Isomorphic human-robot upper limb master-slave incremental mapping method and system

Abstract

The application discloses a kind of isomorphic man-machine upper limb master-slave incremental mapping method and system, method includes: determining muscle coordination lag quantity;According to the increment direction and increment amplitude of joint angle increment signal, the motion gain adjustment direction of each joint is determined in combination with muscle coordination lag quantity;Contact force / torque signal of slave end mechanical arm end effector and the current angle of each joint are obtained;According to contact force / torque signal, the deviation direction of end effector actual motion trajectory and expected motion trajectory is calculated, and the load-motion coupling coefficient of each joint is determined in combination with current joint angle;With motion gain adjustment direction and load-motion coupling coefficient, the incremental mapping proportion coefficient of each joint is determined;Incremental mapping proportion coefficient is used to scale joint angle increment signal, and the angle increment instruction of each joint of slave end mechanical arm is obtained and output execution. Operator physiological signal and slave end environment interaction force are used to realize the adaptive adjustment of mapping proportion coefficient.

Description

Technical Field

[0001] This invention belongs to the field of human-computer interaction control technology, and in particular relates to a master-slave incremental mapping method and system for isomorphic human-computer upper limbs. Background Technology

[0002] Homogeneous human-machine upper limb master-slave control systems are widely used in fields such as teleoperation, medical rehabilitation, and industrial robotics. The operator controls the slave robotic arm to perform tasks through upper limb movements, and the incremental joint angles of the master operator are mapped to drive the corresponding movements of each joint of the slave robotic arm.

[0003] Existing master-slave incremental mapping methods have the following main technical drawbacks: First, the mapping ratio is fixed or only manually adjustable by the operator. In existing systems, the mapping ratio from the master joint angle increment to the slave joint angle increment is usually a preset constant or manually switched by the operator via a switch. When the slave load or the operator's intention changes, the fixed mapping ratio cannot adapt, resulting in the slave responding too quickly or too slowly, reduced operational accuracy, or decreased safety.

[0004] Second, the existing methods fail to utilize the operator's physiological state information. They only use the operator's joint angles as input signals, without considering the impact of the operator's muscle state, force patterns, and movement strategies on master-slave control. The operator's muscle coordination patterns differ significantly between rapid and fine motor movements, and existing methods cannot distinguish these differences or adjust the mapping relationship accordingly.

[0005] Third, the influence of environmental interaction forces on the end effector is not considered. When the end effector operates in a confined environment, the contact forces between the end effector and the environment affect the motion response of each joint through dynamic coupling. Existing methods map each joint independently without considering the influence of load on joint motion coupling, resulting in unstable motion or collisions between the end effector and the environment in confined environments.

[0006] Therefore, a method is needed that can adaptively adjust the scaling factor of each joint increment using the operator's physiological signals and the interaction force of the slave environment. Summary of the Invention

[0007] This invention provides a master-slave incremental mapping method and system for isomorphic human-machine upper limbs to solve at least one of the above-mentioned technical problems.

[0008] In a first aspect, the present invention provides a master-slave incremental mapping method for isomorphic human-machine upper limbs, comprising: Acquire the incremental joint angle signals of each joint of the operator's upper limb and the operator's surface electromyography signals; The activation sequence of each muscle group is extracted from the surface electromyography signal, and the muscle coordination lag of the operator's current joint is determined based on the activation sequence. Based on the incremental direction and amplitude of each joint in the incremental joint angle signal, and combined with the muscle coordination hysteresis, the direction of motion gain adjustment for each joint is determined. Acquire the contact force signal and contact torque signal between the end effector of the slave robotic arm and the working environment, as well as the current joint angle of each joint of the slave robotic arm; The deviation direction between the actual motion trajectory and the desired motion trajectory of the end effector of the slave robotic arm is calculated based on the contact force signal and the contact torque signal. The load-motion coupling coefficient of each joint is determined based on the deviation direction and the current joint angle. The incremental mapping ratio coefficient of each joint is determined by the motion gain adjustment direction of each joint and the load-motion coupling coefficient of each joint. The incremental angle signal of each joint is scaled using the incremental mapping scaling coefficient of each joint to obtain the angle increment command of each joint of the slave robotic arm, and the angle increment command is output to the joint controller of the slave robotic arm for execution.

[0009] Secondly, the present invention provides an isomorphic human-machine upper limb master-slave incremental mapping system, comprising: The first acquisition module is configured to acquire the joint angle increment signals of each joint of the operator's upper limb and the operator's surface electromyography signals. The first determining module is configured to extract the activation sequence of each muscle group from the surface electromyography signal, and determine the muscle coordination lag of the operator's current joint based on the activation sequence. The second determining module is configured to determine the motion gain adjustment direction of each joint based on the incremental direction and amplitude of each joint in the incremental joint angle signal, combined with the muscle coordination hysteresis. The second acquisition module is configured to acquire the contact force signal and contact torque signal between the end effector of the slave robotic arm and the working environment, as well as the current joint angle of each joint of the slave robotic arm; The calculation module is configured to calculate the deviation direction between the actual motion trajectory and the desired motion trajectory of the end effector of the slave robotic arm based on the contact force signal and the contact torque signal, and to determine the load-motion coupling coefficient of each joint based on the deviation direction and the current joint angle. The third determining module is configured to determine the incremental mapping ratio coefficient of each joint by combining the motion gain adjustment direction of each joint with the load-motion coupling coefficient of each joint. The output module is configured to scale the joint angle increment signal using the incremental mapping scaling coefficient of each joint to obtain the angle increment command of each joint of the slave robotic arm, and output the angle increment command to the joint controller of the slave robotic arm for execution.

[0010] Thirdly, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the isomorphic human-machine upper limb master-slave incremental mapping method according to any embodiment of the present invention.

[0011] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the processor performs the steps of the isomorphic human-machine upper limb master-slave incremental mapping method according to any embodiment of the present invention.

[0012] The isomorphic human-machine upper limb master-slave incremental mapping method and system of this application have the following beneficial effects: The activation windows of agonist and antagonist muscles are extracted from surface electromyography (EMG) signals. The difference between the co-activation duration and the alternating activation duration determines whether the operator is using a speed-first or precision-first strategy. This determination does not rely on the absolute value of EMG amplitude, but only on the temporal structure characteristics of the activation windows, making it immune to differences in EMG amplitude between different operators.

[0013] In speed-priority mode, a larger increment amplitude results in stronger gain enhancement, enabling the slave end to quickly respond to large-value operation intentions. In precision-priority mode, a larger increment amplitude results in stronger gain suppression, enabling the slave end to smoothly respond to large-value operation intentions and avoid overshoot in fine-grained operations. The gain adjustment directions are completely opposite in the two modes, realizing adaptive gain direction switching based on the operation strategy.

[0014] By comparing the deviation between the actual and desired motion directions of the end effector and the direction of motion, and combining this with the sign of the equivalent contact torque, it is determined whether each joint is in an environmental assistance or environmental resistance state under the current load conditions. This coupling coefficient enhances the mapping ratio in speed-priority mode and suppresses the mapping ratio in precision-priority mode, thus achieving adaptive utilization and avoidance of environmental interaction forces at the end. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. 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.

[0016] Figure 1 A flowchart of a master-slave incremental mapping method for isomorphic human-machine upper limbs provided in an embodiment of the present invention; Figure 2 This is a structural block diagram of an isomorphic human-machine upper limb master-slave incremental mapping system provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0017] 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.

[0018] Please see Figure 1 The diagram shows a flowchart of a master-slave incremental mapping method for isomorphic human-machine upper limbs according to this application.

[0019] like Figure 1 As shown, the master-slave incremental mapping method for the upper limbs of isomorphic human-machine interfaces specifically includes the following steps: Step S101: Obtain the joint angle increment signal of each joint of the operator's upper limb and the surface electromyography signal of the operator.

[0020] In this step, accelerometers are installed at each joint of the operator's upper limbs to continuously collect acceleration signals in the direction of movement of each joint at a preset sampling frequency. Perform an integral operation on the acceleration signal to obtain a velocity signal, perform an integral operation on the velocity signal to obtain a displacement signal, and use the displacement signal as the joint angle increment signal; An electrode array is installed on the skin surface of the relevant muscle groups of the operator, and the electromyographic signals at each electrode are synchronously collected at the preset sampling frequency to obtain the original electromyographic signals of each muscle group. The original electromyographic (EMG) signal is subjected to bandpass filtering and rectification smoothing to obtain preprocessed EMG signals for each muscle group, and the preprocessed EMG signals are used as the surface EMG signals.

[0021] In one specific embodiment, accelerometers are first installed at various joints of the operator's upper limbs. The accelerometers are microelectromechanical systems (MEMS) triaxial accelerometers with a range of ±8g, a resolution of 0.1mg, and an output frequency of 1000Hz. The accelerometers are fixed to the distal segment surfaces of each joint using medical double-sided tape or elastic bandages, aligning the sensitive axis of the accelerometer with the direction of movement of that joint. The joints include at least the shoulder, elbow, and wrist joints, with one accelerometer corresponding to each joint.

[0022] Acceleration signals in the motion direction of each joint are continuously acquired at a preset sampling frequency of 1000Hz to obtain the original acceleration signal sequence of each joint. Each sampling point in the original acceleration signal sequence corresponds to a timestamp and an acceleration value, with the unit of acceleration value being meters per square second.

