Power assisting method and device for assisting iron tower climbing

Abstract

The invention discloses a power assisting method and device for assisting iron tower climbing. The method comprises the steps that S101, a lower limb muscle deformation signal and a plantar pressure distribution signal of a target object are collected; predicting a climbing intention of the target object through the lower limb muscle deformation signal; judging a climbing stage of the target object through the plantar pressure signal; s102, generating an initial expected joint trajectory according to the climbing intention and the climbing stage; setting an initial joint stiffness damping parameter according to the initial expected joint trajectory; sensing environment information around the iron tower in real time, and dynamically adjusting an expected joint track and a joint rigidity damping coefficient; s103, the joint stiffness damping coefficient is input into a dielectric elastic soft body; the dielectric elastic soft body simulates the muscle activation degree of a target object and drives all joints in the soft body to output assisting torque to the target object. According to the technical scheme of the embodiment of the invention, high efficiency and safety of iron tower climbing operation are ensured.

Description

Technical Field

[0001] The present invention relates to the field of tower climbing assistance, and in particular to a method and device for assisting tower climbing. Background Art

[0002] With the rapid development of infrastructure such as electricity and communications, the demand for tower climbing operations is increasing. Tower climbing is a high-risk, high-intensity task. Operators need to complete high-altitude operations in a complex environment, which places extremely high demands on physical strength and safety.

[0003] Traditional climbing methods rely primarily on manpower. Workers not only have to overcome their own gravity but also contend with the complexities of the tower structure (such as uneven crossarm spacing and corroded areas). This results in significant physical exertion, low efficiency, and a high level of safety hazards. In recent years, lower-limb exoskeleton assist devices have been gradually introduced as an auxiliary tool for tower climbing operations. These devices provide additional assistance to workers through mechanical structures, reducing the physical burden during the climb.

[0004] However, existing lower limb exoskeleton assist devices use motors or hydraulic drives, which have stiff output and delayed response, and cannot simulate the supple characteristics of human muscles, resulting in uncoordinated climbing movements and increased physical exertion of users. Moreover, due to the lack of active perception of tower structures (such as rust and cracks) and environmental risks, it is difficult to achieve safe path planning. Summary of the Invention

[0005] The embodiments of the present invention aim to solve at least one of the deficiencies in the above-mentioned technologies, and ensure the efficiency and safety of tower climbing operations by sensing the climbing intentions of operators and the environment in which the tower is climbed.

[0006] In a first aspect, the present invention provides a method for assisting tower climbing, comprising:

[0007] S101, collecting lower limb muscle deformation signals and plantar pressure distribution signals of a target subject; predicting the target subject's climbing intention based on the lower limb muscle deformation signals; and determining the target subject's climbing stage based on the plantar pressure signals;

[0008] S102, generating an initial expected joint trajectory according to the climbing intention and the climbing stage; setting initial joint stiffness and damping parameters according to the initial expected joint trajectory; sensing environmental information around the tower in real time, and dynamically adjusting the expected joint trajectory and joint stiffness and damping coefficients;

[0009] S103, inputting the joint stiffness damping coefficient into the dielectric elastic soft body; the dielectric elastic soft body simulates the muscle activation of the target object and drives each joint in the soft body to output an assist torque to the target object.

[0010] Furthermore, S101 specifically includes:

[0011] S201, capturing lower limb muscle deformation signals of the quadriceps femoris and calf muscles of the target subject in real time through a flexible electronic skin array; extracting intrinsic mode function features of the lower limb muscle deformation signals by performing CEEMDAN signal decomposition on the lower limb muscle deformation signals;

[0012] S202, inputting the intrinsic mode function into a preset time-dynamic model, wherein the output of the time-dynamic model is an estimated result of muscle activation; and obtaining muscle activation information based on the estimated result of muscle activation information;

[0013] S203, collecting pressure distribution information through a plantar pressure sensor; inputting the muscle activation information and the pressure distribution information into a preset dual-channel LSTM neural network, the output of which is the target object's climbing intention and climbing stage.

