A master-slave teleoperation method based on adaptive signal and dynamic control optimization

Abstract

The application specifically relates to a master-slave remote operation method based on adaptive signal and dynamic control optimization, and steps include: acquiring a master terminal signal, performing Kalman filtering on the master terminal signal to obtain a filtered master terminal signal; based on the filtered master terminal signal, constructing a buffer signal sequence, performing smoothing processing on the buffer signal sequence to obtain a smoothed master terminal signal; the smoothed master terminal signal passes through a slave proportional-differential controller to generate an acceleration control signal, the acceleration control signal is integrated and simultaneously subjected to saturation constraint to obtain a displacement instruction for controlling the movement of a mechanical arm. Under unstable time delay signal conditions, the master-slave remote operation method significantly improves the high-frequency servo precision, trajectory tracking performance and dynamic response stability of a slave mechanical arm, and provides technical support for stable and efficient operation of a master-slave remote operation system.

Description

Technical Field

[0001] This invention belongs to the field of robot teleoperation control technology, specifically relating to a master-slave teleoperation method based on adaptive signals and dynamic control optimization. Background Technology

[0002] Master-slave teleoperation systems are widely used in medical surgery, remote maintenance, and hazardous environment operations. This system controls the slave robotic arm via operating signals from the master end to perform precise tasks. However, in practical applications, the slave robotic arm typically requires high-frequency servo precision to ensure effective handling of complex dynamic tasks. Due to the jitter and non-uniformity of the master-end signal, directly transmitting it to the slave end for control may cause unstable system response, or even dynamic instability.

[0003] In existing technologies, the irregularity of the master-end signal mainly stems from the delay and instability of the communication link. This causes the signal to fluctuate irregularly during transmission, making it impossible to transmit to the slave end in real time at uniform time intervals. This signal irregularity directly affects the servo accuracy and dynamic response performance of the slave robotic arm, especially when performing high-frequency dynamic tasks, which may lead to increased trajectory tracking errors and instability in dynamic response.

[0004] To address this issue, existing solutions typically employ signal filtering techniques to smooth the master-end signal. Common filtering methods, such as weighted moving average, median filtering, or Kalman filtering, while effectively suppressing noise and fluctuations in the signal, do not adequately consider the high-frequency servo accuracy requirements of the slave end during design. Furthermore, when the slave controller handles extreme signals, the system may exhibit a nonlinear response, thereby affecting the system's robustness and stability.

[0005] Therefore, designing a strategy that comprehensively considers real-time filtering of master signals and efficient control of slave signals to improve the servo accuracy and dynamic response performance of slave robotic arms under irregular signal conditions has become a key problem that urgently needs to be solved in the field of master-slave teleoperation systems. Summary of the Invention

[0006] This invention proposes a master-slave teleoperation method based on adaptive signal and dynamic control optimization, aiming to solve the problem of uneven and real-time transmission of master-end signals to slave-end robotic arms due to communication link instability and latency. This is particularly important when the slave-end robotic arm performs high-frequency dynamic tasks, requiring precise servo control and response performance. This invention introduces a buffer sequence to smooth the master-end signal and combines it with a proportional-derivative (PD) control algorithm to optimize the motion performance of the slave-end robotic arm, significantly improving trajectory tracking accuracy and dynamic response stability.

[0007] The technical means employed in this invention are as follows:

[0008] A master-slave teleoperation method based on adaptive signal and dynamic control optimization includes the following steps:

[0009] S1. Obtain the master signal, perform Kalman filtering on the master signal to obtain the filtered master signal;

[0010] S2. Based on the filtered main signal, construct a buffered signal sequence, and smooth the buffered signal sequence to obtain the smoothed main signal;

[0011] S3. The smoothed master signal is used to generate an acceleration control signal through a slave proportional-derivative controller. The acceleration control signal is integrated and saturated to obtain the displacement command used to control the movement of the robotic arm.

[0012] Furthermore, the filtered main-end signal is obtained based on the following formula:

[0013]

[0014] Among them, K k For Kalman gain, z k Let H be the observed value of the master signal at time k, and H be the observation matrix. The filtered master signal is the observed value at time k-1. This represents the observed value of the filtered master signal at time k.

[0015] Furthermore, based on the filtered main signal, a buffered signal sequence is constructed, including:

[0016] Construct a buffer sequence of length M, and store the filtered main signal into the buffer sequence to obtain the buffer signal sequence. The calculation formula for the buffer signal sequence is as follows:

[0017]

[0018] in, For buffered signal sequences, The filtered master signal is the observed value at time k. The filtered master signal is observed at time k-M+1. The value is the observed value of the filtered master signal at time k-M+2.