[0023] An integral operation is performed on the acceleration signal to obtain a velocity signal. This integral operation uses the trapezoidal numerical integration method, starting from the beginning of the acceleration signal sequence, averaging the acceleration values ​​between two adjacent sampling points, multiplying by the sampling interval, and accumulating point by point to obtain the velocity value. An integral operation is then performed on the velocity signal using the same trapezoidal numerical integration method to obtain a displacement signal. The unit of the displacement signal is meters, representing the linear displacement of the operator's joint in the direction of motion. This displacement signal is used as the joint angle increment signal. When the joint motion is rotational, there is a fixed proportional relationship between the displacement signal and the joint angle increment, which is equal to the reciprocal of the joint's rotation radius. The joint angle increment signal is a radian value.

[0024] An electrode array is mounted on the skin surface of the operator's relevant muscle groups. These relevant muscle groups include at least an agonist muscle group that drives the movement of each joint and an antagonist muscle group whose function is opposite to that of the agonist muscle group. The agonist and antagonist muscle groups are distributed on both sides of the joint. For the shoulder joint, the relevant muscle groups include the anterior deltoid as the agonist muscle group and the latissimus dorsi as the antagonist muscle group; for the elbow joint, the relevant muscle groups include the biceps brachii as the agonist muscle group and the triceps brachii as the antagonist muscle group; for the wrist joint, the relevant muscle groups include the flexor carpi radialis as the agonist muscle group and the extensor carpi radialis as the antagonist muscle group.

[0025] The electrode array is a surface dry electrode array comprising multiple electrode units arranged along the muscle fiber direction, with a spacing of 20 mm between the electrode units. Each electrode unit contains an Ag / AgCl electrode pad with a diameter of 10 mm. Before electrode installation, the skin surface is cleaned to remove surface oil and stratum corneum, ensuring a contact impedance of less than 5 kΩ between the electrode and the skin. Each electrode is connected to a bioelectric signal amplifier via wires. The amplifier has a gain of 1000, a common-mode rejection ratio greater than 100 dB, and an input impedance greater than 10 megohms.

[0026] Electromyographic signals at each electrode are synchronously acquired at the preset sampling frequency of 1000Hz to obtain the raw electromyographic signals of each muscle group. The raw electromyographic signals are microvolt-level voltage signals with a frequency range covering 10Hz to 500Hz.

[0027] The raw electromyographic signals of each muscle group were subjected to bandpass filtering. The cutoff frequencies of the bandpass filter were 20Hz and 450Hz, filtering out low-frequency motion artifacts below 20Hz and high-frequency noise above 450Hz. A fourth-order Butterworth filter was used for the bandpass filtering to ensure a flat amplitude response within the passband and no phase distortion.

[0028] The bandpass-filtered electromyographic (EMG) signal is rectified by taking the absolute value, flipping the negative values ​​to positive values. The rectified EMG signal is then smoothed using a sliding window averaging method, with a window length of 100 milliseconds corresponding to 100 sampling points, resulting in preprocessed EMG signals for each muscle group. These preprocessed EMG signals are defined as the envelope curves of the EMG signals, characterizing the time-varying intensity of electromyographic activity. These preprocessed EMG signals are then used as the surface electromyographic (EMG) signals.

[0029] Step S102: Extract the activation sequence of each muscle group from the surface electromyography signal, and determine the muscle coordination lag of the operator's current joint based on the activation sequence.

[0030] In this step, surface electromyography (EMG) signals of each muscle group are acquired. A preset sliding window is slid across the surface EMG signals of each muscle group, and the short-time energy value of the EMG signal within the window is calculated at each window position to obtain the short-time energy sequence corresponding to each muscle group. The baseline value is extracted from the short-time energy sequence, and the baseline value plus a preset multiple of the baseline standard deviation is used as the activation threshold of the muscle group. The starting position where the short-time energy values ​​of a consecutive preset number of sampling points in the short-time energy sequence all exceed the activation threshold is used as the activation start time of the muscle group. The starting position where the short-time energy values ​​of a consecutive preset number of sampling points in the short-time energy sequence all fall below the activation threshold is used as the activation end time of the muscle group. The time interval between the activation start time and the activation end time is used as the activation window of the muscle group. For a certain joint, the activation window of the first muscle group that drives the joint to move in a first direction is obtained as the first activation window, and the activation window of the second muscle group that drives the joint to move in a second direction is obtained as the second activation window, wherein the first direction is opposite to the second direction. The earliest of the start times of the first activation window and the second activation window is taken as the muscle control start time of the joint, and the later of the end times of the first activation window and the second activation window is taken as the muscle control end time of the joint. The time interval between the muscle control start time and the muscle control end time is taken as the joint control window. Furthermore, to avoid a situation where the time spans of the first and second activation windows differ significantly, and the total duration of the joint control window is primarily determined by the activation window of a single muscle group, in a preferred embodiment, the joint control window undergoes validity verification and normalization: the duration T1 of the first activation window and the duration T2 of the second activation window are calculated, with Tmin = min(T1, T2) and Tmax = max(T1, T2); if Tmin / Tmax is less than a preset window consistency threshold, such as 0.5, it is determined that the time span difference between the two muscle groups is too large. In this case, the total duration of the joint control window is not directly used as the normalization denominator, but the average of T1 and T2 is used as the normalization denominator, i.e.: Muscle coordination lag = (Total activation duration - Alternating activation duration) / ((T1 + T2) / 2); The calculation results are then limited to a range between -1 and 1. If Tmin / Tmax is not less than a preset window consistency threshold, normalization is still performed based on the total duration of the joint control window. This process eliminates the impact of excessive differences in the durations of the two activation windows on the calculation of muscle coordination lag, improving computational stability for different operators and tasks.

[0031] Within the joint control window, starting from the moment the muscle control begins, the first activation window and the second activation window are marked sequentially along the time axis at each moment, indicating whether the first activation window and the second activation window are in an active or inactive state, thus obtaining the state sequence corresponding to the joint control window. The state type at each moment in the state sequence includes one of four states: only the first muscle group is activated, only the second muscle group is activated, both muscle groups are activated simultaneously, and neither muscle group is activated. The total duration of simultaneous activation of two muscle groups in the state sequence is taken as the co-activation duration, and the total duration of activation of only the first muscle group and activation of only the second muscle group in the state sequence is taken as the alternating activation duration. The muscle coordination lag of the joint is obtained by dividing the difference between the co-activation duration and the alternating activation duration by the total duration of the joint control window.

[0032] In one specific embodiment, the surface electromyography (EMG) signals of each muscle group obtained in step S101 are first acquired. The surface EMG signals are EMG envelope signals after bandpass filtering, rectification, and smoothing, and the surface EMG signals of each muscle group are aligned on the time axis.

[0033] A sliding scan is performed on the surface electromyography (EMG) signals of each muscle group using a preset sliding window. The window length is determined based on the physiological response time of muscle activation, with 150 sampling points corresponding to 150 milliseconds. The step size of the sliding window is one sampling point, meaning each sampling moment serves as the end position of the window. At each window position, the root mean square (RMS) value of the EMG signal within the window is calculated as the short-time energy value for that window. The formula is the square root of the sum of the squares of the amplitudes of each sampling point within the window, divided by the window length. The short-time energy values ​​calculated at each window position are arranged in chronological order to obtain the short-time energy sequence corresponding to each muscle group. The short-time energy sequence has the same duration as the original EMG signal, but high-frequency fluctuations between sampling points are eliminated after window smoothing.

[0034] Baseline estimation is performed on the short-time energy sequence. The baseline estimation employs a moving median filter method, using a 2-second window length. The median of the short-time energy values ​​within the window is calculated as the baseline value at the center of the window. The baseline values ​​at each time point are arranged chronologically to obtain the baseline sequence. The baseline sequence is then subtracted from the short-time energy sequence to obtain the baseline-debasemented short-time energy sequence. The root mean square value of the baseline-debasemented short-time energy sequence is used as the baseline standard deviation.

[0035] The activation threshold of the muscle group at the current moment is calculated by adding three times the baseline standard deviation to the baseline value in the baseline sequence. When the short-time energy values ​​of five consecutive sampling points in the short-time energy sequence all exceed the activation threshold at the corresponding moment, the moment of the first sampling point among the five consecutive sampling points is taken as the activation start moment of the muscle group. When the short-time energy values ​​of five consecutive sampling points in the short-time energy sequence are all lower than the activation threshold at the corresponding moment, the moment of the first sampling point among the five consecutive sampling points is taken as the activation end moment of the muscle group. The time interval between the activation start moment and the activation end moment is taken as the activation window of the muscle group. When there is no situation in the short-time energy sequence where all five consecutive sampling points exceed the activation threshold, the start moment of the short-time energy sequence is taken as the activation start moment of the muscle group; when there is no situation in the short-time energy sequence where all five consecutive sampling points are lower than the activation threshold, the end moment of the short-time energy sequence is taken as the activation end moment of the muscle group.

[0036] For a given joint, the activation window of a first muscle group that drives the joint to move in a first direction is obtained as the first activation window, where the first direction is the joint flexion direction. The activation window of a second muscle group that drives the joint to move in a second direction is obtained as the second activation window, where the second direction is the joint extension direction and is opposite to the first direction.

[0037] The earlier of the start times of the first activation window and the second activation window is taken as the muscle control start time of the joint. The later of the end times of the first activation window and the second activation window is taken as the muscle control end time of the joint. The time interval between the muscle control start time and the muscle control end time is taken as the joint control window.

[0038] Within the joint control window, with the sampling interval as the time step, starting from the muscle control initiation time, the first activation window and the second activation window are sequentially checked along the time axis at each moment to determine whether they are in an active or inactive state. When the first activation window covers the moment but the second activation window does not, the moment is marked as an active state for only the first muscle group; when the first activation window does not cover the moment but the second activation window does, the moment is marked as an active state for only the second muscle group; when both the first and second activation windows cover the moment, the moment is marked as an active state for both muscle groups; when neither the first nor the second activation window covers the moment, the moment is marked as an inactive state for both muscle groups. The state labels for all moments are arranged chronologically to obtain the state sequence corresponding to the joint control window. Each element in the state sequence corresponds to a sampling moment, and the element value is one of four state types.

