Upper extremity exoskeleton control method, system, and apparatus for assisted insulator replacement

Abstract

The present application relates to the technical field of power equipment maintenance, and discloses a method, system and device for controlling an upper limb exoskeleton for assisting in replacing an insulator. The method comprises collecting image data of the insulator and the on-site environment, and obtaining motion vector information from the image data based on a preset optical flow method; analyzing the motion vector information based on a deep learning algorithm to obtain motion state information of the insulator; obtaining adjustment motion parameters of the upper limb exoskeleton based on a dynamics fusion algorithm according to the motion state information of the insulator in combination with position information of the upper limb exoskeleton; and issuing a control instruction to the upper limb exoskeleton according to the adjustment motion parameters. The present application can monitor the dynamic changes of the insulator and the surrounding environment through the optical flow method, accurately identify and analyze the insulator in motion by combining deep learning analysis, timely adjust the motion trajectory of the upper limb exoskeleton, ensure that the upper limb exoskeleton stably and accurately tracks and operates the insulator, reduce the physical burden of the operator, and improve the operation safety and efficiency.

Description

Technical Field

[0001] This invention relates to the field of power equipment maintenance technology, and in particular to a method, system and device for controlling an upper limb exoskeleton to assist in replacing insulators. Background Technology

[0002] With the rapid development of the power industry, the complexity and maintenance difficulty of the power grid are increasing. In the maintenance of high-voltage transmission lines, replacing insulators is a common and important task. As a key component of transmission lines, the performance of insulators directly affects the safe and stable operation of the power grid.

[0003] Traditional insulator replacement technology has many shortcomings. On the one hand, manual operation makes it difficult to precisely control the force and angle during the replacement process, especially when facing complex and ever-changing working environments, which makes it easier for deviations to occur and affect the quality of the work. On the other hand, traditional technology lacks the ability to monitor and adapt to the working environment in real time, and cannot adjust the work strategy in a timely manner according to changes in the environment, which further limits the efficiency and safety of the work.

[0004] In summary, traditional insulator replacement methods suffer from drawbacks such as heavy physical burden, low efficiency, insufficient precision, and poor environmental adaptability. These drawbacks not only increase the risks of the operation but also limit the efficiency and safety of power equipment maintenance. Therefore, it is particularly important to develop an upper limb exoskeleton control algorithm to assist in insulator replacement. Summary of the Invention

[0005] The main objective of this invention is to provide a method, system, and device for controlling an upper limb exoskeleton to assist in replacing insulators, aiming to solve at least one of the aforementioned technical problems.

[0006] To achieve the above objectives, the present invention provides a method for controlling an upper limb exoskeleton to assist in replacing insulators, comprising:

[0007] Image data of insulators and the field environment are collected, and motion vector information is obtained from the image data based on a preset optical flow method;

[0008] The motion vector information is analyzed using a deep learning algorithm to obtain the motion state information of the insulator;

[0009] Based on the dynamic fusion algorithm, the adjustment motion parameters of the upper limb exoskeleton are obtained by combining the motion state information of the insulator with the position information of the upper limb exoskeleton.

[0010] Control commands are sent to the upper limb exoskeleton based on the adjusted motion parameters.

[0011] In some embodiments, acquiring image data of the insulator and the surrounding environment, and obtaining motion vector information from the image data based on a preset optical flow method, includes:

[0012] Real-time acquisition of raw images of insulators and the field environment, and preprocessing of the raw images to obtain image data;

[0013] Principal component analysis was performed on several insulator image samples to extract key feature vectors;

[0014] Construct a direction weight matrix based on the correlation between the key feature vectors and the direction;

[0015] Based on the aforementioned direction weight matrix, the gradient method in the preset optical flow method is improved to obtain the improved gradient method;

[0016] Motion vector information is obtained based on the improved gradient method and image data.

[0017] In some embodiments, obtaining motion vector information based on the improved gradient method and image data includes:

[0018] The first gradient, the second gradient, and the temporal gradient are calculated based on the improved gradient method and the image data.

[0019] Calculate the first direction motion vector and the second direction motion vector based on the first direction gradient, the second direction gradient, the time direction gradient, and the direction weight matrix, respectively.

[0020] Motion vector information is obtained based on the first direction motion vector and the second direction motion vector.

[0021] In some embodiments, the analysis of the motion vector information based on a deep learning algorithm to obtain the motion state information of the insulator includes:

[0022] Deep learning models are built based on multi-layer cascaded attention networks;

[0023] The motion vector information is input into the deep learning model to identify insulators in motion and obtain key motion features related to the insulators; wherein, the key motion features include the insulator's motion direction and velocity information;

[0024] The motion state information of the insulator is generated based on the key motion characteristics.

[0025] In some embodiments, the dynamic fusion algorithm obtains the adjustment motion parameters of the upper limb exoskeleton based on the motion state information of the insulator and the position information of the upper limb exoskeleton, including:

[0026] Set the desired position of the end effector of the upper limb exoskeleton, and obtain the current position and joint angles of the end effector of the upper limb exoskeleton.

[0027] A dynamic weight vector is obtained by combining the motion state information of the insulator with the historical operational data of the upper limb exoskeleton.

[0028] The joint angle adjustment amount is obtained based on the desired position, current position, joint angles, and dynamic weight vector using a dynamic fusion algorithm.

[0029] The joint angle adjustment amount is used as the adjustment motion parameter of the upper limb exoskeleton.

[0030] In some embodiments, obtaining the dynamic weight vector based on the insulator's motion state information and the historical operational data of the upper limb exoskeleton includes:

[0031] Acquire historical operational data of the upper limb exoskeleton under a preset environment;

[0032] Get real-time information on current wind direction and speed;

[0033] The complexity of the working environment is assessed by combining the insulator's motion state information with the historical operation data and the current wind direction and speed information, resulting in a dynamic weight vector.

[0034] In some embodiments, the method further includes:

[0035] Establish a mapping model between environmental parameters and insulator motion disturbance;

[0036] Real-time monitoring of environmental parameters at the work site;

[0037] The environmental parameters are analyzed in real time to obtain the trend of environmental parameter changes, and the trend of environmental parameter changes is used to determine whether it will interfere with the movement of the insulator.

[0038] If so, then a compensation instruction is generated based on the mapping model and the environmental parameters;

[0039] The motion trajectory of the upper limb exoskeleton is adjusted and corrected based on the compensation command.

[0040] In some embodiments, establishing the mapping model between environmental parameters and insulator motion disturbance includes:

[0041] The original environmental parameters are generated based on wind speed, temperature, humidity, and equipment vibration frequency.

[0042] Data samples were generated based on experimental observations of insulator motion under different combinations of original environmental parameters;

[0043] Based on the data samples, a multivariate linear regression model based on particle swarm optimization is trained to obtain a mapping model between environmental parameters and insulator motion disturbance.