[0019] Furthermore, the buffered signal sequence is smoothed to obtain the smoothed main signal, including:

[0020] The buffered signal sequence is input into the buffer sequence module. Using the sliding window technique, the historical filtered master signal is stored for a period of time. Then, the historical filtered master signal is reconstructed using a linear interpolation algorithm to obtain the smoothed master signal. The calculation formula for the smoothed master signal is as follows:

[0021]

[0022] in, The smoothed main-end signal is the observation value at time k. The filtered master signal is the observed value at time k+1. Here, t represents the observed value of the filtered master signal at time k-1, and t is the current time point. k-1 Let t be the time point at time k-1. k+1 This is the time point at time k+1.

[0023] Furthermore, the smoothed master signal is used by the slave proportional-derivative controller to generate an acceleration control signal, including:

[0024] The smoothed master signal is transmitted to the slave signal, and the joint angle is obtained by inversely solving the smoothed master signal. The joint angle is then input to the slave proportional-derivative controller. Through deep coupling, the error between the buffered master signal and the current position of the slave is calculated to obtain the acceleration control signal used to adjust the control output. The acceleration control signal calculation formula is shown below:

[0025]

[0026] Among them, a k Let x be the acceleration control signal at time k. desired Joint angle, For the target velocity, x k Let k be the position of the slave end. Let K be the velocity at time k from the slave end. p For proportional gain, K d This is the differential gain.

[0027] Furthermore, the acceleration control signal is integrated to obtain the velocity command and displacement command. The formula for calculating the velocity command is shown below:

[0028]

[0029] in, Let k be the velocity of the slave end at time k. Let a be the velocity of the slave end at time k-1. k Let be the acceleration control signal at time k, and Δt be the time step between time k and time k-1.

[0030] The formula for calculating the displacement command is as follows:

[0031]

[0032] in, Let x be the velocity of the slave end at time k. k Let x be the displacement of the slave end at time k. k-1 Let Δt be the displacement of the slave end at time k-1, and Δt be the time step between time k and time k-1.

[0033] Furthermore, the formula for saturating the acceleration control signal is shown below:

[0034]

[0035] in, Let k be the velocity at time k. Let k be the acceleration at time k. For acceleration limitation, Speed ​​limit.

[0036] Compared with the prior art, the present invention has the following advantages:

[0037] 1. This invention utilizes an online filter to smooth irregular signals transmitted from the master end in real time, eliminating signal jitter and noise interference, and providing a more stable input signal for the slave end.

[0038] 2. In this invention, the slave robotic arm adopts a deeply coupled proportional-derivative (PD) control algorithm, which effectively improves the adaptability and servo accuracy to complex dynamic tasks.

[0039] 3. In this invention, to prevent extreme behavior of the system response, an integral saturation term is introduced to constrain the velocity and acceleration, thereby improving the stability and robustness of the system.

[0040] Based on the above reasons, this invention can be widely applied in fields such as robot teleoperation control. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a block diagram of the overall remote operation control of the present invention;

[0043] Figure 2This is a position diagram of the slave controller tracking irregular signal trajectories according to the present invention.

[0044] Figure 3 This is a graph showing the tracking speed of the slave controller for irregular signal trajectories according to the present invention.

[0045] Figure 4 This is a graph showing the acceleration curve of the slave controller tracking an irregular signal trajectory according to the present invention. Detailed Implementation

[0046] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0047] like Figure 1 As shown, this invention provides a master-slave teleoperation method based on adaptive signals and dynamic control optimization, the specific steps of which include:

[0048] S1. Obtain the master signal, perform Kalman filtering on the master signal to obtain the filtered master signal.

[0049] In a master-slave remote operating system, the master signal s k Due to the instability and latency of communication links, as well as hand tremors, irregularities and noise occur during transmission. To address this, this invention designs a real-time signal smoothing method based on buffer sequences to reduce signal fluctuations caused by communication delays and noise. The Kalman filter, through recursive estimation and measurement processes, can perform real-time smoothing of the main signal under noisy conditions, thereby obtaining a more accurate and stable signal.

[0050] First, let the discretized representation of the master signal be s. k , where k represents the discrete sampling time. The master signal is filtered using a Kalman filter algorithm to obtain the filtered master signal. The formula for calculating the filtered master signal is as follows:

[0051]

[0052] Among them, K k For Kalman gain, z k Let H be the observed value of the master signal at time k, and H be the observation matrix. The filtered master signal is the observed value at time k-1. This represents the observed value of the filtered master signal at time k.

[0053] This invention uses an adaptive Kalman filter algorithm to adjust the filter parameters (such as the covariance matrix) in real time and automatically adapt to different noise levels according to changes in the signal.