[0039] The total number of consecutive elements in the state sequence marked as simultaneously activated by both muscle groups is counted, multiplied by the sampling interval to obtain the co-activation duration. The sum of the total number of consecutive elements in the state sequence marked as activated only by the first muscle group and activated only by the second muscle group is counted, multiplied by the sampling interval to obtain the alternating activation duration. The total number of elements in the state sequence is counted, multiplied by the sampling interval to obtain the total duration of the joint control window.

[0040] The co-activation duration is subtracted from the alternating activation duration. The difference is then divided by the total duration of the joint control window. The quotient is used as the muscle coordination lag of the joint. The muscle coordination lag ranges from -1 to +1. When the muscle coordination lag is positive, it indicates that the co-activation degree of the agonist and antagonist muscles is greater than the alternating activation degree, and the operator's current joint is in a precision-priority control mode. When the muscle coordination lag is negative, it indicates that the alternating activation degree of the agonist and antagonist muscles is greater than the co-activation degree, and the operator's current joint is in a speed-priority control mode. When the muscle coordination lag is zero, it indicates that the co-activation degree and the alternating activation degree are equal, and the operator's current joint is in a speed- and precision-balanced control mode.

[0041] Step S103: Based on the incremental direction and amplitude of each joint in the incremental joint angle signal, and in conjunction with the muscle coordination hysteresis, determine the motion gain adjustment direction of each joint.

[0042] In this step, the sign of the joint angle increment signal of a certain joint at the current moment is obtained, the sign is used as the increment direction of the joint, and the absolute value of the joint angle increment signal is used as the increment amplitude of the joint. The muscle coordination lag of the joint is obtained, and the muscle coordination lag is compared with a preset lag threshold. When the muscle coordination lag is greater than the lag threshold, the control strategy of the joint is marked as a first strategy mode. When the muscle coordination lag is not greater than the lag threshold, the control strategy of the joint is marked as a second strategy mode. The first strategy mode is a speed priority mode, and the second strategy mode is a precision priority mode. When the control strategy is marked as the first strategy mode, the incremental amplitude is compared with each level threshold in the preset amplitude gradient sequence to determine the amplitude level interval in the amplitude gradient sequence. The first direction adjustment step size corresponding to the amplitude level interval is used as the motion gain adjustment direction of the joint. Each amplitude threshold divides the value range of the incremental amplitude into multiple amplitude level intervals. The amplitude gradient sequence contains multiple increasing amplitude thresholds. When the control strategy is marked as the second strategy mode, the incremental amplitude is compared with the thresholds of each level in the amplitude gradient sequence one by one to determine the amplitude level interval in the amplitude gradient sequence. The second direction adjustment step size corresponding to the amplitude level interval is used as the motion gain adjustment direction of the joint. The sign of the second direction adjustment step size is opposite to the sign of the first direction adjustment step size. The motion gain adjustment direction is recorded as the gain adjustment direction label of the joint at the current moment, and the value of the gain adjustment direction label is either the enhancement direction or the suppression direction.

[0043] In one specific embodiment, the joint angle increment signal of a certain joint at the current moment is first acquired. This joint angle increment signal is the sampled value of the displacement signal obtained in step S101 at the current moment. The sign of the sampled value is detected; when the sampled value is positive, the increment direction is recorded as positive; when the sampled value is negative, the increment direction is recorded as negative. The absolute value of the sampled value is taken as the increment amplitude of the joint at the current moment, where the increment amplitude is a non-negative real number, and the unit is radians or angles.

[0044] Obtain the muscle coordination lag of the joint at the current moment. The muscle coordination lag is the value calculated in step S102, and its value ranges from -1 to +1. Obtain a preset lag threshold, which is determined based on the physiological boundary between the operator's speed-priority mode and precision-priority mode, and has a value of 0.2. Compare the muscle coordination lag with the lag threshold.

[0045] Furthermore, the determination of the hysteresis threshold of 0.2 is based on the following: A preset number of healthy operators were selected, and their upper limb shoulder, elbow, and wrist joints were tested for speed-priority and precision-priority tasks, respectively. Surface electromyography (EMG) signals of the agonist and antagonist muscles were collected simultaneously, and the muscle coordination hysteresis of each joint was calculated according to step S102. Experimental results showed that when the muscle coordination hysteresis was greater than 0.2, operators generally exhibited a higher degree of co-activation of the agonist and antagonist muscles, increased joint stiffness, faster movement speed, but higher tolerance for trajectory errors, corresponding to a speed-priority strategy. When the muscle coordination hysteresis was not greater than 0.2, operators generally exhibited a higher degree of alternating activation of the agonist and antagonist muscles, smoother joint movement, and smaller trajectory errors, corresponding to a precision-priority strategy. Therefore, 0.2 was used as the default threshold for distinguishing between speed-priority and precision-priority modes. This threshold depends only on the temporal structure of the activation window and not on the absolute value of the EMG amplitude, thus exhibiting good adaptability to differences in EMG amplitude among different operators. In practice, the value of 0.2 can be fine-tuned according to the individual operator's calibration, for example, by selecting a value within the range of 0.15 to 0.25.

[0046] When the muscle coordination lag exceeds the lag threshold, it is determined that the operator's current joint is in a speed-priority control mode, and the joint's control strategy is marked as a first strategy mode, which is the speed-priority mode. In the speed-priority mode, the operator intends to move the joint quickly, and the slave robotic arm needs to enhance its response gain to the operator's movement commands.

[0047] When the muscle coordination hysteresis is less than or equal to the hysteresis threshold, it is determined that the operator's current joint is in a precision-priority control mode, and the joint's control strategy is marked as a second strategy mode, which is the precision-priority mode. In the precision-priority mode, the operator intends to precisely control the joint position, and the slave robotic arm needs to suppress the response gain to the operator's motion commands to avoid overshoot and oscillation.

[0048] A preset amplitude gradient sequence is obtained, which includes multiple increasing amplitude thresholds. The values ​​of the amplitude thresholds are determined based on the physiological range of motion of each joint of the operator's upper limb. For the shoulder joint, the amplitude thresholds are 5 degrees, 15 degrees, and 30 degrees; for the elbow joint, they are 3 degrees, 10 degrees, and 20 degrees; and for the wrist joint, they are 2 degrees, 5 degrees, and 10 degrees. Each amplitude threshold divides the range of incremental amplitude values ​​into multiple amplitude level intervals. The first level interval is greater than or equal to zero and less than the first amplitude threshold; the second level interval is greater than or equal to the first amplitude threshold and less than the second amplitude threshold; the third level interval is greater than or equal to the second amplitude threshold and less than the third amplitude threshold; and the fourth level interval is greater than or equal to the third amplitude threshold and less than or equal to the maximum physiological amplitude.

[0049] When the control strategy is marked as the first strategy mode, the incremental amplitude is compared with each threshold level in the amplitude gradient sequence to determine the amplitude level interval in the amplitude gradient sequence. The first direction adjustment step size corresponding to the amplitude level interval is used as the motion gain adjustment direction of the joint. The first direction adjustment step size is a positive number, representing the direction of gain enhancement. The first direction adjustment step size corresponding to each amplitude level interval increases with the level: the first level interval corresponds to a first step size of 0.1, the second level interval corresponds to a second step size of 0.25, the third level interval corresponds to a third step size of 0.45, and the fourth level interval corresponds to a fourth step size of 0.7.

[0050] When the control strategy is marked as the second strategy mode, the incremental amplitude is compared with each threshold level in the amplitude gradient sequence to determine the amplitude level interval in the amplitude gradient sequence. The second direction adjustment step size corresponding to the amplitude level interval is used as the motion gain adjustment direction of the joint. The second direction adjustment step size is negative, representing the gain suppression direction. The second direction adjustment step size corresponding to each amplitude level interval decreases as the level increases: the first level interval corresponds to a step size of -0.05, the second level interval corresponds to a second step size of -0.15, the third level interval corresponds to a third step size of -0.30, and the fourth level interval corresponds to a fourth step size of -0.50.

[0051] The motion gain adjustment direction is recorded as the gain adjustment direction label of the joint at the current moment. When the motion gain adjustment direction is positive, the gain adjustment direction label is recorded as an enhancement direction; when the motion gain adjustment direction is negative, the gain adjustment direction label is recorded as a suppression direction. The absolute value of the gain adjustment direction label represents the intensity of the gain adjustment; the larger the absolute value, the greater the adjustment intensity.

[0052] Step S104: Obtain the contact force signal and contact torque signal between the end effector of the slave robotic arm and the working environment, as well as the current joint angle of each joint of the slave robotic arm.

[0053] In this step, a six-dimensional force / torque sensor is first installed between the end effector of the slave robot and the wrist of the robot. The six-dimensional force / torque sensor is a strain gauge type, with three orthogonally arranged strain gauge bridges built in. These bridges measure the force components along the three orthogonal axes of the end effector coordinate system and the torque components around those axes. The range of the six-dimensional force / torque sensor is determined based on the rated load of the slave robot, with three times the rated load taken as the sensor's maximum range. The linearity error is less than 0.5% of full scale, and the resolution is less than 0.1% of full scale.