[0044] Furthermore, to achieve the above objectives, the present invention also proposes an upper limb exoskeleton control system for assisting in the replacement of insulators, comprising:

[0045] The optical flow monitoring module is used to collect image data of the insulator and the field environment, and to obtain motion vector information based on the image data using a preset optical flow method;

[0046] The insulator analysis module is used to analyze the motion vector information based on deep learning algorithms to obtain the motion state information of the insulator;

[0047] The trajectory adjustment module is used to obtain the adjustment motion parameters of the upper limb exoskeleton based on the motion state information of the insulator and the position information of the upper limb exoskeleton according to the dynamic fusion algorithm.

[0048] The motion control module is used to send control commands to the upper limb exoskeleton based on the adjusted motion parameters.

[0049] Furthermore, to achieve the above objectives, the present invention also proposes an electronic device comprising: a memory, a processor, and an upper limb exoskeleton control program for assisting insulator replacement stored in the memory and executable on the processor, wherein the upper limb exoskeleton control program for assisting insulator replacement is configured to implement the upper limb exoskeleton control method for assisting insulator replacement as described above.

[0050] This invention provides a method for controlling an upper limb exoskeleton to assist in insulator replacement. The method includes: acquiring image data of the insulator and the surrounding environment; obtaining motion vector information from the image data based on a preset optical flow method; analyzing the motion vector information using a deep learning algorithm to obtain the insulator's motion state information; obtaining adjustment motion parameters for the upper limb exoskeleton based on the insulator's motion state information and the exoskeleton's position information using a dynamic fusion algorithm; and issuing control commands to the exoskeleton based on the adjusted motion parameters. This invention uses optical flow to monitor the dynamic changes of the insulator and its surrounding environment, and combines this with deep learning analysis to accurately identify and analyze the moving insulator. This allows for timely adjustment of the exoskeleton's trajectory, ensuring stable and accurate tracking and operation of the insulator. This not only reduces the physical burden on operators but also improves operational safety and efficiency. Attached Figure Description

[0051] Figure 1 This is a schematic diagram of the structure of an electronic device in the hardware operating environment involved in the embodiments of the present invention;

[0052] Figure 2 This is a flowchart illustrating an embodiment of the upper limb exoskeleton control method for assisting insulator replacement according to the present invention;

[0053] Figure 3 This is a flowchart illustrating the control algorithm involved in the embodiments of the present invention;

[0054] Figure 4 This is a flowchart illustrating the key control algorithm involved in the embodiments of the present invention;

[0055] Figure 5 This is a structural block diagram of an embodiment of the upper limb exoskeleton control system for assisting insulator replacement according to the present invention;

[0056] Figure 6 This is a schematic diagram of an example module of the control system involved in an embodiment of the present invention.

[0057] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0058] 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 a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0059] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0060] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention.

[0061] Reference Figure 1 , Figure 1This is a schematic diagram of the electronic device structure of the hardware operating environment involved in the embodiments of the present invention.

[0062] like Figure 1 As shown, the electronic device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface). The memory 1005 may be high-speed random access memory (RAM) or stable non-volatile memory (NVM), such as a disk drive. The memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001.

[0063] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0064] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a network communication module, a user interface module, and an upper limb exoskeleton control program for assisting in the replacement of insulators.

[0065] exist Figure 1 In the electronic device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the electronic device of the present invention can be set in the electronic device. The electronic device calls the upper limb exoskeleton control program for assisting insulator replacement stored in the memory 1005 through the processor 1001, and executes the upper limb exoskeleton control method for assisting insulator replacement provided in the embodiment of the present invention.

[0066] This invention proposes a method, system, and device for controlling an upper limb exoskeleton to assist in replacing insulators.

[0067] This invention provides a method for controlling an upper limb exoskeleton to assist in replacing insulators, referring to... Figure 2, Figure 2 This is a flowchart illustrating an embodiment of the upper limb exoskeleton control method for assisting in insulator replacement according to the present invention.

[0068] like Figure 2 As shown, the upper limb exoskeleton control method for assisting in insulator replacement includes:

[0069] Step S100: Collect image data of the insulator and the field environment, and obtain motion vector information based on the image data using a preset optical flow method;

[0070] Step S200: Analyze the motion vector information based on a deep learning algorithm to obtain the motion state information of the insulator;

[0071] Step S300: Based on the dynamic fusion algorithm, the adjustment motion parameters of the upper limb exoskeleton are obtained by combining the motion state information of the insulator with the position information of the upper limb exoskeleton;

[0072] Step S400: Send control commands to the upper limb exoskeleton according to the adjusted motion parameters.

[0073] It should be noted that the execution subject in this embodiment can be an electronic device, which can be a computer device with data processing function, or other devices that can achieve the same or similar functions. This embodiment does not limit this; in this embodiment, a computer device is used as an example for explanation.

[0074] In one embodiment, image data of insulators and the surrounding environment are acquired, and motion vector information is obtained from the image data based on a preset optical flow method. This includes: acquiring raw images of the insulators and the surrounding environment in real time, and preprocessing the raw images to obtain image data; performing principal component analysis on several insulator image samples to extract key feature vectors; constructing a direction weight matrix based on the correlation between the key feature vectors and directions; improving the gradient method in the preset optical flow method based on the direction weight matrix to obtain an improved gradient method; and obtaining motion vector information based on the improved gradient method and the image data.

[0075] In one embodiment, obtaining motion vector information based on the improved gradient method and image data includes: calculating a first directional gradient, a second directional gradient, and a temporal gradient based on the improved gradient method and image data; calculating a first directional motion vector and a second directional motion vector based on the first directional gradient, the second directional gradient, the temporal gradient, and a directional weight matrix, respectively; and obtaining motion vector information based on the first directional motion vector and the second directional motion vector.

[0076] Specifically, such as Figure 3As shown, monitoring equipment is installed, and optical flow monitoring is performed: real-time monitoring of the dynamic changes of the insulator and its surrounding environment to obtain motion vector information. Optical flow refers to the velocity field of pixels generated by the movement of objects in an image. When an object moves in an image sequence, the grayscale value of its corresponding pixel changes position between different frames. Optical flow is a physical quantity describing the direction and velocity of this pixel's movement between adjacent frames. In this embodiment, the motion information of the insulator, such as its direction and velocity, can be obtained by analyzing optical flow. Optical flow algorithms include, but are not limited to, gradient-based methods (such as the Lucas-Kanade algorithm), matching-based methods, energy-based methods, and phase-based methods. In this embodiment, a custom optical flow algorithm (improved gradient method) is used, which is an improved algorithm of the gradient-based method.