[0054] S2. Based on the filtered main signal, construct a buffer signal sequence, and smooth the buffer signal sequence to obtain the smoothed main signal.

[0055] Construct a buffer sequence of length M. The filtered main signal is stored in a buffer sequence to obtain the buffer signal sequence. The calculation formula for the buffer signal sequence is as follows:

[0056]

[0057] in, For buffered signal sequences, The filtered master signal is the observed value at time k. The filtered master signal is observed at time k-M+1. The value is the observed value of the filtered master signal at time k-M+2.

[0058] The buffered signal sequence is input into the buffer sequence module. Using the sliding window technique, the historical filtered master signal is stored for a period of time. Then, the historical filtered master signal is reconstructed using a linear interpolation algorithm to obtain the smoothed master signal. The calculation formula for the smoothed master signal is as follows:

[0059]

[0060] in, The smoothed main-end signal is the observation value at time k. The filtered master signal is the observed value at time k+1. Here, t represents the observed value of the filtered master signal at time k-1, and t is the current time point. k-1 Let t be the time point at time k-1. k+1 This is the time point at time k+1.

[0061] By buffering the smoothed signal, latency issues that may arise due to communication delays are eliminated. The buffer module stores and updates the signal incrementally, ensuring the continuity and consistency of the signal received from the receiving end and reducing signal irregularities caused by communication link instability. Signal updates in the buffer sequence follow a "first-in, first-out" (FIFO) rule, ensuring that the latest signal is added incrementally while historical signals are removed sequentially.

[0062] S3. The smoothed master signal is used to generate an acceleration control signal through a slave proportional-derivative controller. The acceleration control signal is integrated and saturated to obtain the displacement command used to control the movement of the robotic arm.

[0063] The smoothed master signal is transmitted to the slave end. The smoothed master signal is then decomposed to obtain the joint angle. The joint angle is input to the proportional-derivative (PD) controller. Through deep coupling, the error between the buffered master signal and the current position of the slave robotic arm is calculated. The slave's execution action is adjusted in real time to ensure high-precision trajectory tracking and stable dynamic response. This yields the PD control formula and acceleration control signal. The acceleration control signal calculation formula is shown below:

[0064]

[0065] Among them, a k Let x be the acceleration control signal at time k. desired Joint angle, For the target velocity, x k and K represents the position and velocity of the slave robotic arm at time k, respectively. p and K d These are the proportional gain and the derivative gain, respectively.

[0066] The target speed here is generally 0, and the target of master-slave teleoperation is a stopped state.

[0067] By integrating the acceleration control signal, velocity and displacement commands are obtained. The displacement commands are then used to control the movement of the robotic arm, thereby improving the dynamic response and control accuracy of the system.

[0068] The formula for calculating the speed command is as follows:

[0069]

[0070] in, Let k be the speed of the slave robotic arm at time k. Let a be the speed of the slave robot arm at time k-1. k Let Δt be the acceleration control signal at time k, and Δt be the time step between time k and time k-1. The formula for calculating the displacement command is as follows:

[0071]

[0072] in, Let x be the speed of the slave robot arm at time k. k Let x be the displacement of the slave robot arm at time k. k-1Let Δt be the displacement of the slave robot arm at time k-1, and let Δt be the time step between time k and time k-1.

[0073] To prevent the system from responding too quickly or excessively, a saturation constraint is applied to the acceleration control signal to limit the maximum values ​​of acceleration and velocity, ensuring that the system response does not exceed a preset safety range. The calculation formula for the saturation constraint is shown below:

[0074]

[0075] in, Let k be the velocity of the end-effector at time k. Let K be the acceleration of the end-effector at time k. For acceleration limitation, Speed ​​limit.

[0076] By introducing saturation terms into the control system, the PD control system transforms from a traditional second-order linear system into a system with nonlinear characteristics. This transformation directly affects the system's dynamic behavior and response characteristics, primarily manifested in the limitation of acceleration and velocity, thereby effectively preventing the system from responding too quickly or excessively. To better enable the nonlinear system to achieve tracking performance, the response time, maximum velocity, and maximum acceleration constraints of the aforementioned system are assessed below.

[0077] Assume the system equations for slave-end PD control with normal integral are as follows:

[0078]

[0079] Where x(t) is the actual position of the system at time t. Let x(t) be the position and x(t) be the velocity of the system at time t. Let x(t) be the position x(t), ζ be the system acceleration at time t, and ω be the damping ratio. n x is the natural angular frequency. desired (t) represents the target position of the system at time t.