[0014] Furthermore, S102 specifically includes:

[0015] S301: Formulate an initial climbing strategy based on the climbing intention and the climbing stage; input the initial climbing strategy into a two-layer optimization model, wherein the upper layer of the model aims to generate initial expected joint trajectories based on the initial climbing strategy, and the lower layer of the model aims to set initial joint stiffness and damping coefficients based on the initial expected joint trajectories;

[0016] S302: Establishing a bionic compound eye vision network and a spiking neural network; obtaining multispectral image information of the iron tower through the bionic compound eye vision network; and identifying structural hidden danger information of the iron tower through the spiking neural network;

[0017] S303, based on the multispectral image information of the tower and the structural hidden danger information of the tower, the climbing strategy is corrected in real time, and the corrected climbing strategy is input into the two-layer optimization model to dynamically adjust the expected joint trajectory and the joint stiffness damping coefficient.

[0018] Furthermore, S303 specifically includes:

[0019] S401, adjusting the adhesion between the target object and the tower according to the structural hidden danger information of the tower, and generating a new climbing strategy;

[0020] S402: Identify, using a surface acoustic wave sensor, whether the new climbing strategy poses a structural resonance risk; if not, use the new climbing strategy as the current climbing strategy; if so, trigger tactile feedback and transcranial electrical stimulation to modify the new climbing strategy, and use the modified climbing strategy as the current climbing strategy;

[0021] S403: Input the current climbing strategy into the two-layer optimization model to dynamically adjust the expected joint trajectory and joint stiffness damping coefficient.

[0022] Furthermore, the determination condition for the structural resonance risk is: detecting the resonance frequency offset Δf by a surface acoustic wave sensor, and when Δf>5%, there is a structural resonance risk.

[0023] Furthermore, it also includes:

[0024] The climbing strategy is divided into a real-time control subgroup and a path optimization subgroup by using a swarm optimization algorithm, and different motion modes are set for different target objects;

[0025] The real-time control subgroup is used to optimize the assist torque of the target object, and the path optimization subgroup is used to optimize the joint trajectory of the target object.

[0026] Furthermore, it also includes:

[0027] Adapt the motion patterns of different target objects through transfer learning methods, and balance the relationship between work efficiency, work safety, and work energy consumption through the reward function;

[0028] The reward function is defined as:

[0029] R = w1×efficiency+w2×safety-w3×energy consumption

[0030] Among them, w1, w2, and w3 are dynamic weight coefficients.

[0031] In a second aspect, the present invention provides a power assisting device for assisting tower climbing, comprising:

[0032] The acquisition module is used to collect the target object's lower limb muscle deformation signals and plantar pressure distribution signals; predict the target object's climbing intention based on the lower limb muscle deformation signals; and determine the target object's climbing stage based on the plantar pressure signals.

[0033] A generation module is configured to generate an initial expected joint trajectory according to the climbing intention and the climbing stage; set initial joint stiffness and damping parameters according to the initial expected joint trajectory; and sense environmental information around the tower in real time to dynamically adjust the expected joint trajectory and joint stiffness and damping coefficients;

[0034] The power assist module is used to input the joint stiffness damping coefficient into the dielectric elastic soft body; the dielectric elastic soft body simulates the muscle activation of the target object and drives each joint in the soft body to output a power assist torque to the target object.

[0035] In a third aspect, an embodiment of the present invention provides an electronic device, comprising:

[0036] at least one processor; and

[0037] a memory communicatively connected to the at least one processor; wherein,

[0038] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the method for assisting tower climbing according to any embodiment of the present invention.

[0039] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a processor to implement the method of assisting tower climbing as described in any embodiment of the present invention when executed.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The embodiment of the present invention collects lower limb muscle deformation signals and plantar pressure distribution signals, analyzes the climbing intention and climbing stage of the target object, generates initial expected joint trajectories and initial joint stiffness damping parameters; and dynamically adjusts the expected joint trajectories and joint stiffness damping coefficients according to environmental information around the tower; finally, the dielectric elastic soft body drives each joint in the soft body to output an assist torque to the target object, thereby achieving smooth assist for the target object and ensuring the safety and efficiency of the operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only preferred embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 A schematic flow chart of a method for assisting tower climbing provided in the first embodiment of the present invention;

[0044] Figure 2 A schematic structural diagram of a power-assisting device for assisting tower climbing provided in a second embodiment of the present invention;

[0045] Figure 3 This is a structural diagram of an electronic device provided in Example 3 of the present invention. DETAILED DESCRIPTION

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0047] Example 1

[0048] Figure 1 This is a flow chart of an assisting method for assisting tower climbing provided in the first embodiment of the present invention. This embodiment is applicable to situations where high-altitude operations are completed in complex environments. The method can be performed by an assisting device for assisting tower climbing provided in the second embodiment of the present invention, which can be implemented in software and / or hardware.

