A non-overshoot adaptive joint control method and system for a gynecological rehabilitation robot
By improving the particle swarm optimization algorithm to tune the PID controller parameters online, the problem of overshoot in obstetric and gynecological rehabilitation robots was solved, achieving overshoot-free and fast-response joint control, thus improving the safety and applicability of the rehabilitation robot.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NORTHEAST FORESTRY UNIV
- Filing Date
- 2026-04-30
- Publication Date
- 2026-07-24
AI Technical Summary
Existing obstetric and gynecological rehabilitation robot controllers are prone to overshoot and have insufficient dynamic performance, failing to meet the safety and comfort requirements of postpartum patients.
An improved particle swarm optimization algorithm (TDE-PSO) is used to tune the PID controller parameters online. Combined with Logistic chaotic mapping, adaptive inertial weights and elite time delay feedback mechanism, it ensures no positive overshoot and good dynamic response.
It achieves overshoot-free control of joint movements in obstetric and gynecological rehabilitation robots, significantly improving safety and dynamic response speed, adapting to individual differences among patients, and is suitable for clinical application.
Smart Images
Figure CN122442633A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a joint control method and system, and particularly to an overshoot-free adaptive joint control method and system for obstetric and gynecological rehabilitation robots. Background Technology
[0002] Rehabilitation robots are widely used in neurorehabilitation and postoperative training. Their core function is to drive repetitive movements of the patient's limbs via motors to promote neural remodeling and muscle strength recovery. During this process, the smoothness and shock-free nature of joint movement are crucial. If the controller overshoots, even a displacement of only 2–3 mm, it can lead to muscle strain, joint discomfort, or even secondary injury, seriously affecting the safety of rehabilitation.
[0003] In the field of postpartum rehabilitation in obstetrics and gynecology, the functional recovery of the pelvis, sacroiliac joints, and lower limb joints is a core component of rehabilitation. Rehabilitation robots are increasingly being used in neurological rehabilitation, post-operative training in obstetrics and gynecology, and postpartum pelvic floor and pelvic function rehabilitation. Their core function is to drive the patient's limbs and joints through motors to perform standardized, repetitive rehabilitation movements, promoting muscle function recovery, joint stability reconstruction, and postpartum physiological structural repair. Due to the fragility of postpartum tissues, the smoothness and shock-free nature of joint movements are crucial during rehabilitation.
[0004] Currently, most rehabilitation robots use traditional PID controllers, whose parameter tuning mainly relies on the Ziegler-Nichols critical proportional method, engineer-based trial-and-error methods, or standard intelligent optimization algorithms (such as PSO and GA). The Ziegler-Nichols method requires the system to be in a critical oscillation state, which is dangerous to operate and unsuitable for postpartum women with sensitive constitutions and fragile tissues. The trial-and-error method is time-consuming and inconsistent, making it difficult to adapt to individual differences among postpartum women. Although standard intelligent algorithms such as PSO can automatically find the optimal solution, they suffer from premature convergence and are easily affected by initial values, and are still prone to positive overshoot in practical applications, failing to meet the high safety requirements of obstetric and gynecological rehabilitation. In addition, existing research often uses comprehensive performance indicators such as ITAE and ISE as optimization targets, without making "no overshoot" a hard constraint, which leads to the algorithm sacrificing motion safety in pursuit of fast response.
[0005] In such systems, real-time signals such as joint position, angle, and torque collected by sensors serve as the core feedback basis for the entire control system. These signals not only provide error input for the PID controller, enabling closed-loop control, but also form the basis for calculating performance indicators such as ITAE in the improved PSO optimization algorithm, directly determining the joint motion accuracy, stability, and safety. Therefore, there is an urgent need for an adaptive control method that can automatically tune PID parameters, ensure no positive overshoot, and possess good dynamic response performance to improve the clinical applicability and patient experience of obstetric and gynecological rehabilitation robots.