[0077] For example, a high-definition camera is installed at a suitable position on the exoskeleton to continuously acquire image sequences (raw images) of the insulator and its surrounding environment at a fixed frame rate. To reduce noise interference with subsequent optical flow calculations, the acquired raw images are preprocessed to obtain image data. For example, such as... Figure 4 As shown, Gaussian filtering is used to perform convolution processing on each frame of the image, effectively filtering high-frequency noise in the image, making the image smoother, and laying a stable data foundation for optical flow calculation.

[0078] Specifically, this embodiment employs an optical flow calculation method based on an improved gradient approach. First, feature extraction is performed on a large number of insulator sample images from different scenarios, yielding multiple feature vectors (key feature vectors). The correlation between each feature vector and the motion direction of objects in the image is then analyzed in depth. Based on these correlations, a direction weight matrix is ​​constructed. This direction weight matrix is ​​incorporated to significantly improve the accuracy of capturing the motion vectors of insulators and surrounding objects. Finally, the motion vectors of adjacent pixels in the image sequence (or image data) are calculated, providing detailed and reliable motion information for subsequent deep learning analysis modules.

[0079] For example, in the optical flow monitoring process, the image sequence captured by the monitoring device, such as a camera, is processed by a custom optical flow algorithm (improved gradient method). This custom optical flow algorithm is based on the improved gradient method and introduces a direction weight matrix. Let the gradients of the image in the x (first direction) and y (second direction) directions be I and I, respectively. x and I y The gradient in the time direction is I t The orientation weight matrix is ​​W. Principal component analysis is performed on a large number of insulator image samples from different scenarios to extract key feature vectors. A weight matrix (orientation weight matrix W) is constructed based on the correlation between the key feature vectors and their orientation. The motion vector (motion vector information) V = (V x V yThe formula for calculating ) is:

[0080]

[0081] Among them, V x V is the motion vector in the first direction; y The second direction motion vector; n is the number of pixels involved in the calculation within the local image region; w i,x and w i,y These are the weights in the x and y directions at the i-th pixel, respectively. This algorithm (based on optical flow calculation using an improved gradient method) can more accurately capture the motion vectors of insulators and surrounding objects, providing a more precise data foundation for subsequent analysis.

[0082] In one embodiment, the motion vector information is analyzed based on a deep learning algorithm to obtain the motion state information of the insulator, including: constructing a deep learning model based on a multi-layer cascaded attention network; inputting the motion vector information into the deep learning model to identify the insulator in motion and obtain key motion features related to the insulator; wherein, the key motion features include the insulator's motion direction and velocity information; and generating the insulator's motion state information based on the key motion features.

[0083] Specifically, such as Figure 3 and Figure 4 As shown, a deep learning algorithm is used to analyze optical flow data (motion vector information) to identify insulators in motion and obtain their motion state information.

[0084] For example, the deep learning model constructed by deep learning analysis can be a multi-layer cascaded attention network. The input of the deep learning model is the motion vector information obtained by optical flow. An attention mechanism module is introduced after each convolution operation. Let the input feature map of the l-th layer be F. l Attention weight A l Calculated using the following formula:

[0085] A l =sigmoid(W1·(W2·F) l +b2)+b1)

[0086] Wherein, W1 and W2 are weight matrices; b1 and b2 are bias terms. The weight matrices W1 and W2, as well as the bias terms, are determined through backpropagation training on a large number of optical flow data samples containing insulators in different motion states. This model (deep learning model) can automatically focus on key motion features related to insulators, suppress irrelevant background interference, and more accurately identify and analyze the motion state of insulators. By learning the features of insulators under different motion directions and velocities in a large number of samples, it accurately outputs the motion direction and velocity information of insulators, significantly improving the accuracy of identification and analysis.

[0087] Specifically, deep learning analysis involves constructing a multi-layered cascaded attention network to analyze optical flow data. The model input originates from motion vector information obtained by optical flow monitoring. The structure of the multi-layered cascaded attention network includes multiple convolutional layers, pooling layers, and attention mechanism modules. After each convolutional operation is completed, an attention mechanism is introduced to enable the model to automatically focus on and capture key motion features related to the insulator, effectively suppressing irrelevant background interference and enhancing the accuracy of insulator motion state analysis. For example, a wide range of optical flow data samples of insulators and their surrounding environment in different scenarios are collected, and the position, direction of motion, velocity and motion state information of the insulators are labeled in detail. These labeled sample data are divided into training set, validation set and test set. During the training process, the stochastic gradient descent algorithm is used to update the model parameters. In each iteration, a batch of samples is randomly selected from the training set, and the loss of the model on this batch of samples is calculated. At the same time, the model performance is evaluated on the validation set regularly to prevent the model from overfitting. When the model performance no longer improves on the validation set, training is stopped, and the final test is performed on the test set to obtain the trained multi-layer cascaded attention network, which serves as the deep learning model for deep learning analysis. This ensures that the deep learning model can accurately identify moving insulators and accurately analyze their motion state.

[0088] Understandably, motion vector information obtained using optical flow monitoring undergoes preprocessing before deep learning analysis. This preprocessing includes noise reduction and normalization of the motion vector information acquired by optical flow. For example, for noise reduction, an adaptive median filtering algorithm can be used. Based on the statistical characteristics of the motion vectors of pixels within a local region, the size of the filtering window is dynamically adjusted. Let the local region window size be S, and the mean value of the motion vectors within the window is calculated. and standard deviation σ V When σ V When the value is greater than the threshold T, the window S is increased; conversely, the window S is decreased. For example, the normalization process can employ a max-min normalization algorithm to normalize the motion vector V to the [0, 1] interval, as shown in the formula:

[0089]

[0090] Among them, V min and V max These represent the minimum and maximum values ​​of the motion vector over a given period of time. Preprocessed data (preprocessed motion vector information) can be better utilized in deep learning analysis, improving the efficiency and accuracy of the analysis.

[0091] In one embodiment, the adjustment motion parameters of the upper limb exoskeleton are obtained based on the motion state information of the insulator and the position information of the upper limb exoskeleton using a dynamic fusion algorithm. This includes: setting the desired position of the end effector of the upper limb exoskeleton; obtaining the current position of the end effector and the angles of each joint; obtaining a dynamic weight vector based on the motion state information of the insulator and the historical operation data of the upper limb exoskeleton; obtaining the joint angle adjustment amount based on the desired position, current position, joint angles, and dynamic weight vector using the dynamic fusion algorithm; and using the joint angle adjustment amount as the adjustment motion parameter of the upper limb exoskeleton.

[0092] In one embodiment, a dynamic weight vector is obtained by combining the motion state information of the insulator with the historical operation data of the upper limb exoskeleton, including: acquiring the historical operation data of the upper limb exoskeleton in a preset environment; acquiring the current wind direction and wind speed information in real time; and evaluating the complexity of the operation environment based on the motion state information of the insulator combined with the historical operation data and the current wind direction and wind speed information to obtain the dynamic weight vector.