[0049] like Figure 1 As shown, the method specifically includes:

[0050] S101, collecting lower limb muscle deformation signals and plantar pressure distribution signals of a target object; predicting the target object's climbing intention based on the lower limb muscle deformation signals; and determining the target object's climbing stage based on the plantar pressure signals.

[0051] It should be noted that the target subjects are workers performing high-altitude tower climbing operations. Before climbing, they need to wear a tower climbing assist device. This tower climbing assist device is an innovative lower-limb exoskeleton assist device. The target subjects' lower-limb muscle deformation signals can be collected using surface electromyography sensors, and the target subjects' plantar pressure signals can be collected using plantar pressure sensors. Climbing intentions include: not climbing, climbing upward, and climbing downward. Climbing stages include: preparing to climb, climbing, and completing the climb.

[0052] It is understood that various physiological signal acquisition components such as surface electromyography sensors and plantar pressure sensors can be installed inside the tower climbing assist device to collect multiple modal physiological signals from the target object. For example, a temperature sensor can be set to collect the target object's body temperature.

[0053] In a specific application scenario, for example, workers need to inspect and repair an old, 60-meter-high transmission tower. The tower's crossarms are corroded in many places, and some bolts are loose. The workers need to wear the tower climbing assist device of Example 2 of the present invention to perform the climbing task.

[0054] During the donning process of the device, the operator stands next to a special donning assistive frame and starts by aligning the leg part of the power device for assisting tower climbing with his or her lower limbs; by adjusting the straps, ensure that the hip, knee and ankle joints are tightly fitted and comfortable; the straps are elastic and adjustable and can adapt to different body shapes; then, fix the waist support component so that it fits firmly to the waist and provides overall support. The entire donning process takes about 3 minutes.

[0055] After the device is put on, the operator turns on the power of the device and performs initial settings on the handheld control terminal. The basic dimensions and adaptation parameters of the exoskeleton are set according to the operator's physical parameters, such as height 175 cm and weight 75 kg. At the same time, the surface electromyography sensors and plantar pressure sensors are calibrated to ensure accurate data collection. For example, the surface electromyography sensors need to be tightly attached to the surface of key muscle groups such as the quadriceps and hamstrings, and signal conduction is enhanced by applying special conductive gel.

[0056] Furthermore, S101 specifically includes:

[0057] S201, real-time acquisition is performed through the flexible electronic skin array to capture the lower limb muscle deformation signals of the quadriceps femoris and calf muscles of the target object; by performing CEEMDAN signal decomposition on the lower limb muscle deformation signals, the intrinsic mode function characteristics of the lower limb muscle deformation signals are extracted.

[0058] Among them, the flexible electronic skin array is made of graphene / liquid metal composite material, and quantifies muscle contraction strength through dual-mode sensing of resistive strain and capacitive coupling.

[0059] Because surface electromyography (EMG) sensors are susceptible to electromagnetic interference, sweat, and environmental vibration when identifying motion intent, resulting in significant errors in intent recognition, embodiments of the present invention integrate surface electromyography (EMG) sensors within a flexible electronic skin array to detect muscle electrical signals. Furthermore, the wearability and flexibility of the flexible electronic skin allow it to conform to the skin better, improving the accuracy and comfort of surface electromyography (EMG) signal detection.

[0060] Specifically, as workers approached the tower ladder and prepared to begin climbing, surface electromyography sensors collected initial contraction signals from the quadriceps and calf muscles. These signals were then decomposed using the CEEMDAN algorithm to extract key intrinsic mode function features.

[0061] CEEMDAN is a signal decomposition method based on adaptive noise control. During the extraction of each order intrinsic mode function (IMF), an adaptive noise factor related to the current IMF is added. This adaptive noise factor is calculated as a weighted sum of the residual of the previous order IMF and the current noise factor. As IMFs are extracted, the impact of the noise gradually decreases, thus avoiding the problem of residual noise. Furthermore, during the extraction of each order IMF, the algorithm adaptively adjusts the amplitude and frequency of the noise factor based on the current residual, ensuring convergence.