[0006] This invention belongs to the interdisciplinary field of intelligent rehabilitation equipment and automatic control technology, specifically involving a PID control parameter self-tuning method based on an improved particle swarm optimization (TDE-PSO) algorithm. It is applicable to joint servo control systems in medical assistive devices with stringent requirements for motion safety and comfort, such as obstetric and gynecological rehabilitation robots, lower limb exoskeletons, and postpartum pelvic rehabilitation training equipment. The purpose of this invention is to provide an overshoot-free adaptive joint control method and system for obstetric and gynecological rehabilitation robots, achieving safe, fast, and robust joint motion control. Summary of the Invention
[0007] The purpose of this invention is to solve the problems of PID parameter tuning relying on manual intervention, easy overshoot, and insufficient dynamic performance in the prior art, and to propose an overshoot-free adaptive joint control method for obstetric and gynecological rehabilitation robots.
[0008] The above objectives are achieved through the following technical solutions: An overshoot-free adaptive joint control method for an obstetric and gynecological rehabilitation robot, the method being implemented through the following steps: Step 1: Obtain the angle feedback signal of the rehabilitation robot joints in real time using an angle sensor; Step 2: Online tuning of the PID controller parameters based on the improved particle swarm optimization algorithm, including: proportional K p Integral K i Differential K d The process of obtaining the optimal PID parameters includes: Step 21: Establish a joint dynamics model for the rehabilitation robot to comprehensively consider the dynamic characteristics of the joint motor rotor inertia, limb load mass, and transmission system damping. The dynamic characteristics of the joints of the rehabilitation robot are constructed as a second-order transfer function, expressed as: In the formula, s represents the complex frequency variable in the Laplace transform; Step 22: When initializing the particle swarm, the initial positions of the particles are generated using a Logistic chaotic mapping. Steps 2 and 3: Design adaptive inertia weight w: The inertia weight w is an inertia weight that decreases non-linearly with the number of iterations; Step 24: Introduce elite time delay feedback: Every preset number of iterations, apply a small perturbation to the current global best particle and re-evaluate its fitness value; Step 25: Use the time-weighted absolute error integral ITAE as the optimization objective function J; The objective function J is expressed as: Where J is the optimization objective function and also the ITAE performance metric. t It is a time variable. e ( t ) represents the deviation between the set angle and the actual angle of the joint, t _f This is the simulation termination time; Step 26: Output the optimal PID parameters: proportional K p Integral K i Differential K d ; Step 3: Apply the optimized PID parameters obtained from the tuning to the PID controller to generate control commands. The motor drive module receives the control commands and drives the joint motor to operate. The joint motor causes the joint to move, thereby achieving motion control of the joint without positive overshoot.
[0009] Furthermore, the small perturbation mentioned in step two is a random number that follows a Gaussian distribution with a mean of 0 and a standard deviation of 0.01.
[0010] Furthermore, the improved particle swarm optimization algorithm method also includes the step of setting the parameter search boundary of the PID controller: Ratio K p The search range is 0.5 ≤ K. p ≤2.0; integral K i The search range is 0.01 ≤ K i ≤ 0.4; Differential K d The search range is 0 ≤ K d ≤ 0.1.
[0011] The purpose of this invention is to solve the problems of PID parameter tuning relying on manual intervention in the prior art, which is prone to overshoot and insufficient dynamic performance, and to propose an overshoot-free adaptive joint control system for obstetric and gynecological rehabilitation robots.
