, e(k) is the current speed error, e(k-1) is the speed error at the previous moment, and e(k-2) is the speed error between the two previous moments.

[0016] The current loop regulation uses a positional PID algorithm, and the regulation formula is as follows:

[0017] u(k)=K p *e i (k)+K i *Σe i (j)+K d *[e i (k)-e i (k-1)];

[0018] Where u(k) is the current loop output control signal, e i (k) represents the current current error, e i (k-1) represents the current error at the previous moment, and j represents the integration number.

[0019] Furthermore, in step 3, the calculation process for the compensation amount of the pre-compensation algorithm is as follows:

[0020] First, establish the friction torque model of the end gripper:

[0021] f(ω) = f_c·ω + f_v·ω;

[0022] Where f(ω) is the frictional torque, f_c is the Coulomb frictional torque, f_v is the viscous friction coefficient, and ω is the angular velocity of the drive motor.

[0023] Secondly, the data acquisition system delay time Based on the torque control error e_t(k), friction torque f(ω), and system delay time Calculate the pre-compensation amount .

[0024] in This is the delay compensation coefficient. This represents the torque control error at the delayed moment.

[0025] The compensation amount ΔT is injected into the current loop current setpoint to complete the pre-compensation.

[0026] Furthermore, in step 4, the adaptive algorithm includes a travel identification module, a parameter adjustment module, and an adaptation execution module. The specific execution process is as follows:

[0027] 4.1 The stroke recognition module collects the stroke data of the end gripper, uses a sliding window filtering algorithm to denoise the stroke data, and extracts three feature parameters: stroke range L, stroke speed v, and stroke position x.

[0028] 4.2 The parameter adjustment module identifies the stroke type based on the extracted feature parameters, establishes a mapping relationship between the stroke feature parameters and the control board output parameters, dual closed-loop adjustment coefficients, and pre-compensation parameters, and dynamically adjusts the control board PWM output frequency, current output range, speed loop and current loop PID parameters, and pre-compensation algorithm compensation coefficients.

[0029] 4.3 The adaptation execution module sends the adjusted parameters to the control board. The control board outputs control signals based on the adjusted parameters, and drives the motor to move the end gripper according to the target torque and the current stroke, thereby realizing adaptive force control.

[0030] Furthermore, in step 1, the force sensor is a high-precision tensile and compressive force sensor with a measurement accuracy of not less than ±0.001N and a sampling frequency of not less than 1kHz.

[0031] Furthermore, in step 2, the sawtooth torque adjustment accuracy is not less than ±0.005N, and the torque pulsation amplitude is controlled within 5%.

[0032] Furthermore, the adaptive algorithm supports an end gripper stroke range of 10mm–50mm and a stroke speed adjustment range of 10mm / s–100mm / s.

[0033] Furthermore, the control board uses a DSP control chip with an operating frequency of no less than 100MHz to ensure the real-time performance of the control algorithm.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] A dual closed-loop control strategy of speed loop and current loop is adopted. The current loop controls the current precisely and the speed loop regulates the speed in a coordinated manner, so as to achieve precise control of sawtooth torque, effectively suppress torque pulsation, and the sawtooth torque control accuracy is not less than ±0.005N, and the torque pulsation amplitude is ≤5%. The control stability and accuracy are significantly better than the single closed-loop scheme.

[0036] By introducing a pre-compensation algorithm, combined with the friction torque model and system delay compensation, interference such as system errors and friction torque are compensated in advance, avoiding the problem of post-compensation lag, and improving the torque control accuracy to ±0.01N, which meets the requirements of high-end scenarios such as precision assembly and micro-part gripping.

[0037] The design incorporates a stroke adaptive algorithm. By collecting stroke parameters in real time and dynamically adjusting control parameters, the same control board can adaptively adapt to grippers with different stroke ends. This eliminates the need to replace the control board, meeting the force control requirements of grippers of various specifications. It solves the problem of poor control board versatility, reduces equipment costs, and improves operational flexibility.

[0038] The method is highly adaptable, supporting strokes of 10mm-50mm and speed adjustments of 10mm / s-100mm / s. It is suitable for high-precision force control scenarios of end effectors in various industrial robots. The structure is simple and easy to implement, requiring no large-scale modification of existing robot systems, and has good engineering application value and promotion prospects. Attached Figure Description

[0039] Figure 1 This is the structural frame of the high-precision force control system for the end effector gripper of the industrial embodied robot of the present invention;

[0040] Figure 2 This is a block diagram illustrating the principle of the dual closed-loop control strategy of this invention.

[0041] Figure 3 This is a schematic diagram of the pre-compensation algorithm of the present invention;

[0042] Figure 4 This is a schematic diagram of the adaptive algorithm flow of the present invention.