[0062] Any signal is composed of several intrinsic mode functions (IMFs). At any time, a signal can contain several intrinsic mode functions. If the intrinsic mode functions overlap with each other, a composite signal is formed.

[0063] S202: Inputting the intrinsic mode function into a preset time-dynamic model, the output of the time-dynamic model is the estimated result of the muscle activation; and obtaining muscle activation information based on the estimated result of the information muscle activation.

[0064] Among them, muscle activation can stably reflect muscle contraction; muscle activation information includes: the average absolute value, root mean square value and integral absolute value of muscle activation, etc.; the average absolute value (MAV) refers to the average absolute value of muscle activation; the root mean square value (RMS) refers to the square root of the average square of muscle activation; the integral absolute value (IAV) refers to the integral value of the absolute value of muscle activation.

[0065] In this embodiment of the present invention, extensive model training is used to establish a relationship between the IMF signature and muscle activation. Once the IMF signature is input into the model, the model solves the IMF signature to estimate muscle activation. Based on this muscle activation, the model then calculates the average absolute value, root mean square value, and integrated absolute value of the muscle activation.

[0066] S203, collecting pressure distribution information through the plantar pressure sensor; inputting the muscle activation information and the pressure distribution information into a preset dual-channel LSTM neural network, the output of which is the target object's climbing intention and climbing stage.

[0067] Specifically, the first channel of a dual-channel LSTM neural network processes muscle activation information to predict the target's climbing intention, while the second channel analyzes the trajectory of the center of plantar pressure to determine the target's climbing phase. Through extensive model training, the system establishes a link between muscle activation information and climbing intention, and between pressure distribution information and climbing phase. By introducing an attention mechanism into the dual-channel LSTM neural network, the system fuses muscle activation and pressure distribution information to determine the operator's movement intention.

[0068] S102, generating an initial expected joint trajectory according to the climbing intention and climbing stage; setting initial joint stiffness and damping parameters according to the initial expected joint trajectory; sensing the environmental information around the tower in real time, and dynamically adjusting the expected joint trajectory and joint stiffness and damping coefficients.

[0069] Environmental information surrounding the tower includes its multispectral image and structural hazard information. Joint stiffness and damping coefficients include stiffness and damping coefficients. The stiffness coefficient represents the relationship between force and displacement, while the damping coefficient represents the relationship between force and velocity.

[0070] Furthermore, S102 specifically includes:

[0071] S301: Formulate an initial climbing strategy based on the climbing intention and climbing stage; input the initial climbing strategy into a two-layer optimization model, where the upper layer of the model aims to generate an initial expected joint trajectory based on the initial climbing strategy, and the lower layer of the model aims to set the initial joint stiffness damping coefficient based on the initial expected joint trajectory.

[0072] Preferably, through cloud-based collaborative technology, the edge is responsible for collecting muscle activation information and pressure distribution information in real time and performing real-time impedance control, while the cloud deploys a dual-channel LSTM neural network and a two-layer optimization model to formulate climbing strategies and optimize the global climbing path.

[0073] The climbing strategy includes the climbing path, among other things. Through a two-layer optimization model, the exoskeleton gently assists the worker in lifting their legs and taking the first step. The upper layer of the model utilizes the Model Predictive Control (MPC) algorithm, while the lower layer utilizes the Particle Swarm Optimization (PSO). The MPC algorithm solves a finite-time optimization problem to determine the optimal control input at the current moment. The first value of this control input is then applied to the system, and the process is repeated at the next sampling moment, resulting in continuous rolling optimization. The PSO utilizes information sharing among individuals within a group, allowing the group's movement to evolve from disorder to order within the problem-solving space, thereby obtaining a feasible solution to the problem.

[0074] S302: Establish a bionic compound eye vision network and a pulse neural network; obtain multispectral image information of the iron tower through the bionic compound eye vision network; and identify structural hidden danger information of the iron tower through the pulse neural network.

[0075] Specifically, a bionic compound eye vision network is used to obtain multispectral images of the tower, and a pulse neural network is used to identify hidden dangers of structural corrosion, cracks, and loose bolts, and the risk level is marked; thereby completing the perception of the surrounding environment of the climber.