[0012] The above objectives are achieved through the following technical solutions: An overshoot-free adaptive joint control system for an obstetric and gynecological rehabilitation robot includes: Angle sensor is used to acquire angle feedback signals of the joints of the rehabilitation robot in real time; The motor drive module is used to receive control commands generated by the PID controller and drive the joint motor to operate. The joint motor makes the joint move, realizing motion control of the joint without positive overshoot. A PID controller, which has an embedded computer program configured to perform the overshoot-free adaptive joint control method as described in any one of claims 1-3; A host computer or embedded processor is used to run the improved particle swarm optimization algorithm and output the optimal PID parameters; The improved particle swarm optimization algorithm includes: A joint dynamics model of the rehabilitation robot was established to comprehensively consider the dynamic characteristics of the joint motor rotor inertia, limb load mass, and transmission system damping. The dynamic characteristics of the joints of the rehabilitation robot are constructed as a second-order transfer function, expressed as: In the formula, s represents the complex frequency variable in the Laplace transform; Then, when initializing the particle swarm, the initial positions of the particles are generated using a Logistic chaotic mapping. Next, an adaptive inertia weight w is designed: the inertia weight w is an inertia weight that decreases non-linearly with the number of iterations; Then, an elite time delay feedback mechanism is introduced: every preset number of iterations, a small perturbation is applied to the current global best particle and its fitness value is re-evaluated; The time-weighted absolute error integral is then optimized using ITAE as the objective function J. The objective function J is expressed as: Where J is the optimization objective function and also the ITAE performance metric. t It is a time variable. e ( t ) represents the deviation between the set angle and the actual angle of the joint, t _f This is the simulation termination time; Then output the optimal PID parameter: proportional K p Integral K i Differential K d .
[0013] Furthermore, the small perturbation is a random number following a Gaussian distribution with a mean of 0 and a standard deviation of 0.01; the improved particle swarm optimization algorithm also includes the step of setting the parameter search boundary of the PID controller: The ratio K p The search range is 0.5 ≤ K. p ≤2.0; integral K i The search range is 0.01 ≤ K i ≤ 0.4; Differential K d The search range is 0 ≤ Kd ≤ 0.1.
[0014] Furthermore, the angle sensor is a rotary encoder or an inertial measurement unit (IMU).
[0015] Furthermore, the PID controller is deployed in an ARM Cortex-M series microcontroller or an industrial control computer.
[0016] Furthermore, the small perturbation is a random number that follows a Gaussian distribution with a mean of 0 and a standard deviation of 0.01.
[0017] An apparatus for an overshoot-free adaptive joint control method for an obstetric and gynecological rehabilitation robot includes: processor; A memory in which executable instructions of the processor are stored; The processor is configured to perform the steps of the overshoot-free adaptive joint control method for an obstetric and gynecological rehabilitation robot by executing the executable instructions.
[0018] A computer-readable storage medium for storing a program, which, when executed, implements the steps of the described method for an overshoot-free adaptive joint control for an obstetric and gynecological rehabilitation robot.
[0019] The beneficial effects of this invention are as follows: To achieve the above objectives, this invention proposes an improved particle swarm optimization algorithm (TDE-PSO algorithm) that integrates chaotic initialization, adaptive inertial weighting, and elite time delay feedback mechanisms for online optimization of PID controller parameters for rehabilitation robot joints. Specifically, this is reflected in the following aspects: 1. Achieving zero positive overshoot control significantly improves the motion safety of obstetric and gynecological rehabilitation robots: Experimental verification shows that the PID controller parameters tuned using the TDE-PSO algorithm of this invention can achieve a step response overshoot of -2.00% in the joint motion system of obstetric and gynecological rehabilitation robots, which is in a state of slight undershoot or even zero overshoot. This can effectively avoid muscle strain, joint discomfort and secondary injury in postpartum patients caused by positive overshoot, which is just right for the rehabilitation characteristics of postpartum patients with weak physical condition, and thus significantly improves the motion control safety of obstetric and gynecological rehabilitation robots.
[0020] 2. Fast dynamic response speed, highlighting the superiority of the control method of the present invention: Under the condition of 2% error band, the joint adjustment time of the control method involved in the present invention is only 2.31 seconds, which shows excellent dynamic response performance. Compared with traditional control methods, it can quickly adapt to the joint movement requirements of obstetric and gynecological rehabilitation robots and improve the efficiency of rehabilitation training. This also fully demonstrates that the overshoot-free adaptive joint control method of the present invention for obstetric and gynecological rehabilitation robots is superior to traditional control methods.