[0054] The six-dimensional force / torque sensor is mounted between the wrist flange of the slave robotic arm and the end effector. The sensor's sensitive axis direction is aligned with the direction of the end effector's coordinate system. The origin of the end effector's coordinate system is located at the geometric center of the end effector, and the three orthogonal axes are along the end effector's motion direction, lateral direction, and normal direction, respectively. The sensor is bolted to the wrist flange, and the end effector is bolted to the other side of the sensor. The sensor housing is marked with coordinate system orientation markings; during installation, ensure that the three sensitive axes of the sensor are aligned with the three axes of the end effector's coordinate system.

[0055] When the end effector of the robotic arm comes into contact with the working environment, the contact force is transmitted through the end effector to the elastic body of the six-dimensional force / torque sensor. The elastic body undergoes slight deformation, causing a change in the resistance value of the strain gauge bridge. This change in resistance is converted into a change in voltage signal by a Wheatstone bridge. The voltage signal is amplified and output to a data acquisition card, resulting in six analog voltage signals, corresponding to force components in three directions and torque components in three directions. These six analog voltage signals are continuously acquired at a preset sampling frequency of 2000Hz, and after analog-to-digital conversion, digitized contact force and contact torque signals are obtained.

[0056] The contact force signal includes force components along the X-axis, Y-axis, and Z-axis of the end effector coordinate system, forming a three-dimensional force vector. The contact torque signal includes torque components around the X-axis, Y-axis, and Z-axis of the end effector coordinate system, forming a three-dimensional torque vector.

[0057] Zero-point correction is performed on the acquired contact force and contact torque signals. The zero-point correction involves acquiring contact force and contact torque signals for a preset time period in a free-space state where the end effector of the robotic arm is not in contact with any environment. The arithmetic mean of the signals from each channel within the preset time period is calculated, and this arithmetic mean is used as the zero-point offset for each channel. During subsequent actual operation, the real-time measurement value of each channel is subtracted from the corresponding zero-point offset to obtain the corrected contact force and contact torque signals, eliminating the influence of sensor zero drift and temperature drift.

[0058] The acquired contact force and contact torque signals are subjected to low-pass filtering to remove high-frequency noise components. The cutoff frequency of the low-pass filter is 100Hz, and a second-order Butterworth low-pass filter is used to filter out high-frequency noise caused by mechanical vibration and environmental electromagnetic interference while retaining the dynamic change information of the contact force.

[0059] Simultaneously, the current joint angle of each joint is read from the joint controller of each joint of the slave-end robotic arm. Each joint of the slave-end robotic arm is equipped with an absolute encoder with a resolution of 0.001 degrees and a measurement range of 0 to 360 degrees. The joint controller reads the current angle value of each encoder at a preset sampling frequency of 1000Hz and stores the current angle value in the register of the joint controller. The current joint angle value is read from each joint controller through the communication interface to obtain the joint angle of each joint at the current moment, in radians. When reading the current joint angle, the sampling time of the six-dimensional force / torque sensor is used as the time reference, and the joint angle value in the joint controller whose timestamp is closest to the sampling time is taken as the current joint angle at that moment, ensuring that the joint angle signal and the contact force / torque signal are aligned on the time axis.

[0060] The corrected contact force signal, the corrected contact torque signal, and the current joint angle are output as input data for step S105.

[0061] Step S105: Calculate the deviation direction between the actual motion trajectory and the desired motion trajectory of the end effector of the slave robotic arm based on the contact force signal and the contact torque signal, and determine the load-motion coupling coefficient of each joint based on the deviation direction and the current joint angle.

[0062] In this step, the actual position sequence and expected position sequence of the end effector of the slave robot arm within a preset time window are obtained. Taking the start time of the preset time window as the reference zero point, the actual displacement vector of the last position point in the actual position sequence relative to the reference zero point is calculated, and the expected displacement vector of the last position point in the expected position sequence relative to the reference zero point is calculated. Calculate the angle between the actual displacement vector and the expected displacement vector, and use the sign of the angle as the deviation direction label. When the deviation direction label is positive, it indicates that the actual displacement direction leads the expected displacement direction. When the deviation direction label is negative, it indicates that the actual displacement direction lags behind the expected displacement direction. The force transmission Jacobian matrix from each joint of the slave-end robotic arm to the end effector is calculated based on the current joint angle. The contact force signal and the contact torque signal are used as inputs, and the inverse matrix of the force transmission Jacobian matrix is ​​used as the transformation matrix to map the contact force signal and the contact torque signal to each joint of the slave-end robotic arm to obtain the equivalent contact torque of each joint. When the equivalent contact torque is positive, it indicates that the contact force applies a positive driving torque to the joint. When the equivalent contact torque is negative, it indicates that the contact force applies a reverse resistance torque to the joint. Multiply the equivalent contact torque of each joint by the deviation direction label, and use the product as the preliminary coupling coefficient of each joint. When the preliminary coupling coefficient is positive, it indicates that the actual movement direction of the joint under the current load condition is consistent with the desired movement direction and is aided by the environment. When the preliminary coupling coefficient is negative, it indicates that the actual movement direction of the joint under the current load condition is opposite to the desired movement direction and is resisted by the environment. The initial coupling coefficients are subjected to a time window moving average process, and the result of the moving average process is used as the load-motion coupling coefficient of each joint.

[0063] In one specific embodiment, the actual position sequence and expected position sequence of the end effector of the slave robot are first obtained within a preset time window. The length of the preset time window is determined based on the motion response time of the slave robot, and is set to 200 milliseconds, corresponding to 200 sampling points. The actual position sequence is the set of the actual three-dimensional spatial coordinates of the end effector in the world coordinate system at each sampling moment within the preset time window, calculated by the forward kinematics model of the slave robot based on the current joint angle of each joint. The expected position sequence is the set of the expected three-dimensional spatial coordinates of the end effector in the world coordinate system at each sampling moment within the preset time window, calculated by the joint angle increment command of each joint through the forward kinematics model. The joint angle increment command is the increment command actually received by the slave robot at each sampling moment before the current moment.

[0064] Using the start time of the preset time window as a reference zero point, the actual displacement vector is obtained by subtracting the three-dimensional spatial coordinates of the actual position point at the reference zero point from the three-dimensional spatial coordinates of the last position point in the actual position sequence. The actual displacement vector is a three-dimensional vector containing displacement components along the X, Y, and Z axes of the world coordinate system, representing the actual spatial displacement of the end effector from the starting position to the final position within the preset time window. Similarly, the expected displacement vector is obtained by subtracting the three-dimensional spatial coordinates of the expected position point at the reference zero point from the three-dimensional spatial coordinates of the last position point in the expected position sequence. The expected displacement vector is also a three-dimensional vector containing displacement components along the X, Y, and Z axes of the world coordinate system, representing the expected spatial displacement of the end effector from the starting position to the final position within the preset time window.

[0065] Calculate the angle between the actual displacement vector and the desired displacement vector. The angle is calculated as follows: calculate the dot product of the actual displacement vector and the desired displacement vector, divide by the product of the magnitudes of the actual and desired displacement vectors, and take the inverse cosine of the quotient to obtain the angle in radians. The sign of the angle is used as a deviation direction label. When the angle is positive, the deviation direction label is positive, indicating that the actual displacement direction leads the desired displacement direction, meaning the actual movement direction of the end effector angularly leads the desired movement direction. When the angle is negative, the deviation direction label is negative, indicating that the actual displacement direction lags behind the desired displacement direction, meaning the actual movement direction of the end effector angularly lags behind the desired movement direction. When the magnitude of either the actual displacement vector or the desired displacement vector is zero, the deviation direction label is set to zero.

[0066] The force transmission Jacobian matrix from each joint of the slave manipulator to the end effector is calculated based on the current joint angles. The force transmission Jacobian matrix is ​​an m×n matrix, where m is the number of degrees of freedom of the end effector and n is the number of joints in the slave manipulator. Each column of the force transmission Jacobian matrix corresponds to a joint, and each row corresponds to a degree of freedom direction of the end effector in the world coordinate system. The elements of the force transmission Jacobian matrix are the partial derivatives of each joint angle with respect to the position and orientation of the end effector.

[0067] Using the contact force signal and the contact torque signal as inputs, the contact force signal contains a three-dimensional force vector, and the contact torque signal contains a three-dimensional torque vector. The two are combined to form a six-dimensional force vector for the end effector. Using the inverse matrix of the force transfer Jacobian matrix as the transformation matrix, the six-dimensional force vector composed of the contact force signal and the contact torque signal is mapped to the joint space of each joint of the slave-end manipulator, obtaining the equivalent contact torque for each joint. The equivalent contact torque is a scalar, with units of Newton-meters. When the equivalent contact torque is positive, it indicates that the contact force applies a positive driving torque to the joint, meaning the direction of the contact force is consistent with the positive direction of the joint, assisting the joint to move in the positive direction. When the equivalent contact torque is negative, it indicates that the contact force applies a negative resisting torque to the joint, meaning the direction of the contact force is opposite to the positive direction of the joint, hindering the joint from moving in the positive direction.

[0068] The equivalent contact torque of each joint is multiplied by the deviation direction label, and the product is used as the initial coupling coefficient for each joint. When both the deviation direction label and the equivalent contact torque are positive, the initial coupling coefficient is positive, indicating that the joint's actual movement direction under the current load condition is consistent with the desired movement direction and is aided by environmental forces, meaning the environmental force accelerates the joint's movement speed. When both the deviation direction label and the equivalent contact torque are negative, the initial coupling coefficient is positive, indicating that the joint's actual movement direction under the current load condition lags behind the desired movement direction and is subject to environmental resistance; the environmental force is opposite to the movement direction, and the joint needs to output a larger driving force to overcome the resistance. When both the deviation direction label and the equivalent contact torque are negative, the initial coupling coefficient is negative, indicating that the joint's actual movement direction under the current load condition is ahead of the desired movement but is subject to environmental resistance; the environmental force hinders the joint's movement, and the joint needs to output a larger driving force to maintain movement. When the deviation direction label is negative and the equivalent contact torque is positive, the initial coupling coefficient is negative, indicating that the joint's actual movement direction is lagging under the current load conditions but is aided by the environment. The environmental force drives the joint to move, and the joint needs to reduce its output torque to avoid excessive speed.