[0080] The approximate characteristic calculation formula for the proportional-derivative controller after introducing saturation constraints is shown below:

[0081]

[0082] Among them, T r The approximate response time of the system is... It is the acceleration under amplitude limiting. It is the speed under amplitude limiting, x desired Joint angle, For acceleration limitation, For speed constraints, Δt is the discrete time interval of the discrete system, and ω nLet ω be the natural angular frequency, ζ be the damping ratio, and x be the actual position of the system. For time speed.

[0083] Example

[0084] To demonstrate the feasibility of the above invention, a simulation experiment was further conducted on a master-slave teleoperation method based on adaptive signal and dynamic control optimization. In the simulation experiment, the maximum acceleration of the slave arm was set to 15 rad / s². 2 The maximum speed is limited to 3 rad / s, and the main signal is a segmented, irregular signal. The simulation tracking experiment results are as follows: Figures 2-4 As shown, the slave end is instructed to follow the following joint trajectory:

[0085]

[0086] Among them, u d (t) is the piecewise function of the main terminal position signal containing irregular signals. It is the corresponding velocity function. That is the corresponding acceleration function.

[0087] like Figures 2 to 4 As shown, simulation verification demonstrates that the introduction of the saturation term improves the stability and robustness of the system, while also enhancing the smoothness of the dynamic response, ensuring the high-frequency servo accuracy and trajectory tracking performance of the slave robot under unstable time delay and signal jitter conditions.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A master-slave teleoperation method based on adaptive signal and dynamic control optimization, characterized by the steps of Comprise: S1. Obtain the master end signal, Kalman filter the master end signal to obtain the filtered master end signal, and obtain the filtered master end signal based on the following formula: wherein, is the Kalman gain, is the observation value of the primary end signal at time k, is the observation matrix, is the observation value of the filtered primary end signal at time k-1, is the observation value of the filtered primary end signal at time k; S2. Based on the filtered master end signal, construct the buffer signal sequence, smooth the buffer signal sequence to obtain the smoothed master end signal; S3. The smoothed master end signal passes through the slave proportional-differential controller to generate the acceleration control signal, integrates the acceleration control signal, and simultaneously saturates the acceleration control signal to obtain the displacement instruction for controlling the movement of the mechanical arm, The smoothed master end signal passes through the slave proportional-differential controller to generate the acceleration control signal, comprising: The smoothed master end signal is transmitted to the slave end, the joint angle is obtained by inversely solving the smoothed master end signal, the joint angle is input to the slave proportional-differential controller, the error between the buffered master end signal and the current position of the slave end is calculated by the method of deep coupling, and the acceleration control signal for adjusting the control output is obtained, and the acceleration control signal calculation formula is as follows: wherein, is the acceleration control signal at time k, is the joint angle, is the target velocity, is the position of the slave end at time k, is the velocity of the slave end at time k, is the proportional gain, is the derivative gain.

2. The master-slave teleoperation method based on adaptive signal and dynamic control optimization of claim 1, wherein, Based on the filtered master end signal, construct the buffer signal sequence, comprising: A buffer sequence with a length of M The filtered main end signal is stored into the buffer sequence to obtain a buffer signal sequence, and the calculation formula of the buffer signal sequence is as follows: wherein, is the buffered signal sequence, is the filtered primary-side signal observation at time k, is the filtered primary-side signal observation at time k-M+1, is the filtered primary-side signal observation at time k-M+2.

3. The master-slave teleoperation method based on adaptive signal and dynamic control optimization of claim 1, wherein, Smooth the buffer signal sequence to obtain the smoothed master end signal, comprising: The buffer signal sequence is input to the buffer sequence module, a period of time of historical filtered master end signal is stored by the sliding window technology, the historical filtered master end signal is reconstructed by the linear interpolation algorithm, and the smoothed master end signal is obtained, and the calculation formula of the smoothed master end signal is as follows: wherein, is an observation value of the smoothed main-end signal at time k, is an observation value of the filtered main-end signal at time k+1, is an observation value of the filtered main-end signal at time k-1, t is a time point of the current time, is a time point of time k-1, is a time point of time k+1.

4. The master-slave teleoperation method based on adaptive signal and dynamic control optimization of claim 1, wherein, Integrate the acceleration control signal to obtain the speed instruction and the displacement instruction, and the calculation formula of the speed instruction is as follows: wherein, is the slave speed at time k, is the slave speed at time k-1, is the acceleration control signal at time k, is the time step between time k and k-1, The calculation formula of the displacement instruction is as follows: wherein, is the slave speed at time k, is the slave displacement at time k, is the slave displacement at time k-1, is the time step between time k and k-1.

5. The master-slave teleoperation method based on adaptive signal and dynamic control optimization of claim 1, wherein, The formula for saturating the acceleration control signal is as follows: wherein, is the velocity at the end k time, is the acceleration at the end k time, is the acceleration limit, is the velocity limit.

Images