[0021] 3. Automatic suppression of derivative terms to enhance system industrial robustness: The overshoot-free adaptive joint control method for obstetric and gynecological rehabilitation robots in this invention can automatically optimize the derivative term parameter Kd of the PID controller to 0.0000 through the TDE-PSO algorithm. This effectively avoids the derivative term amplifying noise in the sensor-acquired signal, reduces the impact of noise on the joint control accuracy, thereby enhancing the industrial robustness of the joint control system of the obstetric and gynecological rehabilitation robot and ensuring stable operation of the system in complex rehabilitation scenarios.
[0022] 4. No manual intervention required, adaptable to individual differences among postpartum patients: The TDE-PSO improved particle swarm optimization algorithm used in this invention can realize online automatic tuning of PID controller parameters without human intervention throughout the process. It can also adapt to changes in limb load of different postpartum patients in real time, which effectively solves the problem that traditional parameter tuning methods are difficult to adapt to individual differences among patients, and further improves the practicality and clinical adaptability of obstetric and gynecological rehabilitation robots.
[0023] 5. Highly deployable and adaptable to real-world clinical applications: The TDE-PSO algorithm proposed in this invention has low computational complexity and does not require complex hardware support. It can be flexibly deployed on ARM Cortex-M series MCUs or host computers and run in real time. It can meet the design requirements of miniaturized and portable obstetric and gynecological rehabilitation robots and is also convenient for promotion and application in clinical rehabilitation institutions, community hospitals and home rehabilitation scenarios, and has good engineering application value. Attached Figure Description
[0024] Figure 1 This invention relates to a block diagram of a non-overshoot adaptive joint control system for an obstetric and gynecological rehabilitation robot; Figure 2 This invention relates to the process of an improved particle swarm optimization algorithm, demonstrating the complete iterative logic from initialization to outputting optimal parameters; Figure 3 This is a diagram showing the convergence curve of the system after optimization using the TDE-PSO algorithm. Figure 4 This is a graph showing the final response curve of the system after optimization using the TDE-PSO algorithm; Figure 5It is the evolution trajectory of PID parameters during the TDE-PSO algorithm optimization process; Figure 6 This is a diagram illustrating the convergence curve of the system after standardization (PSO). Figure 7 This is a diagram of the system response curve after standardization (PSO). Figure 8 This is a comparison diagram of the step responses of the standard PSO, Ziegler-Nichols method, and the TDE-PSO algorithm of this invention; Figure 1 In the diagram, 1 is the error summation node module, 2 is the PID controller, 3 is the second-order transfer function module, 4 is the feedback loop, and 5 is the target signal input terminal. Detailed Implementation
[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Specific implementation method one:
[0027] An overshoot-free adaptive joint control method for obstetric and gynecological rehabilitation robots, such as Figure 2 As shown, the method is implemented through the following steps: Step 1: Obtain the angle feedback signal of the rehabilitation robot joints in real time using an angle sensor; Step 2: Online tuning of the PID controller parameters based on the improved particle swarm optimization algorithm, including: proportional K p Integral K i Differential K d The process of obtaining the optimal PID parameters includes: Step 21: Establish a joint dynamics model for the rehabilitation robot to comprehensively consider the dynamic characteristics of the joint motor rotor inertia, limb load mass, and transmission system damping. The dynamic characteristics of the joints of the rehabilitation robot are constructed as a second-order transfer function, expressed as: In the formula, s represents the complex frequency variable in the Laplace transform. The primary objective of this invention is not to innovate in modeling; the standard normalized second-order model is sufficient to support the implementation of this scheme and conforms to conventional practices in control engineering. Specific physical parameters can be obtained through system identification or calibration and used for actual controller gain tuning without affecting the feasibility of this invention.