[0043] Figure Labels

[0044] 1-End gripper; 2-Drive motor; 3-Control board; 4-Force sensor. Detailed Implementation

[0045] The present invention will be further described in detail below with reference to specific embodiments, so that those skilled in the art can understand it.

[0046] Example 1

[0047] A high-precision force control method for the end effector gripper of an industrial embodied robot includes the following steps:

[0048] Step 1: Construct the force control system for the end effector gripper of the industrial robot. The system includes the end effector gripper 1, drive motor 2, force sensor 4, and control board 3. The drive motor is a coreless brushless DC motor with a rated speed of 3000 r / min and a rated torque of 0.5 N·m. The force sensor 4 is a high-precision tensile / compression sensor with a measurement accuracy of ±0.001 N and a sampling frequency of 1 kHz. The control board 3 uses a DSP chip with an operating frequency of 100 MHz and a PWM output frequency of 0–100 kHz. The force sensor 4 is installed at the connection between the end effector gripper 1 and the drive motor 2 to collect the actual output torque of the end effector gripper. The control board 3 is electrically connected to the motor and is used for motor drive, algorithm execution, and command output.

[0049] Step 2: Construct a sawtooth torque regulation mechanism using a dual closed-loop control strategy of speed loop and current loop.

[0050] The current loop uses the three-phase current of the motor as the control object, with a sampling period of 0.1ms. It acquires the three-phase current signal in real time and compares it with the current setpoint. It uses a positional PID algorithm to adjust the output current control signal, and the PID parameter is set to K. p =0.5, K i =0.1, K d =0.05.

[0051] The speed loop uses motor speed as the control object. The speed signal is acquired via an encoder with a sampling period of 0.1ms. After comparison with the speed setpoint, an incremental PID algorithm is used for adjustment. The adjustment result is used as the current setpoint for the current loop. The PID parameter is set to K. p =0.3, K i =0.08, K d =0.02.

[0052] With dual closed-loop control, the sawtooth torque regulation accuracy reaches ±0.005N, and the torque pulsation amplitude is ≤5%.

[0053] Step 3: Introduce a pre-compensation algorithm to improve torque control accuracy. The force sensor collects the actual output torque of the end gripper in real time, and calculates the torque control error e_t(k) = target torque − actual output torque based on a preset target torque of 0.5N; establish the friction torque model f(ω) = f_c·ω + f_v·ω, where f_c = 0.002N·m, f_v = 0.001N·s / rad; and calculate the system delay time. =0.2ms, delay compensation coefficient =0.1, calculate the pre-compensation amount ΔT=e_t(k)+f(ω)+0.1·e_t(k-0.2); inject the compensation amount into the current loop current setpoint, and the torque control accuracy reaches ±0.01N after compensation.

[0054] Step 4: Use an adaptive algorithm to achieve adaptive control board travel.

[0055] 4.1 The stroke recognition module collects stroke data through displacement sensors, filters and denoises through a sliding window, and extracts the stroke range L, stroke speed v, and stroke position x.

[0056] 4.2 The parameter adjustment module identifies the stroke type and establishes parameter mapping based on stroke characteristics: For short strokes (L=30mm, v=0.5mm / s), the PWM output frequency is 50kHz, the current output range is 0.1-1A, and the speed loop K... p =0.35, K i =0.1, K d=0.03, current loop Kp=0.55, K i =0.12、K d =0.06, compensation coefficient Kτ=0.12; for long stroke (L=100mm, v=5mm / s), the PWM output frequency is 30kHz, the current output range is 0.2–1.5A, and the speed loop K p =0.25, K i =0.06, K d =0.01, current loop K p =0.45, K i =0.08, Kd=0.04, compensation coefficient Kτ=0.08.

[0057] 4.3 The adaptation execution module sends the parameters to the control board, and the control board outputs the adaptation signal to drive the motor to move, thereby realizing adaptive force control.

[0058] In this embodiment, the same control board is adapted to end grippers with strokes of 30mm and 100mm respectively, both achieving torque control accuracy of ±0.01N, sawtooth torque pulsation amplitude ≤5%, and good adaptive effect, effectively solving the technical problem that the control board cannot adapt to different strokes.