[0093] Specifically, such as Figure 3 As shown, motion trajectory adjustment: Based on the motion state information of the insulator, combined with the current position and posture of the exoskeleton, motion parameters are obtained to adjust the motion trajectory of the exoskeleton.

[0094] For example, when calculating and adjusting motion parameters, a kinematic and dynamic fusion algorithm based on dynamic weights can be used, where the desired position of the exoskeleton end effector is P. d The current position is P, and the angles of each joint are θ = (θ1, θ2, ..., θ). n ), dynamic weight vector W d =(w d1 ,w d2 ,...,w dn ), whereby, based on the historical operational data of the exoskeleton and the current operational environment complexity assessment, the dynamic weight vector W is determined. d Using the kinematic forward kinematics formula P = f k (θ) and the inverse dynamics formula Combined with dynamic weights (dynamic weight vector W) d Calculate and adjust the motion parameters (joint angle adjustment amount). The formula for calculating the joint angle adjustment amount Δθ is:

[0095]

[0096] Where J is the Jacobian matrix; λ and μ are control gains, determined through experimental and simulation optimization. This embodiment employs a kinematic and dynamic fusion algorithm based on dynamic weights, which can more flexibly and accurately adjust the motion trajectory according to the insulator's motion state and the actual situation of the exoskeleton.

[0097] In practical applications, upper limb exoskeletons are used to coordinate with the movement of insulators for operation, and the desired position P of the exoskeleton end effector is... d Typically, the position needs to be determined based on the insulator's motion state information, and the insulator's current position can be used as the desired position P of the exoskeleton end effector. d Alternatively, considering the delay in exoskeleton movement and the movement trend of the insulator, the position of the insulator can be predicted to obtain a more suitable desired position P. d Dynamic weight vector W d The importance of each joint in motion control is adjusted. The insulator's swing amplitude, speed, direction, and other motion state information will change the complexity of the exoskeleton's working environment. An environmental complexity assessment model can be established. This model obtains an environmental complexity assessment value based on the insulator's motion state information. A dynamic weight vector W is then determined based on the environmental complexity assessment value and the exoskeleton's historical operational data. d For example, first, the basic weights of each joint under different environmental complexities are obtained based on historical operation data, and then adjusted according to the current environmental complexity.

[0098] It should be noted that after obtaining the insulator's motion state information through deep learning analysis, based on this information and the current position and posture of the exoskeleton, kinematic and dynamic algorithms are used to calculate the various motion parameters required to adjust the exoskeleton's motion trajectory (adjustment motion parameters). In complex environments, the weight of joints closer to the target direction increases, thereby enhancing the joint's influence in motion trajectory adjustment. Finally, based on the insulator's real-time motion state and the exoskeleton's current status, the joint angle adjustment amount required to adjust the exoskeleton's motion trajectory is flexibly and accurately calculated as the key motion parameters (adjustment motion parameters). Based on the calculated adjustment motion parameters, such as joint angle adjustment amount, velocity, and acceleration, the adjustment motion parameters are converted into control commands that the exoskeleton drive device can recognize. These control commands include key information such as the target angle and target velocity of each joint. Subsequently, the control commands are transmitted to the exoskeleton drive device through a communication interface. The drive device adjusts the movement of each joint of the exoskeleton according to the control commands to adjust the exoskeleton's motion trajectory, ensuring accurate tracking of the moving insulator and completing the insulator replacement operation.

[0099] In one embodiment, the method further includes: establishing a mapping model between environmental parameters and insulator motion interference; monitoring environmental parameters at the work site in real time; performing real-time analysis based on the environmental parameters to obtain the trend of environmental parameter changes, and determining whether the trend of environmental parameter changes interferes with the insulator motion; if so, generating a compensation instruction based on the mapping model and the environmental parameters; and adjusting and correcting the motion trajectory of the upper limb exoskeleton based on the compensation instruction.

[0100] In one embodiment, establishing a mapping model between environmental parameters and insulator motion disturbance includes: generating original environmental parameters based on wind speed, temperature, humidity, and equipment vibration frequency; generating data samples based on experimental observations of insulator motion under different combinations of original environmental parameters; and training a multivariate linear regression model based on particle swarm optimization based on the data samples to obtain the mapping model between environmental parameters and insulator motion disturbance.

[0101] Specifically, such as Figure 3 As shown, environmental parameter perception and compensation involves real-time monitoring of environmental parameters at the work site. Based on a pre-established mapping model, the impact of environmental factors on insulator movement is determined, and compensation commands are sent to the upper limb exoskeleton to adjust the movement trajectory when necessary. Specifically, constructing the mapping model between environmental parameters and insulator movement interference can employ a multiple linear regression model based on particle swarm optimization. Extensive experimental observations of insulator movement are conducted under different combinations of environmental parameters to obtain data samples. Using the particle swarm optimization algorithm, the optimal regression coefficient that minimizes the objective function value is found, thereby determining the mapping relationship between environmental parameters and the amount of insulator movement interference, thus forming the mapping model.

[0102] For example, in the process of environmental parameter sensing and compensation, when establishing the mapping model between environmental parameters and insulator motion disturbance, a multiple linear regression model based on particle swarm optimization is adopted, assuming the environmental parameters (original environmental parameters) are X = (x1, x2, ..., x...). m The data sample (X) includes wind speed, temperature, humidity, equipment vibration frequency, and insulator motion disturbance quantity Y. Through extensive experimental observations of insulator motion under different combinations of environmental parameters, data samples were obtained. i ,Y i The particle swarm optimization algorithm is used to find the optimal regression coefficients β = (β0, β1, ..., βN), where i = 1, 2, ..., N. m ), so that the objective function In the minimum, each particle in the particle swarm represents a set of regression coefficients, whose position and velocity are continuously updated in the search space. After multiple iterations of optimization, the final regression coefficients are determined, forming a mapping model. This model (mapping model) can more accurately reflect the influence of environmental parameters on insulator movement, providing a reliable basis for the generation of compensation commands.

[0103] In one embodiment, control commands are sent to the upper limb exoskeleton based on the adjusted motion parameters.

[0104] Specifically, motion smoothing processing can be performed between the motion trajectory adjustment and the exoskeleton drive device. This smooths the control commands output by the motion trajectory adjustment, preventing abrupt changes and impacts during exoskeleton movement. For example, a smoothing algorithm based on Bézier curves can be used, where the control command sequence is C = (c1, c2, ..., c...). n By determining the control points of the Bézier curve, the control command is converted into a smooth curve. The determination of the Bézier curve control points is based on the motion characteristics and operational requirements of the exoskeleton. According to the maximum acceleration and speed limits of the exoskeleton joints and the smoothness requirements of the insulator replacement operation, the position of the control points is adjusted. For the time parameter t∈[0,1], the smoothed control command C smooth (t) is calculated using the Bézier curve formula:

[0105]

[0106] Among them, B i,n (t) is a Bernstein polynomial. This motion-smoothing control command enables smoother exoskeleton movement, extends equipment lifespan, and improves operator comfort.