[0028] Step 22: When initializing the particle swarm, a Logistic chaotic mapping is used to generate the initial positions of the particles in order to enhance the diversity of the population and avoid getting trapped in local optima. Steps 2 and 3: Design an adaptive inertia weight w to balance global search and local exploitation capabilities: the inertia weight w is an inertia weight that decreases non-linearly with the number of iterations; Step 24: Introduce an elite time delay feedback mechanism to simulate the "delayed verification" process and prevent the algorithm from converging too early: Every preset number of iterations, apply a small perturbation to the current global best particle and re-evaluate its fitness value. Step 25: Use ITAE (time-weighted absolute error integral) as the optimization objective function J; The objective function J is expressed as: Where J is the optimization objective function and also the ITAE performance metric. t It is a time variable. e ( t ) represents the deviation between the set angle and the actual angle of the joint, t _f This is the simulation termination time; Step 26: Output the optimal PID parameters: proportional K p Integral K i Differential K d ; Step 3: After the algorithm terminates, the optimal PID parameters obtained from the tuning are applied to the PID controller to generate control commands. The motor drive module receives the control commands and drives the joint motor to operate. The joint motor causes the joint to move, thereby achieving motion control of the joint without positive overshoot.
[0029] The joint control method described in this invention can ensure that the overshoot of the joint step response does not exceed ±2%, and the settling time (2% error band) is less than 2.5 seconds. Specific Implementation Method Two:
[0030] This embodiment of the method for overshoot-free adaptive joint control of an obstetric rehabilitation robot differs from the first embodiment in that the small perturbation in step two is a random number that follows a Gaussian distribution with a mean of 0 and a standard deviation of 0.01. Specific implementation method three:
[0031] This embodiment of the adaptive joint control method for an obstetric and gynecological rehabilitation robot without overshoot differs from specific embodiments one or two in that the improved particle swarm optimization algorithm method further includes a step of setting the parameter search boundary of the PID controller. This ensures that the optimization algorithm searches for optimal parameters only within a set "reasonable range," avoiding "blind search," "divergence," or the acquisition of an unstable controller. It also prevents the algorithm from deviating from preset rules or objectives. Ratio K p The search range is 0.5 ≤ K. p ≤2.0; integral K i The search range is 0.01 ≤ K i ≤ 0.4; Differential K d The search range is 0 ≤ K d ≤ 0.1. Specific implementation method four:
[0032] This embodiment provides a non-overshoot adaptive joint control system for an obstetric and gynecological rehabilitation robot, the control system comprising: Angle sensor is used to acquire angle feedback signals of the joints of the rehabilitation robot in real time; The motor drive module is used to receive control commands generated by the PID controller and drive the joint motor to operate. The joint motor makes the joint move, realizing motion control of the joint without positive overshoot. A PID controller, which has an embedded computer program configured to perform the overshoot-free adaptive joint control method as described in any one of claims 1-3; A host computer or embedded processor is used to run the improved particle swarm optimization algorithm and output the optimal PID parameters, enabling algorithm deployment and human-computer interaction. The improved particle swarm optimization algorithm includes, Improve particle swarm optimization systems, such as Figure 1 As shown, it includes: Error summation node module 1 is used to calculate the difference between the actual angle signal and the target angle command, and output the error signal to the controller. The TDE-PSO module (not shown in the figure) is used to execute the improved particle swarm optimization algorithm and tune the optimal parameters for PID controller 2 in the background. PID controller 2 is used to output a control quantity based on the error signal to realize closed-loop regulation of joint motion; The second-order transfer function module 3 is located at the joint and is part of the motor drive module. The second-order transfer function module 3 is used to output the actual joint angle. The angle sensor (not shown in the figure) acquires the actual joint angle feedback signal of the rehabilitation robot in real time and transmits it to the error summation node module 1 at the input end through the feedback loop 4. Target signal input terminal 5 is used to input the target signal (step command) and give the system the angle to be reached; In the second-order transfer function module, a joint dynamics model of the rehabilitation robot is established at 3 locations to comprehensively consider the dynamic characteristics of the joint motor rotor inertia, limb load mass, and transmission system damping. The dynamic characteristics of the joints of the rehabilitation robot are constructed as a second-order transfer function, expressed as: In the formula, s represents the complex frequency variable in the Laplace transform. The primary objective of this invention is not to innovate in modeling; the standard normalized second-order model is sufficient to support the implementation of this scheme and conforms to conventional practices in control engineering. Specific physical parameters can be obtained through system identification or calibration and used for actual controller gain tuning without affecting the feasibility of this invention.