[0076] The tower climbing assistance system incorporates environmental perception components such as lidar and cameras, forming a bionic compound eye visual network and a spiking neural network. For example, when a worker climbs approximately 10 meters and encounters an area with large crossarm spacing, the lidar scans the surroundings in real time, and using the improved RANSAC algorithm, the point cloud data quickly identifies the edges of the tower's crossarm structure. The camera provides a 360-degree field of view, supporting ultraviolet light detection of oxidation signatures and infrared temperature differential detection. The spiking neural network processes multispectral data in real time using spatiotemporal pulse coding, outputting the coordinates of the risk area and a confidence level.

[0077] The RANSAC algorithm is an algorithm used in computer vision. It can calculate the mathematical model parameters of a data set containing abnormal data, thereby obtaining valid sample data. The spiking neural network (SNN) is a deep learning algorithm with a strong biological foundation. It can process biological stimuli by mimicking the working principles of biological neurons, explain the brain's intelligent behavior, and interface with the biological nervous system.

[0078] S303, based on the multispectral image information of the tower and the structural hidden danger information of the tower, the climbing strategy is corrected in real time, and the corrected climbing strategy is input into the two-layer optimization model to dynamically adjust the expected joint trajectory and joint stiffness damping coefficient.

[0079] Specifically, cubic spline interpolation is used to generate a safe climbing path. During this process, the operator's body posture changes, and the surface electromyography signal and plantar pressure data also change accordingly. These changes are continuously and accurately perceived through the multimodal data fusion algorithm, and the dynamic adaptive control algorithm adjusts the joint stiffness and damping parameters in a timely manner. For example, when the operator needs to go around an obstacle sideways, the hip and knee joints of the exoskeleton dynamically adjust the stiffness and damping according to the algorithm instructions, providing appropriate assistance and support to ensure smooth and safe movements. At the same time, the hardware acceleration module ensures the fast operation of the algorithm, the 1kHz actuator current loop control enables the magnetorheological actuator to respond quickly, the 100Hz impedance parameter adjustment ensures comfortable joint movement, and the 10Hz path planning update fine-tunes the climbing path according to real-time environmental changes.

[0080] Furthermore, S303 specifically includes:

[0081] S401, according to the structural hidden danger information of the tower, the adhesion between the target object and the tower is adjusted to generate a new climbing strategy.

[0082] S402: Using a surface acoustic wave sensor, identify whether the new climbing strategy poses a structural resonance risk. If not, use the new climbing strategy as the current climbing strategy. If so, trigger tactile feedback and transcranial electrical stimulation to correct the new climbing strategy, and use the corrected climbing strategy as the current climbing strategy.

[0083] S403: Input the current climbing strategy into the two-layer optimization model to dynamically adjust the expected joint trajectory and joint stiffness damping coefficient.

[0084] The criteria for determining structural resonance risk are: Surface acoustic wave sensors detect the resonant frequency offset Δf. When Δf exceeds 5%, a Level 3 alarm is triggered, indicating the presence of a structural resonance risk. The adhesion strength of the magnetorheological fluid can be controlled by current, with an adjustable range of 50-500N. The surface acoustic wave sensors can detect the resonant frequency offset Δf of the tower's crossarms.

[0085] Based on information about the tower's structural hazards, the device adjusts the adhesion between the target object and the tower. For example, when a user steps over a corroded crossarm, the device's knee joint dynamically adjusts the electric field strength (2kV to 4kV) based on the curvature of the path, smoothly transitioning the contraction rate from 20% to 45% and increasing the assist torque from 50N·m to 120N·m, simulating the natural force curve of muscle. In the foot joint, the magnetorheological current increases from 0.5A to 1.5A, boosting the adhesion from 100N to 300N, preventing the foot from slipping on the corroded surface.

[0086] When there is a risk of structural resonance, tactile feedback and transcranial electrical stimulation are triggered. For example, a surface acoustic wave sensor detects a crossarm resonant frequency shift of Δf = 6% (baseline frequency 1kHz → 940Hz), which is determined to be a risk of structural instability. The tactile feedback array triggers 80Hz high-frequency vibrations at the foot, and transcranial electrical stimulation (tES) applies a 1.5mA current to stimulate the user's motor cortex. The current in the foot portion of the device suddenly increases to 2A, and the adhesion force is locked at 500N to provide rigid support. At the same time, the drive portion of the device contracts in the opposite direction, outputting a torque of 150N·m to assist the user in moving sideways out of the danger zone.