[0107] Understandably, the environmental parameter perception and compensation system also features environmental risk early warning capabilities. When environmental parameters exceed safety thresholds, it not only sends compensation commands to adjust the movement trajectory but also issues warning signals to the operators. These safety thresholds can be determined through analysis of extensive power operation accident data and exoskeleton performance test data under different environments. Warning signals include, but are not limited to, audible and visual alarms, which alert operators through flashing indicator lights on the exoskeleton and loudspeaker sounds. Simultaneously, it can upload current environmental parameters and warning information to a remote monitoring center in real time, allowing managers to promptly grasp the on-site situation of power equipment maintenance operations and take appropriate measures to further ensure operational safety.

[0108] For example, various sensors such as wind speed sensors, temperature and humidity sensors, and vibration sensors are deployed in the working environment to collect environmental parameters such as wind force, temperature, humidity, and equipment vibration frequency and amplitude in real time. The data collected by the sensors can be transmitted to a processor for environmental parameter sensing and compensation via wired or wireless communication. The processor analyzes the collected environmental parameters in real time, calculates the trend of environmental parameter changes, and the deviation between the current value and the preset safety threshold, thereby judging the environmental changes. When the processor for environmental parameter sensing and compensation determines that environmental factors have a significant impact on the movement of the insulator and may affect the operating accuracy of the exoskeleton, it calculates a correction suggestion for the motion parameters based on the mapping model and generates a compensation command. The compensation command may include correction values ​​for joint angle adjustment, speed, and acceleration motion parameters. Then, the compensation command is sent through the communication interface, and the calculated motion parameters (adjusted motion parameters) are corrected according to the compensation command to ensure that the exoskeleton can still stably and accurately track and operate the insulator in complex and changing environments.

[0109] It should be noted that the method described in this embodiment is illustrated below with two specific implementation examples:

[0110] Example 1: On a power transmission line in a mountainous area, the insulators needed to be replaced due to long-term erosion by the natural environment. During the operation, the wind speed was high, reaching 8 m / s, and the wind direction was unstable with gusts. This had a significant impact on the stability of the insulators and the operational precision of the exoskeleton. The mountainous terrain where the power transmission line is located is complex, with trees and rocks all around, which further increased the difficulty of the operation.

[0111] A high-definition camera on the exoskeleton captures image sequences of the insulator and its surrounding environment at a frame rate of 30 frames per second. Due to the complex lighting conditions in the mountainous area, the captured images (raw images) may contain noise. Therefore, the raw images are first preprocessed using Gaussian filtering, and high-frequency noise is effectively filtered out through convolution operations to smooth the images and obtain image data, laying a stable data foundation for subsequent optical flow calculations. Motion vectors are calculated based on an improved gradient method. Within a certain local region, the gradient I in the x-direction of the image is... x =0.3, gradient I in the y-direction y =0.2, time gradient I t =0.1, the number of pixels involved in the calculation n=10, the orientation weight matrix W is constructed by performing principal component analysis on a large number of insulator image samples in different mountainous scenes, extracting key feature vectors, and based on the correlation between feature vectors and orientation.

[0112] Taking a certain pixel as an example, suppose its weight w in the x-direction i,x =0.6, y-direction weight w i,y =0.4, substitute into the formula:

[0113]

[0114] Calculate the motion vector V = (V_0.05) of the pixels in this region. x V y This provides basic data for subsequent analysis.

[0115] The motion vector information (motion vector V) obtained by optical flow is input into a multi-layer cascaded attention network. This network structure contains multiple convolutional layers, pooling layers, and attention mechanism modules. The attention mechanism module is introduced after each convolutional operation. Assume that the input feature map F of the l-th layer is... l Through the formula:

[0116] A l =sigmoid(W1·(W2·F) l +b2)+b1)

[0117] Calculate attention weight A l The weight matrices W1 and W2, and the bias terms b1 and b2 are determined during model training. When processing motion vector information, the deep learning model can automatically focus on the key motion features of the insulator, suppress irrelevant interference such as the swaying of trees in the mountainous background, and identify the motion state information of the insulator under strong wind with large amplitude and high speed, and the direction of the sway changing with the wind direction.

[0118] Given the desired position P of the exoskeleton end effector d The current position is P, and the angles of each joint are θ = (θ1, θ2, ..., θ). n Based on historical operational data of exoskeletons in mountainous areas with strong winds, and an assessment of the operational environment complexity caused by frequent changes in wind direction and speed, a dynamic weight vector W is determined. d =(w d1 ,w d2 ,...,w dn The weight of joints closer to the swing direction of the insulator will increase, thereby enhancing their influence in adjusting the motion trajectory.

[0119] Using the kinematic forward kinematics formula P = f k (θ) and the inverse dynamics formula The joint angle adjustment Δθ is calculated using dynamic weighting, and the formula is as follows:

[0120]

[0121] Among them, the Jacobian matrix J is determined according to the mechanical structure of the exoskeleton, and the control gains λ and μ are optimized through multiple experiments to calculate key motion parameters such as the joint angle adjustment required to adjust the exoskeleton's motion trajectory, so that the exoskeleton's motion trajectory adapts to the insulator's motion.

[0122] A wind speed sensor collects wind speed data in real time. The processor determines if the wind speed exceeds the safe threshold of 8 m / s. Using a particle swarm optimization-based multiple linear regression model, let the wind speed be x1 = 8 m / s, the wind direction be x2 (assumed to be expressed as an angle), and other environmental parameters (such as temperature, humidity, equipment vibration frequency, etc.) be x3,...,x m Based on a large number of experimental observation data samples of insulator motion under different combinations of wind speed, wind direction and other environmental parameters, the optimal regression coefficient β=(β0,β1,...,β) was found using the particle swarm optimization algorithm. m ), so that the objective function Minimize and determine the mapping relationship between environmental parameters and insulator motion disturbance, forming a mapping model.

[0123] The insulator motion interference quantity Y is calculated based on the mapping model, and a compensation command is sent to the motion trajectory adjustment module to correct the motion parameters. At the same time, the indicator lights on the exoskeleton flash rapidly, and the speaker emits a sharp alarm sound to remind the operator to pay attention to safety. The current environmental parameters (wind speed, wind direction) and early warning information are uploaded to the remote monitoring center in real time so that the monitoring personnel can keep abreast of the situation on site.

[0124] Example 2: Some insulators in a substation are aging and need to be replaced. The operation takes place during the hot and humid summer, with the temperature reaching 38°C and the relative humidity reaching 80%. The heat generated by the operation of electrical equipment in the substation further exacerbates the high temperature in the local environment. This hot and humid environment may not only affect the performance of the insulators, but also cause fatigue for the exoskeleton operators and reduce the accuracy of their operations.