[0033] Subsequently, when initializing the particle swarm, a Logistic chaotic mapping is used to generate the initial positions of the particles in order to enhance population diversity and avoid getting trapped in local optima. Next, an adaptive inertia weight w is designed to balance global search and local exploitation capabilities: the inertia weight w is an inertia weight that decreases non-linearly with the number of iterations; Then, an elite time delay feedback mechanism is introduced to simulate the "delayed verification" process and prevent the algorithm from converging too early: every preset number of iterations, a small perturbation is applied to the current global best particle and its fitness value is re-evaluated. Then, ITAE (time-weighted absolute error integral) was used as the optimization objective function J; The objective function J is expressed as: Where J is the optimization objective function and also the ITAE performance metric. t It is a time variable. e ( t ) represents the deviation between the set angle and the actual angle of the joint, t _f This is the simulation termination time; Then output the optimal PID parameter: proportional K p Integral K i Differential K d . Specific implementation method five:
[0034] The overshoot-free adaptive joint control system for an obstetric and gynecological rehabilitation robot in this embodiment differs from the fourth embodiment in that the small perturbation is a random number that follows a Gaussian distribution with a mean of 0 and a standard deviation of 0.01. The improved particle swarm optimization algorithm also includes a step of setting the parameter search boundary for the PID controller, ensuring that the optimization algorithm searches for optimal parameters only within a set "reasonable range," avoiding "blind search," "divergence," or the acquisition of an unstable controller. It also prevents the algorithm from deviating from preset rules or objectives. The ratio K p The search range is 0.5 ≤ K. p ≤2.0; integral K i The search range is 0.01 ≤ K i≤ 0.4; Differential K d The search range is 0 ≤ K d ≤ 0.1. Specific implementation method six:
[0035] This embodiment of the adaptive joint control system for obstetric and gynecological rehabilitation robots without overshoot differs from Specific Embodiment Five in that the angle sensor is a rotary encoder or an inertial measurement unit (IMU). Specific implementation method seven:
[0036] This embodiment of the invention provides an overshoot-free adaptive joint control system for an obstetric and gynecological rehabilitation robot. Unlike embodiment six, the PID controller is deployed in an ARM Cortex-M series microcontroller or an industrial control computer. Detailed implementation method eight:
[0037] This embodiment of the adaptive joint control system for obstetric and gynecological rehabilitation robots without overshoot differs from specific embodiment seven in that the small perturbation is a random number that follows a Gaussian distribution with a mean of 0 and a standard deviation of 0.01. Specific implementation method nine:
[0038] This embodiment of the device for an overshoot-free adaptive joint control method for an obstetric and gynecological rehabilitation robot includes: processor; A memory in which executable instructions of the processor are stored; The processor is configured to perform the steps of the overshoot-free adaptive joint control method for an obstetric rehabilitation robot according to any one of claims 1 to 3 by executing the executable instructions. Specific Implementation Method Ten:
[0039] This embodiment provides a computer-readable storage medium for storing a program that, when executed, implements the steps of the method according to any one of claims 1 to 3.
[0040] The overshoot-free adaptive joint control method of the present invention for obstetric and gynecological rehabilitation robots is applied to a rehabilitation robot for hip joint rehabilitation of obstetric and gynecological patients. The rehabilitation robot performs the following steps at the hip joint: System modeling: Equivalent hip joint dynamics to ; Algorithm parameter settings: Number of particles: 30; Maximum number of iterations: 50; Search scope: K p ∈[0.5, 3], K i∈[0.01, 0.8], K d ∈[0, 1].