[0069] A time-window moving average is applied to the initial coupling coefficients. The window length for the moving average is 100 milliseconds, or 100 sampling points. Using the current time as the end point of the window, the initial coupling coefficient values ​​corresponding to the 100 sampling points prior to the current time are taken, and the arithmetic mean of these 100 initial coupling coefficient values ​​is calculated. This arithmetic mean is used as the load-motion coupling coefficient for each joint at the current time. This moving average process eliminates high-frequency fluctuations in the initial coupling coefficients, allowing the load-motion coupling coefficients to change smoothly over time.

[0070] Furthermore, the window length of the moving average processing is related to the dynamic response time of the slave-end manipulator. The joint servo speed loop bandwidth of the slave-end manipulator is typically 10Hz to 20Hz, corresponding to a main dynamic response time of approximately 50ms to 100ms; simultaneously, after the contact force signal is low-pass filtered at 100Hz, its main dynamic component period is greater than 10ms. Setting the moving average window length to 100ms, i.e., 100 sampling points at a sampling frequency of 1000Hz, can match the main dynamic response time scale of the slave-end manipulator. When the window is too short, high-frequency noise and contact force jitter in the initial coupling coefficient are difficult to filter out sufficiently; when the window is too long, it will introduce a large phase lag, reducing the real-time response capability of the load-motion coupling coefficient to changes in environmental assistance / resistance. Therefore, a 100ms window can achieve a balance between smoothing effect and real-time performance, making the load-motion coupling coefficient of each joint both stable and able to track the interaction state of the slave-end environment in a timely manner.

[0071] Step S106: Determine the incremental mapping ratio coefficient of each joint by combining the motion gain adjustment direction of each joint with the load-motion coupling coefficient of each joint.

[0072] In this step, the direction of joint motion gain adjustment and the load-motion coupling coefficient are obtained; Obtain the preset basic mapping ratio coefficient of the joint, which is the default mapping ratio coefficient of the joint when there is no load and the operator is in a moderate motion intention. When the motion gain adjustment direction is the enhancement direction, the absolute value of the load-motion coupling coefficient is added to the preset enhancement adjustment coefficient to obtain the first proportional factor, and the product of the preset basic mapping proportional coefficient and the first proportional factor is used as the enhanced mapping proportional coefficient of the joint. When the motion gain adjustment direction is the suppression direction, the absolute value of the load-motion coupling coefficient is subtracted from the preset suppression adjustment coefficient to obtain the second proportional factor, and the product of the preset basic mapping proportional factor and the second proportional factor is used as the suppressed mapping proportional factor of the joint. The enhanced post-mapping scaling factor or the suppressed post-mapping scaling factor is used as the incremental mapping scaling factor of the joint.

[0073] In one specific embodiment, the motion gain adjustment direction of a certain joint determined in step S103 and the load-motion coupling coefficient of the same joint determined in step S105 are first obtained. The motion gain adjustment direction is either an enhancement direction or a suppression direction, and the load-motion coupling coefficient is a real number.

[0074] Obtain the preset base mapping ratio coefficient for the joint. This preset base mapping ratio coefficient is the default mapping ratio coefficient for the joint when there is no external load and the operator is in a moderate movement intention state. This coefficient is pre-calibrated based on the range of motion ratio between the slave robotic arm and the operator's upper limb. For the shoulder joint, the preset base mapping ratio coefficient is 0.8, indicating that for every 1 degree rotation of the operator's shoulder joint, the slave robotic arm's shoulder joint rotates 0.8 degrees. For the elbow joint, the preset base mapping ratio coefficient is 1.0, indicating that for every 1 degree rotation of the operator's elbow joint, the slave robotic arm's elbow joint rotates 1.0 degrees. For the wrist joint, the preset base mapping ratio coefficient is 1.2, indicating that for every 1 degree rotation of the operator's wrist joint, the slave robotic arm's wrist joint rotates 1.2 degrees.

[0075] Determine the label of the motion gain adjustment direction. When the motion gain adjustment direction is an enhancement direction, execute the enhancement direction mapping ratio calculation process; when the motion gain adjustment direction is a suppression direction, execute the suppression direction mapping ratio calculation process.

[0076] When the motion gain adjustment direction is the enhancement direction, the absolute value of the load-motion coupling coefficient is added to a preset enhancement adjustment coefficient to obtain a first scaling factor. The preset enhancement adjustment coefficient is a constant greater than 1, with a value of 1.2. The first scaling factor is the sum of the preset enhancement adjustment coefficient and the absolute value of the load-motion coupling coefficient, and its value ranges from greater than 1 to less than or equal to 2.2. When the absolute value of the load-motion coupling coefficient increases, the first scaling factor increases accordingly, indicating that the joint has greater environmental resistance under the current load conditions, and the mapping ratio needs a greater enhancement to overcome the resistance. When the absolute value of the load-motion coupling coefficient is zero, the first scaling factor is equal to 1.2, indicating that the mapping ratio is slightly enhanced when there is no environmental interaction.

[0077] The preset base mapping scaling factor is multiplied by the first scaling factor, and the product is used as the enhanced post-mapping scaling factor for the joint. If the enhanced post-mapping scaling factor is greater than the preset base mapping scaling factor, it indicates that the joint requires increased slave-end response gain under the current load conditions.

[0078] When the motion gain adjustment direction is the suppression direction, the absolute value of the load-motion coupling coefficient is subtracted from the preset suppression adjustment coefficient to obtain the second scaling factor. The preset suppression adjustment coefficient is a positive number less than 1, with a value of 0.7. The second scaling factor is the difference between the absolute value of the load-motion coupling coefficient and the preset suppression adjustment coefficient, and its value ranges from greater than or equal to 0.1 to less than 1. When the absolute value of the load-motion coupling coefficient increases, the second scaling factor decreases accordingly, indicating that the greater the environmental resistance of the joint under the current load conditions, the greater the suppression required for the mapping ratio to avoid collision between the slave end and the environment. When the absolute value of the load-motion coupling coefficient is zero, the second scaling factor is equal to 0.7, indicating that the mapping ratio is slightly suppressed when there is no environmental interaction.

[0079] The preset base mapping scaling factor is multiplied by the second scaling factor, and the product is used as the suppressed mapping scaling factor for that joint. If the suppressed mapping scaling factor is less than the preset base mapping scaling factor, it indicates that the joint needs to reduce the slave-end response gain under the current load conditions.

[0080] The enhanced mapping ratio or the suppressed mapping ratio is output as the incremental mapping ratio of the joint. When the motion gain adjustment direction is the enhancement direction, the incremental mapping ratio is greater than the preset base mapping ratio; when the motion gain adjustment direction is the suppression direction, the incremental mapping ratio is less than the preset base mapping ratio.

[0081] The incremental mapping ratio is limited to a preset safe mapping ratio range. The lower limit of the safe mapping ratio range is 0.2 times the preset base mapping ratio, and the upper limit of the safe mapping ratio range is 2.5 times the preset base mapping ratio. When the calculated enhanced mapping ratio is greater than the upper limit, the upper limit is used as the incremental mapping ratio of the joint; when the calculated suppressed mapping ratio is less than the lower limit, the lower limit is used as the incremental mapping ratio of the joint.

[0082] Step S107: The joint angle increment signal is scaled using the incremental mapping scaling coefficient of each joint to obtain the angle increment command of each joint of the slave robotic arm, and the angle increment command is output to the joint controller of the slave robotic arm for execution.

[0083] In this step, the incremental mapping ratio coefficients of each joint are obtained, and an incremental mapping ratio matrix is ​​constructed using the incremental mapping ratio coefficients of each joint as diagonal elements. The incremental mapping ratio matrix is ​​a diagonal matrix, and all off-diagonal elements are zero. The joint angle increment signals of each joint are arranged in the joint order to obtain the joint angle increment vector; Multiply the incremental mapping ratio matrix with the joint angle increment vector, and use the resulting vector as the initial angle increment command for each joint. For each joint component in the initial angle increment command, obtain the current angle value of the joint and the upper and lower limits of the joint angle safety range. Calculate the predicted angle value after adding the joint component to the current angle value. When the predicted angle value exceeds the upper or lower limit of the joint angle safety range, reduce the joint component proportionally to make the predicted angle value equal to the upper or lower limit of the safety range, and obtain the limited angle increment component corresponding to the joint. Traverse all joints to obtain the limited angle increment components corresponding to each joint, and combine them into a limited angle increment command vector. The angle increment command vector after the amplitude limit is used as the final angle increment command for each joint of the slave robotic arm. The final angle increment command is output to the joint controller of each joint of the slave robotic arm, so that each joint performs movement according to the final angle increment command.

[0084] In one specific embodiment, the incremental mapping ratio coefficients for each joint determined in step S106 are first obtained. The number of incremental mapping ratio coefficients is equal to the total number of joints in the operator's upper limb, with one incremental mapping ratio coefficient corresponding to each joint.

[0085] An incremental mapping ratio matrix is ​​constructed using the incremental mapping ratio coefficients of each joint as diagonal elements. The incremental mapping ratio matrix is ​​an n×n diagonal matrix, where n is the total number of joints in the operator's upper limb. The element in the i-th row and i-th column of the incremental mapping ratio matrix equals the incremental mapping ratio coefficient of the i-th joint, and all off-diagonal elements in the matrix are zero. When n equals 3, the incremental mapping ratio matrix is ​​a three-dimensional square matrix, with the incremental mapping ratio coefficients of the shoulder, elbow, and wrist joints arranged sequentially on the diagonal.