[0125] The camera captures images at 25 frames per second. Since high temperature and humidity may cause images to be blurry or noisy, the captured images (raw images) are first preprocessed using Gaussian filtering. Convolution processing is performed on each frame of the image to effectively filter high-frequency noise in the image, making the image smoother and obtaining image data to provide stable data for subsequent optical flow calculations.

[0126] When calculating motion vectors, it is assumed that in a specific region I x =0.25, I y =0.18, I t =0.08, the number of pixels involved in the calculation n=8, the directional weight matrix W is constructed based on principal component analysis of a large number of insulator image samples under different substation scenarios. Taking a certain pixel as an example, if the weight w in the x direction... i,x =0.5, y-direction weight w i,y =0.5, substitute into the formula:

[0127]

[0128] The motion vector V of the pixels in this region is calculated to obtain dynamic change information of the insulator and its surrounding environment.

[0129] Motion vector information is input into a multi-layered cascaded attention network, and the layers in the model are connected by the following formula:

[0130] A l =sigmoid(W1·(W2·F) l +b2)+b1)

[0131] Calculate attention weight A l The weight matrices W1 and W2 and the bias terms b1 and b2 are determined during the training of the deep learning model. The deep learning model can automatically focus on capturing key motion features related to the insulator, identify the slow displacement of the insulator due to condensation on the surface caused by high temperature and high humidity, and the motion state change information caused by the small deformation due to thermal expansion and contraction.

[0132] Determine the desired position P of the exoskeleton end effector d The current position is P, and the angles of each joint are θ = (θ1, θ2, ..., θ). n Based on historical operational data of the exoskeleton in high-temperature and high-humidity environments, and an assessment of the operational environment complexity caused by factors such as equipment heat dissipation within the current substation, a dynamic weight vector W is determined. d =(w d1 ,w d2 ,...,w dn The joint weights near the insulator displacement direction will be appropriately increased.

[0133] Using the kinematic forward kinematics formula P = f k (θ) and the inverse dynamics formula The joint angle adjustment Δθ is calculated using dynamic weighting, and the formula is as follows:

[0134]

[0135] Among them, the Jacobian matrix J is determined based on the mechanical structure of the exoskeleton, and the control gains λ and μ are set through experimental optimization. The key motion parameters required to adjust the movement trajectory of the exoskeleton are calculated to ensure that the exoskeleton can accurately approach and operate the insulator.

[0136] Temperature and humidity sensors collect temperature and humidity data. The processor determines that both the temperature of 38℃ and the relative humidity of 80% exceed the safety threshold. Using environmental parameters of temperature x1 = 38℃ and humidity x2 = 80% as input, a multiple linear regression model based on particle swarm optimization is used. Calculate the insulator motion disturbance, where the regression coefficient β=(β0,β1,...,β)m The optimal value was determined by using a particle swarm optimization algorithm to find the optimal value based on a large number of experimental observation data samples of insulator motion under different combinations of temperature, humidity and other environmental parameters.

[0137] When it is calculated that environmental factors significantly interfere with the movement of the insulator, a compensation command is sent to the motion trajectory adjustment module to correct the motion parameters. At the same time, the indicator lights on the exoskeleton flash at a specific frequency, and the speaker emits a soft but continuous alarm sound to remind the operator to pay attention to the working environment. The current environmental parameters and early warning information are uploaded to the remote monitoring center in real time so that the monitoring personnel can pay attention to the on-site situation and make corresponding decisions at any time.

[0138] Compared with existing technologies, the upper limb exoskeleton control method for replacing insulators using this embodiment has the following beneficial effects: By using optical flow to monitor the dynamic changes of the insulator and its surrounding environment in real time, and combining this with deep learning analysis to accurately identify and analyze the moving insulator, the exoskeleton's movement trajectory can be adjusted in a timely manner. This ensures that the exoskeleton can stably and accurately track and operate the insulator, which not only reduces the physical burden on operators but also improves the safety and efficiency of the operation. Furthermore, through environmental parameter sensing and compensation, environmental parameters at the work site, such as wind speed, temperature, humidity, and equipment vibration frequency, can be monitored in real time. Based on a pre-established mapping model, the influence of environmental factors on the insulator's movement can be determined. When environmental factors significantly interfere with the insulator's movement, compensation commands can be automatically sent to correct the movement parameters of the trajectory adjustment, ensuring that the exoskeleton can still operate stably in complex and changing environments. This environmental adaptability and robustness make this control method widely applicable in fields such as power equipment maintenance.

[0139] This embodiment provides a method for controlling an upper limb exoskeleton to assist in insulator replacement. The method includes: acquiring image data of the insulator and the surrounding environment; obtaining motion vector information from the image data based on a preset optical flow method; analyzing the motion vector information using a deep learning algorithm to obtain the insulator's motion state information; obtaining adjustment motion parameters for the upper limb exoskeleton based on the insulator's motion state information and the exoskeleton's position information using a dynamic fusion algorithm; and issuing control commands to the exoskeleton based on the adjusted motion parameters. This embodiment uses optical flow to monitor the dynamic changes of the insulator and its surrounding environment, and combines this with deep learning analysis to accurately identify and analyze the moving insulator. This allows for timely adjustment of the exoskeleton's trajectory, ensuring stable and accurate tracking and operation of the insulator. This not only reduces the physical burden on operators but also improves operational safety and efficiency.

[0140] Furthermore, this embodiment of the invention also proposes a storage medium storing an upper limb exoskeleton control program for assisting insulator replacement. When the upper limb exoskeleton control program for assisting insulator replacement is executed by a processor, it implements the steps of the upper limb exoskeleton control method for assisting insulator replacement as described above.

[0141] Reference Figure 5 , Figure 5 This is a structural block diagram of an embodiment of the upper limb exoskeleton control system for assisting in the replacement of insulators according to the present invention.

[0142] like Figure 5 As shown, the upper limb exoskeleton control system for assisting in insulator replacement includes:

[0143] The optical flow monitoring module 10 is used to collect image data of the insulator and the field environment, and to obtain motion vector information based on the image data according to the preset optical flow method.

[0144] The insulator analysis module 20 is used to analyze the motion vector information based on a deep learning algorithm to obtain the motion state information of the insulator;

[0145] The trajectory adjustment module 30 is used to obtain the adjustment motion parameters of the upper limb exoskeleton based on the motion state information of the insulator and the position information of the upper limb exoskeleton according to the dynamic fusion algorithm.

[0146] The motion control module 40 is used to send control commands to the upper limb exoskeleton according to the adjusted motion parameters.