[0041] An overshoot-free adaptive joint control method (TDE-PSO algorithm) for obstetric and gynecological rehabilitation robots was run to obtain the optimal parameters: K p =4.24, K i =0.89, K d =2.1 Performance verification: ITAE = 3.660 Overshoot = -0.04% Adjustment time (2%) = 1.59 seconds Comparative experiment: Ziegler-Nichols method: 0% overshoot, ITAE = 8.78 Standard PSO: 1% overshoot, ITAE = 0.4363, settling time 12 seconds It is evident that the method of this invention significantly outperforms existing methods in both safety and dynamic performance. This controller can be integrated into the main control system of a rehabilitation robot, automatically performing parameter tuning before each training session to ensure personalized and safe control.
[0042] This invention initializes the particle swarm using Logistic chaotic mapping, designs nonlinear decreasing inertial weights, and introduces an elite particle perturbation mechanism to tune PID parameters online with ITAE as the optimization objective. Applying this method to a typical second-order joint dynamics model G(s) = 1 / (s^2 + s + 1), simulation results show that the system's step response overshoot is -2.00% (no positive overshoot), the settling time is less than 2.5 seconds, and the differential gain automatically approaches zero, effectively improving motion safety and noise immunity. The TDE–PSO optimization algorithm proposed in this invention produces a PID controller that is superior to the traditional Ziegler–Nichols method and ordinary PSO: smaller overshoot, faster response, and better overall control performance. Therefore, this invention significantly outperforms existing methods in both safety and dynamic performance.
[0043] This invention is applicable to medical assistive devices such as exoskeletons, upper limb and obstetric and gynecological lower limb rehabilitation training robots, and has good clinical application prospects.
[0044] The embodiments disclosed herein are preferred embodiments, but are not limited thereto. Those skilled in the art can readily grasp the spirit of the present invention based on the above embodiments and make different extensions and variations, but as long as they do not depart from the spirit of the present invention, they are all within the protection scope of the present invention.
Claims
1. An overshoot-free adaptive joint control method for obstetric and gynecological rehabilitation robots, characterized in that: The method is implemented through the following steps: Step 1: Obtain the angle feedback signal of the rehabilitation robot joints in real time using an angle sensor; Step 2: Online tuning of the PID controller parameters based on the improved particle swarm optimization algorithm, including: proportional K p Integral K i Differential K d The process of obtaining the optimal PID parameters includes: Step 21: Establish a joint dynamics model for the rehabilitation robot to comprehensively consider the dynamic characteristics of the joint motor rotor inertia, limb load mass, and transmission system damping. The dynamic characteristics of the joints of the rehabilitation robot are constructed as a second-order transfer function, expressed as: In the formula, s represents the complex frequency variable in the Laplace transform; Step 22: When initializing the particle swarm, the initial positions of the particles are generated using a Logistic chaotic mapping. Steps 2 and 3: Design adaptive inertia weight w: The inertia weight w is an inertia weight that decreases non-linearly with the number of iterations; Step 24: Introduce elite time delay feedback: Every preset number of iterations, apply a small perturbation to the current global best particle and re-evaluate its fitness value; Step 25: Use the time-weighted absolute error integral ITAE as the optimization objective function J; The objective function J is expressed as: Where J is the optimization objective function and also the ITAE performance metric. t It is a time variable. e ( t The deviation between the set angle and the actual angle of the joint, t _f This is the simulation termination time; Step 26: Output the optimal PID parameters: proportional K p Integral K i Differential K d ; Step 3: Apply the optimized PID parameters obtained from the tuning to the PID controller to generate control commands. The motor drive module receives the control commands and drives the joint motor to operate. The joint motor causes the joint to move, thereby achieving motion control of the joint without positive overshoot.
2. The method for overshoot-free adaptive joint control of an obstetric and gynecological rehabilitation robot according to claim 1, characterized in that: The small perturbation mentioned in step two is a random number that follows a Gaussian distribution with a mean of 0 and a standard deviation of 0.