[0086] The joint angle increment signals of each joint at the current moment are arranged in joint order to obtain a joint angle increment vector. The dimension of the joint angle increment vector is n×1, and its i-th element is equal to the value of the joint angle increment signal of the i-th joint at the current moment. The joint angle increment signal is the sampled value of the displacement signal obtained in step S101 at the current moment, and the unit is radians.

[0087] The incremental mapping ratio matrix is ​​multiplied by the joint angle increment vector, performing a matrix multiplication operation. The result of the matrix multiplication operation is an n×1 vector, where the i-th element equals the incremental mapping ratio coefficient of the i-th joint multiplied by the joint angle increment signal value of the i-th joint. This resulting vector is used as the initial angle increment command for each joint. Each element in the initial angle increment command corresponds to an angle increment command value for a joint, in radians.

[0088] For each joint component in the initial angle increment command, the current angle value and the upper and lower limits of the joint angle safety range are obtained. The current angle value is determined by the joint angle at the current moment read from the joint controller in step S104. The upper and lower limits of the joint angle safety range are determined by the mechanical limits of each joint of the slave-end robotic arm, with the upper limit being the maximum angle value allowed for the joint and the lower limit being the minimum angle value allowed for the joint.

[0089] Calculate the predicted angle value by adding the joint component from the initial angle increment command to the current angle value. When the predicted angle value is greater than the upper limit of the joint angle safety range, the joint component is scaled down proportionally by a scaling factor equal to the upper limit of the joint angle safety range minus the current angle value divided by the joint component, so that the predicted angle value equals the upper limit of the safety range. When the predicted angle value is less than the lower limit of the joint angle safety range, the joint component is scaled down proportionally by a scaling factor equal to the lower limit of the joint angle safety range minus the current angle value divided by the joint component, so that the predicted angle value equals the lower limit of the safety range. When the predicted angle value is between the upper and lower limits of the joint angle safety range, the joint component remains unchanged.

[0090] The process iterates through all joints, performing the aforementioned limiting process on each joint component to obtain the limited angle increment component corresponding to each joint. These limited angle increment components are then arranged in joint order and combined to form a limited angle increment command vector. The limited angle increment command vector has a dimension of n×1, and each element lies within the safe angle increment range of the corresponding joint.

[0091] The limited-angle increment command vector is used as the final angle increment command for each joint of the slave-end robotic arm. The command values ​​for each joint in the final angle increment command are sent to the joint controllers of each joint of the slave-end robotic arm via a communication interface. The communication interface is an EtherCAT, CAN, or Ethernet interface, and the communication cycle is consistent with the preset sampling frequency. After receiving the final angle increment command, each joint controller uses the command value as the angle increment target value for the current control cycle and drives each joint motor to perform movement according to the angle increment target value.

[0092] In summary, the method of this application acquires the incremental joint angle signals and surface electromyography (EMG) signals of each joint of the operator's upper limb; extracts the activation sequence of each muscle group from the EMG signals to determine the muscle coordination lag; determines the motion gain adjustment direction of each joint based on the incremental direction and amplitude of the incremental joint angle signals, combined with the muscle coordination lag; acquires the contact force / torque signals of the end effector of the slave robotic arm and the current angle of each joint; calculates the deviation direction between the actual motion trajectory and the desired motion trajectory of the end effector based on the contact force / torque signals, and determines the load-motion coupling coefficient of each joint based on the current joint angle; determines the incremental mapping ratio coefficient of each joint by combining the motion gain adjustment direction and the load-motion coupling coefficient; scales the incremental joint angle signals using the incremental mapping ratio coefficient to obtain the angle increment commands of each joint of the slave robotic arm and outputs them for execution. Adaptive adjustment of the mapping ratio coefficient is achieved using the operator's physiological signals and the interaction force of the slave environment.

[0093] The experimental platform employs a homogeneous human-machine interface with an upper limb master-slave operating system. The master end is equipped with an accelerometer and surface electromyography electrodes, while the slave end utilizes a six-degree-of-freedom collaborative robotic arm. A six-dimensional force / torque sensor is installed between the wrist and the end effector. Ten healthy operators were selected to perform three types of tasks: grasping-placement, trajectory tracking, and confined space contact. Comparisons included existing fixed mapping ratio methods, existing manually adjusted mapping ratio methods, and the method described in this application. Evaluation metrics included trajectory tracking error, task completion time, peak end-effector contact force, overshoot, and performance standard deviation among different operators.

[0094] Exemplary test results are as follows: In the grasp-place task, the average trajectory tracking error of the fixed mapping ratio method is 8.6 mm, the task completion time is 12.4 s, the peak end contact force is 18.5 N, and the overshoot is 12.5%; the average trajectory tracking error of the manually adjusted mapping ratio method is 7.9 mm, the task completion time is 11.2 s, the peak end contact force is 16.2 N, and the overshoot is 10.8%; the average trajectory tracking error of the method in this application is 5.2 mm, the task completion time is 9.1 s, the peak end contact force is 11.3 N, and the overshoot is 6.2%. Compared with the fixed mapping ratio method, the method in this application reduces the trajectory tracking error by approximately 39.5%, shortens the task completion time by approximately 26.6%, reduces the peak end contact force by approximately 38.9%, and reduces the overshoot by approximately 50.4%. Regarding adaptability to different operators, the standard deviation of task completion time for the fixed mapping ratio method is 2.8 s, and the standard deviation of trajectory tracking error is 1.2 mm. The standard deviation of task completion time for the method described in this application is reduced to 1.6 s, and the standard deviation of trajectory tracking error is reduced to 0.7 mm. These data indicate that the method described in this application outperforms existing fixed mapping and manual adjustment mapping schemes in terms of operational accuracy, response speed, and adaptability to different operators. It can adaptively adjust the mapping ratio coefficient by utilizing operator physiological signals and the interaction force of the slave environment, achieving better technical results.

[0095] The reduction in trajectory tracking error is due to the fact that this scheme extracts the activation sequence of each muscle group from surface electromyography (EMG) signals and calculates the muscle coordination hysteresis based on the difference between co-activation duration and alternating activation duration. This muscle coordination hysteresis depends only on the temporal structure characteristics of the activation window and not on the absolute value of the EMG amplitude, thus enabling a stable distinction between the operator's current speed-priority mode and precision-priority mode. When the operator is in precision-priority mode, a larger joint angle increment corresponds to an adjustment step size in the inhibition direction, reducing the incremental mapping proportional coefficient and allowing the slave robotic arm to respond smoothly, avoiding trajectory overshoot and repeated corrections due to excessive mapping gain. When the operator is in speed-priority mode, a larger joint angle increment corresponds to an adjustment step size in the enhancement direction, enabling the slave to respond quickly to the operator's intentions. Therefore, this scheme achieves a smaller trajectory tracking error than fixed mapping and manual adjustment under different movement intentions.

[0096] The reduced task completion time is due to the fact that this solution determines the direction of motion gain adjustment based on muscle coordination lag and combines this with the load-motion coupling coefficient to determine the incremental mapping ratio. In speed-priority mode, the mapping ratio is appropriately enhanced, allowing the slave robotic arm to more quickly follow the operator's large-amplitude joint angle increments. When the slave receives environmental assistance, the load-motion coupling coefficient adjusts the mapping ratio in a favorable direction, thus utilizing environmental forces to assist movement. When the slave experiences environmental resistance, the mapping ratio is appropriately suppressed to avoid violent collisions and repeated adjustments between the slave and the environment. Compared to manual adjustment, this solution eliminates the need for the operator to be distracted by switching switches or manually adjusting the gain; the adjustment process is automatically completed by physiological signals and interactive forces, thus shortening the task completion time.

[0097] The reduction in peak end-effector contact force is due to the fact that this solution acquires the contact force and torque signals between the end effector of the robotic arm and the working environment, and determines the load-motion coupling coefficient of each joint based on the deviation direction between the actual and desired motion trajectories and the current joint angle. This coefficient reflects whether the joint is in an environment-assisted or environment-resisted state under the current load conditions. When the actual motion direction is ahead and subject to environmental resistance, this solution reduces the joint motion speed by suppressing the mapping ratio, thus reducing the impact of the end effector on the environment. When the actual motion direction is lagging and subject to environmental resistance, this solution increases the joint driving force by enhancing the mapping ratio, helping to overcome the resistance. Simultaneously, a time-window moving average is applied to the initial coupling coefficient to smooth the load-motion coupling coefficient changes, avoiding drastic fluctuations in the mapping ratio caused by sudden changes in contact force. Therefore, the peak end-effector contact force is significantly reduced.

[0098] The reduction in overshoot is due to the fact that this scheme employs a gain adjustment strategy in the opposite direction to the speed-priority mode in the precision-priority mode: the larger the increment amplitude, the larger the adjustment step size in the suppression direction, and the smaller the increment mapping ratio coefficient, thereby avoiding overshoot caused by excessively fast response from the slave robot during precise operations. Furthermore, before outputting the angle increment command, this scheme performs safety range limiting processing on each joint component. When the predicted angle value exceeds the upper or lower limit of the joint angle safety range, the joint component is proportionally reduced, ensuring that the slave robot always moves within the safe angle range. The combined effect of these two mechanisms significantly reduces overshoot.