[0147] Specifically, the optical flow monitoring module 10 includes an optical flow monitoring submodule, used to monitor the dynamic changes of the insulator and its surrounding environment in real time and acquire motion vector information. For example... Figure 6 As shown, in the optical flow monitoring submodule, the image sequence captured by the camera is processed by a custom optical flow algorithm. This algorithm is based on an improved gradient method and introduces a direction weight matrix. Let the gradients of the image in the x and y directions be I0 and I0, respectively. x and I y The gradient in the time direction is I t The orientation weight matrix is ​​W. Principal component analysis is performed on a large number of insulator image samples from different scenarios to extract key feature vectors. A weight matrix W is constructed based on the correlation between the feature vectors and their orientation. The motion vector V = (V... x V y The formula for calculating ) is:

[0148]

[0149] Where n is the number of pixels involved in the calculation within the local region of the image, w i,x and wi,y These are the weights at the i-th pixel in the x and y directions, respectively.

[0150] Specifically, the insulator analysis module 20 includes a deep learning analysis submodule, which uses deep learning algorithms to analyze optical flow data, identify moving insulators, and obtain their motion state information. For example... Figure 6 As shown, the deep learning model constructed by the deep learning analysis submodule is a multi-layer cascaded attention network. The model input is the motion vector information obtained by optical flow. An attention mechanism module is introduced after each convolutional operation. Let the input feature map of the l-th layer be F. l Attention weight A l Calculated using the following formula:

[0151] A l =sigmoid(W1·(W2·F) l +b2)+b1)

[0152] Where W1 and W2 are weight matrices; b1 and b2 are bias terms. This model can automatically focus on key motion features related to the insulator and suppress irrelevant background interference.

[0153] Specifically, the trajectory adjustment module 30 includes a motion trajectory adjustment submodule, which adjusts the exoskeleton's motion trajectory based on the insulator's motion state information, combined with the exoskeleton's current position and posture. For example... Figure 6 As shown, the motion trajectory adjustment submodule uses a kinematics and dynamics fusion algorithm based on dynamic weights when calculating and adjusting motion parameters. Let P be the desired position of the exoskeleton end effector. d The current position is P, and the angles of each joint are θ = (θ1, θ2, ..., θ). n ), dynamic weight vector W d =(w d1 ,w d2 ,...,w dn Based on historical operational data of the exoskeleton and an assessment of the complexity of the current operational environment, a dynamic weight vector W is determined. d Using the kinematic forward kinematics formula P = f k (θ) and the inverse dynamics formula Combined with dynamic weight W d The formula for calculating the joint angle adjustment amount Δθ is as follows:

[0154]

[0155] Where J is the Jacobian matrix, and λ and μ are the control gains.

[0156] Specifically, such as Figure 6As shown, the system also includes an environmental parameter sensing and compensation module, used to monitor environmental parameters at the work site in real time, determine the impact of environmental factors on the insulator's movement based on a pre-established mapping model, and send compensation commands to the movement trajectory adjustment module when necessary. In the environmental parameter sensing and compensation module, when establishing the mapping model between environmental parameters and insulator movement interference, a multiple linear regression model based on particle swarm optimization is used. Let the environmental parameters be X = (x1, x2, ..., x...). m The data sample (X) includes wind speed, temperature, humidity, equipment vibration frequency, and insulator motion disturbance quantity Y. Through extensive experimental observations of insulator motion under different combinations of environmental parameters, data samples were obtained. i ,Y i The particle swarm optimization algorithm is used to find the optimal regression coefficients β = (β0, β1, ..., βN), where i = 1, 2, ..., N. m ), so that the objective function In the minimum, each particle in the particle swarm represents a set of regression coefficients, whose position and velocity are continuously updated in the search space. After multiple iterations of optimization, the final regression coefficients are determined, forming a mapping model.

[0157] For example, when the processor of the environmental parameter perception and compensation module determines that environmental factors significantly interfere with the movement of the insulator and may affect the operation accuracy of the exoskeleton, it calculates a correction suggestion for the motion parameters based on the mapping model and generates a compensation instruction. The compensation instruction includes correction values ​​for the joint angle adjustment amount, speed, and acceleration motion parameters. Then, the compensation instruction is sent to the motion trajectory adjustment submodule through the communication interface. The motion trajectory adjustment submodule corrects the calculated motion parameters according to the compensation instruction.

[0158] For example, a data preprocessing submodule is provided between the optical flow monitoring submodule and the deep learning analysis submodule. This submodule performs noise reduction and normalization processing on the motion vector information obtained by the optical flow method. For noise reduction, an adaptive median filtering algorithm is used. Based on the statistical characteristics of the motion vectors of pixels in the local region, the size of the filtering window is dynamically adjusted. Let the size of the local region window be S, and the mean value of the motion vectors within the window is calculated. and standard deviation σ V When σ V When the value is greater than the threshold T, the window S is increased; otherwise, the window S is decreased. Normalization is performed using the max-min normalization algorithm to normalize the motion vector V to the [0, 1] interval, as shown in the formula:

[0159]

[0160] Where V min and V max These are the minimum and maximum values ​​of the motion vector over a period of time, respectively.

[0161] For example, a motion smoothing submodule is provided between the motion trajectory adjustment submodule and the exoskeleton driving device. This motion smoothing submodule is used to smooth the control commands output by the motion trajectory adjustment submodule to avoid abrupt changes and impacts during exoskeleton movement. A smoothing algorithm based on Bézier curves is adopted, and the control command sequence is set as C = (c1, c2, ..., c n By determining the control points of the Bézier curve, the control command is converted into a smooth curve. The determination of the Bézier curve control points is based on the motion characteristics and operational requirements of the exoskeleton. According to the maximum acceleration and speed limits of the exoskeleton joints and the smoothness requirements of the insulator replacement operation, the position of the control points is adjusted. For the time parameter t∈[0,1], the smoothed control command C smooth (t) is calculated using the Bézier curve formula:

[0162]

[0163] Among them B i,n (t) is a Bernstein polynomial.

[0164] For example, the environmental parameter perception and compensation module also has an environmental risk early warning function. When the environmental parameters are detected to exceed the safety threshold range, it not only sends a compensation command to the motion trajectory adjustment submodule, but also sends an early warning signal to the operator. The safety threshold is determined by analyzing a large amount of power operation accident data and the performance test data of the exoskeleton in different environments. The early warning signal includes an audible and visual alarm, which reminds the operator by flashing the indicator lights on the exoskeleton and emitting an alarm sound through the speaker. At the same time, the system uploads the current environmental parameters and early warning information to the remote monitoring center in real time.

[0165] This embodiment provides an upper limb exoskeleton control system for assisting insulator replacement. By monitoring the dynamic changes of the insulator and its surrounding environment using optical flow, and combining deep learning analysis to accurately identify and analyze the moving insulator, the system can adjust the movement trajectory of the upper limb exoskeleton in a timely manner. This ensures that the upper limb exoskeleton can stably and accurately track and operate the insulator, thereby not only reducing the physical burden on the operator but also improving the safety and efficiency of the operation.