01.
3. The overshoot-free adaptive joint control method for an obstetric and gynecological rehabilitation robot according to claim 1 or 2, characterized in that: The improved particle swarm optimization algorithm method further includes the step of setting the parameter search boundary of the PID controller: Ratio K p The search range is 0.5 ≤ K. p ≤2.0; integral K i The search range is 0.01 ≤ K i ≤ 0.4; Differential K d The search range is 0 ≤ K d ≤ 0.
1.
4. An overshoot-free adaptive joint control system for an obstetric and gynecological rehabilitation robot, characterized in that, The control system includes: Angle sensor is used to acquire angle feedback signals of the joints of the rehabilitation robot in real time; The motor drive module is used to receive control commands generated by the PID controller and drive the joint motor to operate. The joint motor makes the joint move, realizing motion control of the joint without positive overshoot. A PID controller, which has an embedded computer program configured to perform the overshoot-free adaptive joint control method as described in any one of claims 1-3; A host computer or embedded processor is used to run the improved particle swarm optimization algorithm and output the optimal PID parameters; The improved particle swarm optimization algorithm includes: A joint dynamics model of the rehabilitation robot was established to comprehensively consider the dynamic characteristics of the joint motor rotor inertia, limb load mass, and transmission system damping. The dynamic characteristics of the joints of the rehabilitation robot are constructed as a second-order transfer function, expressed as: In the formula, s represents the complex frequency variable in the Laplace transform; Then, when initializing the particle swarm, the initial positions of the particles are generated using a Logistic chaotic mapping. Next, an adaptive inertia weight w is designed: the inertia weight w is an inertia weight that decreases non-linearly with the number of iterations; Then, an elite time delay feedback mechanism is introduced: every preset number of iterations, a small perturbation is applied to the current global best particle and its fitness value is re-evaluated; The time-weighted absolute error integral is then optimized using ITAE as the objective function J. The objective function J is expressed as: Where J is the optimization objective function and also the ITAE performance metric. t It is a time variable. e ( t The deviation between the set angle and the actual angle of the joint, t _f This is the simulation termination time; Then output the optimal PID parameter: proportional K p Integral K i Differential K d .
5. The overshoot-free adaptive joint control system for an obstetric and gynecological rehabilitation robot according to claim 4, characterized in that: The small perturbation is a random number following a Gaussian distribution with a mean of 0 and a standard deviation of 0.01; the improved particle swarm optimization algorithm also includes the step of setting the parameter search boundary of the PID controller. The ratio K p The search range is 0.5 ≤ K. p ≤2.0; integral K i The search range is 0.01 ≤ K i ≤ 0.4; Differential K d The search range is 0 ≤ K d ≤ 0.
1.
6. A non-overshoot adaptive joint control method for an obstetric and gynecological rehabilitation robot according to claim 5, characterized in that: The angle sensor is a rotary encoder or an inertial measurement unit (IMU).
7. A non-overshoot adaptive joint control system for an obstetric and gynecological rehabilitation robot according to claim 6, characterized in that: The PID controller is deployed in an ARM Cortex-M series microcontroller or an industrial control computer.
8. The overshoot-free adaptive joint control system for an obstetric and gynecological rehabilitation robot according to claim 7, characterized in that: The small perturbation is a random number that follows a Gaussian distribution with a mean of 0 and a standard deviation of 0.
01.
9. A device for an overshoot-free adaptive joint control method for an obstetric and gynecological rehabilitation robot, characterized in that, include: processor; A memory in which executable instructions of the processor are stored; The processor is configured to perform the steps of the overshoot-free adaptive joint control method for an obstetric rehabilitation robot according to any one of claims 1 to 3 by executing the executable instructions.
10. A computer-readable storage medium for storing a program, characterized in that, When the program is executed, it performs the steps of the method according to any one of claims 1 to 3.