[0099] The improved adaptability among different operators stems from the fact that this scheme determines muscle coordination lag by using only the start and end times of the activation windows for each muscle group, as well as the duration of simultaneous and alternating activation states of both muscle groups, without relying on the absolute amplitude of the electromyographic (EMG) signal. Even with significant differences in EMG amplitude among different operators, the calculated muscle coordination lag remains consistent as long as the activation sequence structures of the agonist and antagonist muscles are similar. Furthermore, the load-motion coupling coefficient is derived from the six-dimensional force / torque sensor at the slave end and the current angle of each joint, and is independent of individual physiological differences among operators. Therefore, the standard deviation of task completion time and trajectory tracking error among different operators in this scheme are both lower than those of the fixed mapping ratio method, demonstrating better operator adaptability.

[0100] The above data shows that this method determines muscle coordination lag by analyzing the muscle activation sequence in the operator's surface electromyography (EMG) signals, determines the incremental direction and amplitude by analyzing the incremental joint angle signals, and combines the end-contact force / torque signals and the current joint angle to determine the load-motion coupling coefficient. Finally, it jointly determines the incremental mapping ratio coefficient for each joint. This technical approach outperforms existing fixed mapping and manual adjustment mapping schemes in terms of operational accuracy, response speed, contact safety, and adaptability to different operators, achieving better technical results. The above experimental data are illustrative and not intended to limit the scope of protection of this method.

[0101] Please see Figure 2 The diagram shows a structural block diagram of a homogeneous human-machine upper limb master-slave incremental mapping system according to this application.

[0102] like Figure 2 As shown, the isomorphic human-machine upper limb master-slave incremental mapping system 200 includes a first acquisition module 210, a first determination module 220, a second determination module 230, a second acquisition module 240, a calculation module 250, a third determination module 260, and an output module 270.

[0103] The system includes a first acquisition module 210 configured to acquire incremental joint angle signals of each joint of the operator's upper limb and the operator's surface electromyography (EMG) signals; a first determination module 220 configured to extract the activation sequence of each muscle group from the EMG signals and determine the muscle coordination lag of the operator's current joint based on the activation sequence; a second determination module 230 configured to determine the motion gain adjustment direction of each joint based on the incremental direction and amplitude of each joint in the incremental joint angle signals, combined with the muscle coordination lag; and a second acquisition module 240 configured to acquire contact force signals and contact torque signals between the end effector of the slave robotic arm and the working environment, as well as the current joint angles of each joint of the slave robotic arm; and to calculate... Module 250 is configured to calculate the deviation direction between the actual motion trajectory and the desired motion trajectory of the end effector of the slave robotic arm based on the contact force signal and the contact torque signal, and determine the load-motion coupling coefficient of each joint based on the deviation direction and the current joint angle; the third determination module 260 is configured to jointly determine the incremental mapping ratio coefficient of each joint by the motion gain adjustment direction of each joint and the load-motion coupling coefficient of each joint; the output module 270 is configured to scale the joint angle increment signal using the incremental mapping ratio coefficient of each joint to obtain the angle increment command of each joint of the slave robotic arm, and output the angle increment command to the joint controller of the slave robotic arm for execution.

[0104] It should be understood that Figure 2 The modules and references described in the document Figure 1 The steps described in the text correspond to those in the method described above. Therefore, the operations, features, and corresponding technical effects described above also apply to the method described in the text. Figure 2 The various modules in the document will not be described in detail here.

[0105] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, causes the processor to perform the isomorphic human-machine upper limb master-slave incremental mapping method in any of the above method embodiments. In one embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, which are configured as follows: Acquire the incremental joint angle signals of each joint of the operator's upper limb and the operator's surface electromyography signals; The activation sequence of each muscle group is extracted from the surface electromyography signal, and the muscle coordination lag of the operator's current joint is determined based on the activation sequence. Based on the incremental direction and amplitude of each joint in the incremental joint angle signal, and combined with the muscle coordination hysteresis, the direction of motion gain adjustment for each joint is determined. Acquire the contact force signal and contact torque signal between the end effector of the slave robotic arm and the working environment, as well as the current joint angle of each joint of the slave robotic arm; The deviation direction between the actual motion trajectory and the desired motion trajectory of the end effector of the slave robotic arm is calculated based on the contact force signal and the contact torque signal. The load-motion coupling coefficient of each joint is determined based on the deviation direction and the current joint angle. The incremental mapping ratio coefficient of each joint is determined by the motion gain adjustment direction of each joint and the load-motion coupling coefficient of each joint. The incremental angle signal of each joint is scaled using the incremental mapping scaling coefficient of each joint to obtain the angle increment command of each joint of the slave robotic arm, and the angle increment command is output to the joint controller of the slave robotic arm for execution.

[0106] Computer-readable storage media may include a stored program area and a stored data area, wherein the stored program area may store an operating system and an application program required for at least one function; the stored data area may store data created based on the use of the isomorphic human-machine upper limb master-slave incremental mapping system, etc. Furthermore, the computer-readable storage medium may include high-speed random access memory, and may also include memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include memory remotely disposed relative to a processor, which can be connected to the isomorphic human-machine upper limb master-slave incremental mapping system via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0107] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 3 As shown, the device includes a processor 310 and a memory 320. The electronic device may also include an input device 330 and an output device 340. The processor 310, memory 320, input device 330, and output device 340 can be connected via a bus or other means. Figure 3 Taking a bus connection as an example, the memory 320 is the computer-readable storage medium described above. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby implementing the isomorphic human-machine interface upper limb master-slave incremental mapping method described in the above embodiment. The input device 330 can receive input digital or character information and generate key signal inputs related to user settings and function control of the isomorphic human-machine interface upper limb master-slave incremental mapping system. The output device 340 may include a display screen or other display device.

[0108] The aforementioned electronic device can execute the method provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in the embodiments of the present invention.

[0109] In one implementation, the above-described electronic device is applied to a homogeneous human-machine upper limb master-slave incremental mapping system as a client, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Acquire the incremental joint angle signals of each joint of the operator's upper limb and the operator's surface electromyography signals; The activation sequence of each muscle group is extracted from the surface electromyography signal, and the muscle coordination lag of the operator's current joint is determined based on the activation sequence. Based on the incremental direction and amplitude of each joint in the incremental joint angle signal, and combined with the muscle coordination hysteresis, the direction of motion gain adjustment for each joint is determined. Acquire the contact force signal and contact torque signal between the end effector of the slave robotic arm and the working environment, as well as the current joint angle of each joint of the slave robotic arm; The deviation direction between the actual motion trajectory and the desired motion trajectory of the end effector of the slave robotic arm is calculated based on the contact force signal and the contact torque signal. The load-motion coupling coefficient of each joint is determined based on the deviation direction and the current joint angle. The incremental mapping ratio coefficient of each joint is determined by the motion gain adjustment direction of each joint and the load-motion coupling coefficient of each joint. The incremental angle signal of each joint is scaled using the incremental mapping scaling coefficient of each joint to obtain the angle increment command of each joint of the slave robotic arm, and the angle increment command is output to the joint controller of the slave robotic arm for execution.

[0110] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0111] 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 of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A master-slave incremental mapping method for isomorphic human-machine upper limbs, characterized in that, include: Acquire the incremental joint angle signals of each joint of the operator's upper limb and the operator's surface electromyography signals; The activation sequence of each muscle group is extracted from the surface electromyography signal, and the muscle coordination lag of the operator's current joint is determined based on the activation sequence. Based on the incremental direction and amplitude of each joint in the incremental joint angle signal, and combined with the muscle coordination hysteresis, the direction of motion gain adjustment for each joint is determined. Acquire the contact force signal and contact torque signal between the end effector of the slave robotic arm and the working environment, as well as the current joint angle of each joint of the slave robotic arm; The deviation direction between the actual motion trajectory and the desired motion trajectory of the end effector of the slave robotic arm is calculated based on the contact force signal and the contact torque signal. The load-motion coupling coefficient of each joint is determined based on the deviation direction and the current joint angle. The incremental mapping ratio coefficient of each joint is determined by the motion gain adjustment direction of each joint and the load-motion coupling coefficient of each joint. The incremental angle signal of each joint is scaled using the incremental mapping scaling coefficient of each joint to obtain the angle increment command of each joint of the slave robotic arm, and the angle increment command is output to the joint controller of the slave robotic arm for execution.

2. The isomorphic human-machine upper limb master-slave incremental mapping method according to claim 1, characterized in that, The acquisition of the joint angle increment signals of each joint of the operator's upper limb and the operator's surface electromyography signals includes: Accelerometers are installed at each joint of the operator's upper limbs to continuously collect acceleration signals in the direction of joint movement at a preset sampling frequency; Perform an integral operation on the acceleration signal to obtain a velocity signal, perform an integral operation on the velocity signal to obtain a displacement signal, and use the displacement signal as the joint angle increment signal; An electrode array is installed on the skin surface of the relevant muscle groups of the operator, and the electromyographic signals at each electrode are synchronously collected at the preset sampling frequency to obtain the original electromyographic signals of each muscle group. The original electromyographic (EMG) signals are subjected to bandpass filtering and rectification smoothing to obtain preprocessed EMG signals for each muscle group, and the preprocessed EMG signals are used as the surface EMG signals.