[0166] It should be noted that technical details not described in detail in this embodiment of the upper limb exoskeleton control system for assisting insulator replacement can be found in any embodiment of the present invention and applied to the upper limb exoskeleton control method for assisting insulator replacement as described above, and will not be repeated here.

[0167] It should be understood that the above are merely illustrative examples and do not constitute any limitation on the technical solutions of the present invention. In specific applications, those skilled in the art can make settings as needed, and the present invention does not impose any restrictions on this.

[0168] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of this invention. In practical applications, those skilled in the art can select some or all of the workflow to achieve the purpose of this embodiment according to actual needs, and no restrictions are imposed here.

[0169] Furthermore, it should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0170] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0171] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory (ROM) / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0172] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for controlling an upper limb exoskeleton to assist in replacing insulators, characterized in that, include: Image data of insulators and the field environment are collected, and motion vector information is obtained from the image data based on a preset optical flow method; The motion vector information is analyzed using a deep learning algorithm to obtain the motion state information of the insulator; Based on the dynamic fusion algorithm, the adjustment motion parameters of the upper limb exoskeleton are obtained by combining the motion state information of the insulator with the position information of the upper limb exoskeleton. Based on the adjusted motion parameters, control commands are sent to the upper limb exoskeleton; Based on the dynamic fusion algorithm, the adjustment motion parameters of the upper limb exoskeleton are obtained by combining the motion state information of the insulator with the position information of the upper limb exoskeleton. This includes: setting the desired position of the upper limb exoskeleton's end effector; obtaining the current position of the end effector and the angles of each joint; obtaining a dynamic weight vector based on the motion state information of the insulator and the historical operation data of the upper limb exoskeleton; obtaining the joint angle adjustment amount based on the desired position, current position, joint angles, and dynamic weight vector using the dynamic fusion algorithm; and using the joint angle adjustment amount as the adjustment motion parameter of the upper limb exoskeleton. The formula for calculating the joint angle adjustment Δθ is: ； Where J is the Jacobian matrix; λ and μ are the control gains, determined through experimental and simulation optimization; P d Let P be the desired position of the exoskeleton end effector, P be the current position, and θ be the angles of each joint (θ1, θ2, ..., θ). n ), W d This is a dynamic weight vector.

2. The method as described in claim 1, characterized in that, The acquisition of image data of the insulator and the field environment, and the obtaining of motion vector information based on the image data using a preset optical flow method, includes: Real-time acquisition of raw images of insulators and the field environment, and preprocessing of the raw images to obtain image data; Principal component analysis was performed on several insulator image samples to extract key feature vectors; Construct a direction weight matrix based on the correlation between the key feature vectors and the direction; Based on the aforementioned direction weight matrix, the gradient method in the preset optical flow method is improved to obtain the improved gradient method; Motion vector information is obtained based on the improved gradient method and image data.

3. The method as described in claim 2, characterized in that, The step of obtaining motion vector information based on the improved gradient method and image data includes: The first gradient, the second gradient, and the temporal gradient are calculated based on the improved gradient method and the image data. Calculate the first direction motion vector and the second direction motion vector based on the first direction gradient, the second direction gradient, the time direction gradient, and the direction weight matrix, respectively. Motion vector information is obtained based on the first direction motion vector and the second direction motion vector.

4. The method as described in claim 1, characterized in that, The analysis of the motion vector information based on the deep learning algorithm to obtain the motion state information of the insulator includes: Constructing a deep learning model based on a multi-layer cascaded attention network; The motion vector information is input into the deep learning model to identify insulators in motion and obtain key motion features related to the insulators; wherein, the key motion features include the insulator's motion direction and velocity information; The motion state information of the insulator is generated based on the key motion characteristics.

5. The method as described in claim 1, characterized in that, The process of obtaining a dynamic weight vector based on the insulator's motion state information and the historical operational data of the upper limb exoskeleton includes: Acquire historical operational data of the upper limb exoskeleton under a preset environment; Get real-time information on current wind direction and speed; The complexity of the working environment is assessed by combining the insulator's motion state information with the historical operation data and the current wind direction and speed information, resulting in a dynamic weight vector.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: Establish a mapping model between environmental parameters and insulator motion disturbance; Real-time monitoring of environmental parameters at the work site; The environmental parameters are analyzed in real time to obtain the trend of environmental parameter changes, and the trend of environmental parameter changes is used to determine whether it will interfere with the movement of the insulator. If so, then a compensation instruction is generated based on the mapping model and the environmental parameters; The motion trajectory of the upper limb exoskeleton is adjusted and corrected based on the compensation command.

7. The method as described in claim 6, characterized in that, The establishment of the mapping model between environmental parameters and insulator motion disturbance includes: The original environmental parameters are generated based on wind speed, temperature, humidity, and equipment vibration frequency. Data samples were generated based on experimental observations of insulator motion under different combinations of original environmental parameters; Based on the data samples, a multivariate linear regression model based on particle swarm optimization is trained to obtain a mapping model between environmental parameters and insulator motion disturbance.

8. A control system for an upper limb exoskeleton used to assist in replacing insulators, characterized in that, include: The optical flow monitoring module is used to collect image data of the insulator and the field environment, and to obtain motion vector information based on the image data according to a preset optical flow method; The insulator analysis module is used to analyze the motion vector information based on deep learning algorithms to obtain the motion state information of the insulator; The trajectory adjustment module is used to obtain the adjustment motion parameters of the upper limb exoskeleton based on the motion state information of the insulator and the position information of the upper limb exoskeleton according to the dynamic fusion algorithm. The motion control module is used to send control commands to the upper limb exoskeleton based on the adjusted motion parameters; Based on the dynamic fusion algorithm, the adjustment motion parameters of the upper limb exoskeleton are obtained by combining the motion state information of the insulator with the position information of the upper limb exoskeleton. This includes: setting the desired position of the upper limb exoskeleton's end effector; obtaining the current position of the end effector and the angles of each joint; obtaining a dynamic weight vector based on the motion state information of the insulator and the historical operation data of the upper limb exoskeleton; obtaining the joint angle adjustment amount based on the desired position, current position, joint angles, and dynamic weight vector using the dynamic fusion algorithm; and using the joint angle adjustment amount as the adjustment motion parameter of the upper limb exoskeleton. The formula for calculating the joint angle adjustment Δθ is: ； Where J is the Jacobian matrix; λ and μ are the control gains, determined through experimental and simulation optimization; P d Let P be the desired position of the exoskeleton end effector, P be the current position, and θ be the angles of each joint (θ1, θ2, ..., θ). n ), W d This is a dynamic weight vector.

9. An electronic device, characterized in that, The electronic device includes: a memory, a processor, and an upper limb exoskeleton control program for assisting insulator replacement stored in the memory and executable on the processor, the upper limb exoskeleton control program for assisting insulator replacement being configured to implement the upper limb exoskeleton control method for assisting insulator replacement as described in any one of claims 1 to 7.

Images