[0059] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the concept and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A high-precision force control method for the end effector gripper of an industrial embodied robot, characterized in that, Includes the following steps: Step 1: Build an end effector force control system for an industrial robot. The force control system includes an end effector, a drive motor, a force sensor, and a control board. The end effector is a mechanical actuator. The drive motor is a coreless brushless DC motor. The force sensor is used to collect the actual output torque of the end effector. The control board is used for motor driving, algorithm execution, and control command output. Step 2: A dual closed-loop control strategy of speed loop and current loop is adopted to construct a sawtooth torque precision control mechanism. The current loop takes the three-phase current of the drive motor as the control object, collects the three-phase current signal in real time and compares it with the current setpoint, and outputs the current control signal through PID regulation to achieve precise control of the motor current. The speed loop takes the speed of the drive motor as the control object, collects the speed signal in real time and compares it with the speed setpoint, and uses the adjustment result as the current setpoint of the current loop to achieve coordinated control of speed and current, precisely control the sawtooth torque of the end gripper and suppress torque pulsation. Step 3: Introduce a pre-compensation algorithm to improve torque control accuracy; The actual output torque of the end gripper is collected in real time by a force sensor, and the torque control error is calculated by combining it with the preset target torque. Based on the mechanical characteristics of the end gripper, the friction torque model, and the system delay parameters, a pre-compensation model is constructed. The compensation amount is calculated based on the torque control error and the pre-compensation model and injected into the dual closed-loop control loop to compensate for interference factors such as system error and friction torque in advance, so that the torque control accuracy of the end gripper reaches ±0.01N. Step 4: Design an adaptive algorithm to achieve adaptive adaptation of the same control board to different strokes; collect the stroke parameters of the end gripper, including stroke range, stroke speed and stroke position; The adaptive algorithm analyzes and processes the stroke parameters, identifies the stroke type of the end gripper, and dynamically adjusts the output parameters of the control board, the adjustment coefficient of the dual closed-loop control, and the compensation parameters of the pre-compensation algorithm. The control board outputs a control signal adapted to the current stroke based on the adjusted parameters, drives the motor to move the end gripper, and realizes adaptive force control of the end gripper with different strokes by the same control board.

2. The method according to claim 1, characterized in that, In step 2, the speed loop adjustment uses an incremental PID algorithm, and the adjustment formula is as follows: Δu(k) = K p *[e(k) - e(k-1)] + K i *e(k) + K d *[e(k) - 2e(k-1) + e(k-2)]; where Δu(k) is the speed loop output increment, K p is the proportional coefficient, K i is the integral coefficient, K d is the differential coefficient, e(k) is the current speed error, e(k-1) is the last time speed error, and e(k-2) is the speed error two times ago. The current loop regulation uses a positional PID algorithm, and the regulation formula is as follows: u(k)=K p *and i (k)+K i *Σe i (j)+K d *[and i (k)-e i (k-1)]; Where u(k) is the current loop output control signal, e i (k) represents the current current error, e i (k-1) represents the current error at the previous moment, and j represents the integration number.

3. The method according to claim 1, characterized in that, In step 3, the calculation process for the compensation amount in the pre-compensation algorithm is as follows: Establish the friction torque model of the end gripper: f(ω) = f_c*ω + f_v*ω, where f(ω) is the friction torque, f_c is the Coulomb friction torque, f_v is the viscous friction coefficient, and ω is the angular velocity of the drive motor. Data Acquisition System Delay Time Based on the torque control error e_t(k), friction torque f(ω), and system delay time Calculate the pre-compensation amount ,in This is the delay compensation coefficient. To compensate for the torque control error at the delayed moment, the compensation amount ΔT is injected into the current loop current setpoint to complete the pre-compensation.

4. The method according to claim 1, characterized in that, In step 4, the adaptive algorithm includes a travel recognition module, a parameter adjustment module, and an adaptation execution module. The specific execution process is as follows: 4.1 The stroke recognition module collects the stroke data of the end gripper, uses a sliding window filtering algorithm to denoise the stroke data, and extracts three feature parameters: stroke range L, stroke speed v, and stroke position x. 4.2 The parameter adjustment module identifies the stroke type based on the extracted feature parameters, establishes a mapping relationship between the stroke feature parameters and the control board output parameters, dual closed-loop adjustment coefficients, and pre-compensation parameters, and dynamically adjusts the control board PWM output frequency, current output range, speed loop and current loop PID parameters, and pre-compensation algorithm compensation coefficients. 4.3 The adaptation execution module sends the adjusted parameters to the control board. The control board outputs control signals based on the adjusted parameters, and drives the motor to move the end gripper according to the target torque and the current stroke, thereby realizing adaptive force control.

5. The method according to claim 1, characterized in that, In step 1, the force sensor is a high-precision tensile and compressive force sensor with a measurement accuracy of not less than ±0.001N and a sampling frequency of not less than 1kHz.

6. The method according to claim 1, characterized in that, In step 2, the sawtooth torque adjustment accuracy is no less than ±0.005N, and the torque pulsation amplitude is controlled within 5%.

7. The method according to claim 1, characterized in that, The adaptive algorithm supports a stroke range of 10mm–50mm for the end gripper and a stroke speed adjustment range of 10mm / s–100mm / s.

8. The method according to claim 1, characterized in that, The control board uses a DSP control chip with an operating frequency of no less than 100MHz to ensure the real-time performance of the control algorithm.

Images