3. The isomorphic human-machine upper limb master-slave incremental mapping method according to claim 1, characterized in that, The step of extracting the activation sequence of each muscle group from the surface electromyography signal and determining the muscle coordination lag of the operator's current joint based on the activation sequence includes: Acquire surface electromyography (EMG) signals of each muscle group, slide a preset sliding window over the surface EMG signals of each muscle group, calculate the short-time energy value of the EMG signal within the window at each window position, and obtain the short-time energy sequence corresponding to each muscle group. The baseline value is extracted from the short-time energy sequence, and the baseline value plus a preset multiple of the baseline standard deviation is used as the activation threshold of the muscle group. The starting position where the short-time energy values ​​of a consecutive preset number of sampling points in the short-time energy sequence all exceed the activation threshold is used as the activation start time of the muscle group. The starting position where the short-time energy values ​​of a consecutive preset number of sampling points in the short-time energy sequence all fall below the activation threshold is used as the activation end time of the muscle group. The time interval between the activation start time and the activation end time is used as the activation window of the muscle group. For a certain joint, the activation window of the first muscle group that drives the joint to move in a first direction is obtained as the first activation window, and the activation window of the second muscle group that drives the joint to move in a second direction is obtained as the second activation window, wherein the first direction is opposite to the second direction. The earliest of the start times of the first activation window and the second activation window is taken as the muscle control start time of the joint, and the later of the end times of the first activation window and the second activation window is taken as the muscle control end time of the joint. The time interval between the muscle control start time and the muscle control end time is taken as the joint control window. Within the joint control window, starting from the moment the muscle control begins, the first activation window and the second activation window are marked sequentially along the time axis at each moment, indicating whether the first activation window and the second activation window are in an active or inactive state, thus obtaining the state sequence corresponding to the joint control window. The state type at each moment in the state sequence includes one of four states: only the first muscle group is activated, only the second muscle group is activated, both muscle groups are activated simultaneously, and neither muscle group is activated. The total duration of simultaneous activation of two muscle groups in the state sequence is taken as the co-activation duration, and the total duration of activation of only the first muscle group and activation of only the second muscle group in the state sequence is taken as the alternating activation duration. The muscle coordination lag of the joint is obtained by dividing the difference between the co-activation duration and the alternating activation duration by the total duration of the joint control window.

4. The isomorphic human-machine upper limb master-slave incremental mapping method according to claim 1, characterized in that, The step of determining the motion gain adjustment direction of each joint based on the incremental direction and amplitude of each joint in the incremental joint angle signal, combined with the muscle coordination hysteresis, includes: Obtain the sign of the joint angle increment signal of a certain joint at the current moment, use the sign as the increment direction of the joint, and use the absolute value of the joint angle increment signal as the increment amplitude of the joint. The muscle coordination lag of the joint is obtained, and the muscle coordination lag is compared with a preset lag threshold. When the muscle coordination lag is greater than the lag threshold, the control strategy of the joint is marked as a first strategy mode. When the muscle coordination lag is not greater than the lag threshold, the control strategy of the joint is marked as a second strategy mode. The first strategy mode is a speed priority mode, and the second strategy mode is a precision priority mode. When the control strategy is marked as the first strategy mode, the incremental amplitude is compared with each level threshold in the preset amplitude gradient sequence to determine the amplitude level interval in the amplitude gradient sequence. The first direction adjustment step size corresponding to the amplitude level interval is used as the motion gain adjustment direction of the joint. Each amplitude threshold divides the value range of the incremental amplitude into multiple amplitude level intervals. The amplitude gradient sequence contains multiple increasing amplitude thresholds. When the control strategy is marked as the second strategy mode, the incremental amplitude is compared with the thresholds of each level in the amplitude gradient sequence one by one to determine the amplitude level interval in the amplitude gradient sequence. The second direction adjustment step size corresponding to the amplitude level interval is used as the motion gain adjustment direction of the joint. The sign of the second direction adjustment step size is opposite to the sign of the first direction adjustment step size. The motion gain adjustment direction is recorded as the gain adjustment direction label of the joint at the current moment, and the value of the gain adjustment direction label is either the enhancement direction or the suppression direction.

5. The isomorphic human-machine upper limb master-slave incremental mapping method according to claim 1, characterized in that, The step of calculating the deviation direction between the actual motion trajectory and the desired motion trajectory of the end effector of the slave robotic arm based on the contact force signal and the contact torque signal, and determining the load-motion coupling coefficient of each joint based on the deviation direction and the current joint angle, includes: The actual position sequence and expected position sequence of the end effector of the slave robot arm within a preset time window are obtained. Taking the start time of the preset time window as the reference zero point, the actual displacement vector of the last position point in the actual position sequence relative to the reference zero point is calculated, and the expected displacement vector of the last position point in the expected position sequence relative to the reference zero point is calculated. Calculate the angle between the actual displacement vector and the expected displacement vector, and use the sign of the angle as the deviation direction label. When the deviation direction label is positive, it indicates that the actual displacement direction leads the expected displacement direction. When the deviation direction label is negative, it indicates that the actual displacement direction lags behind the expected displacement direction. The force transmission Jacobian matrix from each joint of the slave-end robotic arm to the end effector is calculated based on the current joint angle. The contact force signal and the contact torque signal are used as inputs, and the inverse matrix of the force transmission Jacobian matrix is ​​used as the transformation matrix to map the contact force signal and the contact torque signal to each joint of the slave-end robotic arm to obtain the equivalent contact torque of each joint. When the equivalent contact torque is positive, it indicates that the contact force applies a positive driving torque to the joint. When the equivalent contact torque is negative, it indicates that the contact force applies a reverse resistance torque to the joint. Multiply the equivalent contact torque of each joint by the deviation direction label, and use the product as the preliminary coupling coefficient of each joint. When the preliminary coupling coefficient is positive, it indicates that the actual movement direction of the joint under the current load condition is consistent with the desired movement direction and is aided by the environment. When the preliminary coupling coefficient is negative, it indicates that the actual movement direction of the joint under the current load condition is opposite to the desired movement direction and is resisted by the environment. The initial coupling coefficients are subjected to a time window moving average process, and the result of the moving average process is used as the load-motion coupling coefficient of each joint.

6. The isomorphic human-machine upper limb master-slave incremental mapping method according to claim 1, characterized in that, The step of determining the incremental mapping ratio coefficient of each joint by combining the motion gain adjustment direction of each joint with the load-motion coupling coefficient of each joint includes: Obtain the direction of joint motion gain adjustment and the load-motion coupling coefficient; Obtain the preset basic mapping ratio coefficient of the joint, which is the default mapping ratio coefficient of the joint when there is no load and the operator is in a moderate motion intention. When the motion gain adjustment direction is the enhancement direction, the absolute value of the load-motion coupling coefficient is added to the preset enhancement adjustment coefficient to obtain the first proportional factor, and the product of the preset basic mapping proportional coefficient and the first proportional factor is used as the enhanced mapping proportional coefficient of the joint. When the motion gain adjustment direction is the suppression direction, the absolute value of the load-motion coupling coefficient is subtracted from the preset suppression adjustment coefficient to obtain the second proportional factor, and the product of the preset basic mapping proportional factor and the second proportional factor is used as the suppressed mapping proportional factor of the joint. The enhanced post-mapping scaling factor or the suppressed post-mapping scaling factor is used as the incremental mapping scaling factor of the joint.

7. The isomorphic human-machine upper limb master-slave incremental mapping method according to claim 1, characterized in that, The step of scaling the joint angle increment signal using the incremental mapping scaling coefficient of each joint to obtain the angle increment command of each joint of the slave robotic arm, and outputting the angle increment command to the joint controller of the slave robotic arm for execution includes: Obtain the incremental mapping ratio coefficient of each joint, and construct an incremental mapping ratio matrix with the incremental mapping ratio coefficient of each joint as the diagonal element. The incremental mapping ratio matrix is ​​a diagonal matrix and all off-diagonal elements are zero. The joint angle increment signals of each joint are arranged in the joint order to obtain the joint angle increment vector; Multiply the incremental mapping ratio matrix with the joint angle increment vector, and use the resulting vector as the initial angle increment command for each joint. For each joint component in the initial angle increment command, obtain the current angle value of the joint and the upper and lower limits of the joint angle safety range. Calculate the predicted angle value after adding the joint component to the current angle value. When the predicted angle value exceeds the upper or lower limit of the joint angle safety range, reduce the joint component proportionally to make the predicted angle value equal to the upper or lower limit of the safety range, and obtain the limited angle increment component corresponding to the joint. Traverse all joints to obtain the limited angle increment components corresponding to each joint, and combine them into a limited angle increment command vector. The angle increment command vector after the amplitude limit is used as the final angle increment command for each joint of the slave robotic arm. The final angle increment command is output to the joint controller of each joint of the slave robotic arm, so that each joint performs movement according to the final angle increment command.

8. A homogeneous human-machine upper limb master-slave incremental mapping system, characterized in that, include: The first acquisition module is configured to acquire the joint angle increment signals of each joint of the operator's upper limb and the operator's surface electromyography signals. The first determining module is configured to extract the activation sequence of each muscle group from the surface electromyography signal, and determine the muscle coordination lag of the operator's current joint based on the activation sequence. The second determining module is configured to determine the motion gain adjustment direction of each joint based on the incremental direction and amplitude of each joint in the incremental joint angle signal, combined with the muscle coordination hysteresis. The second acquisition module is configured to acquire the contact force signal and contact torque signal between the end effector of the slave robotic arm and the working environment, as well as the current joint angle of each joint of the slave robotic arm; The calculation module is configured to calculate the deviation direction between the actual motion trajectory and the desired motion trajectory of the end effector of the slave robotic arm based on the contact force signal and the contact torque signal, and to determine the load-motion coupling coefficient of each joint based on the deviation direction and the current joint angle. The third determining module is configured to determine the incremental mapping ratio coefficient of each joint by combining the motion gain adjustment direction of each joint with the load-motion coupling coefficient of each joint. The output module is configured to scale the joint angle increment signal using the incremental mapping scaling coefficient of each joint to obtain the angle increment command of each joint of the slave robotic arm, and output the angle increment command to the joint controller of the slave robotic arm for execution.

9. An electronic device, characterized in that, include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method according to any one of claims 1 to 7.

Images