[0087] In an embodiment of the present invention, foot stability is enhanced by a magnetorheological adhesion-locking dual-state switching mechanism, and human-machine collaborative risk avoidance is achieved by triggering tactile feedback and transcranial electrical stimulation.

[0088] Furthermore, in order to set different motion modes for different target objects, the climbing strategy can be divided into a real-time control subgroup and a path optimization subgroup through the swarm optimization algorithm to set the corresponding motion mode for the target object; among them, the real-time control subgroup is used to optimize the assist torque of the target object, and the path optimization subgroup is used to optimize the joint trajectory of the target object.

[0089] Furthermore, in order to balance the relationship between work efficiency, work safety, and work energy consumption, the transfer learning method in the meta-reinforcement learning framework is used to adapt the motion patterns of different target objects. The reward function is defined as:

[0090] R = w1×efficiency+w2×safety-w3×energy consumption

[0091] Among them, w1, w2, and w3 are dynamic weight coefficients.

[0092] S103, inputting the joint stiffness damping coefficient into the dielectric elastic soft body; the dielectric elastic soft body simulates the muscle activation of the target object and drives each joint in the soft body to output an assist torque to the target object.

[0093] Optionally, based on the output of compliant joint torque by the dielectric elastic soft body, a friction nanogenerator is simultaneously used to convert mechanical vibration energy into electrical energy.

[0094] Optional, dielectric elastic soft body adopts stacked structure, voltage response rate <10ms, output density up to 30N / cm 2 The triboelectric nanogenerator is embedded in the contact surface between the joint hinge and the sole of the foot, and its output power satisfies the following formula:

[0095] P TENG =k·f 1.2 ·A 0.8

[0096] Where f is the vibration frequency, A is the amplitude, and k is the material coupling coefficient.

[0097] Optionally, the working voltage of the dielectric elastic soft body is 0.5-5kV, and the relationship between its deformation rate ε and electric field intensity E satisfies:

[0098] ε=λ·E 2 +μ·E

[0099] Where λ and μ are the nonlinear coefficients of the material.

[0100] Furthermore, a self-consistent energy model can be configured during climbing. For example, a triboelectric nanogenerator (TENG) at the knee joint hinge converts impact vibration (frequency 20Hz, amplitude 2mm) into electrical energy, recovering 0.15J of energy per step and storing it in a supercapacitor. After unloading the device, the deformation rate of the drive unit (threshold <5%) and the state of the magnetorheological fluid (viscosity error <3%) are automatically detected, generating a maintenance report.

[0101] The embodiment of the present invention extracts the intrinsic mode function characteristics in the signal to analyze the muscle activation information of the operator, and at the same time collects the operator's plantar pressure distribution information through the plantar pressure sensor; it realizes high-precision recognition of the user's movement intention, and actively detects potential risks such as rust and cracks in the tower structure to ensure the safety of the climbing path; through the magnetorheological adhesion-locking mechanism, it can dynamically adjust the foot support force to adapt to the needs of different climbing stages, greatly enhancing the stability and adaptability of the system; by adjusting the climbing strategy in real time, a closed-loop control is formed for the entire climbing process to ensure efficient and safe operations.

[0102] Example 2

[0103] Figure 2 This is a schematic diagram of the structure of a power assist device for assisting tower climbing provided by the second embodiment of the present invention. Figure 2 As shown, the device specifically includes:

[0104] The acquisition module 100 is used to collect the target object's lower limb muscle deformation signals and plantar pressure distribution signals; predict the target object's climbing intention based on the lower limb muscle deformation signals; and determine the target object's climbing stage based on the plantar pressure signals.

[0105] The generation module 200 is used to generate an initial expected joint trajectory according to the climbing intention and the climbing stage; set the initial joint stiffness and damping parameters according to the initial expected joint trajectory; and sense the environmental information around the tower in real time to dynamically adjust the expected joint trajectory and the joint stiffness and damping coefficient.

[0106] The power assist module 300 is used to input the joint stiffness damping coefficient into the dielectric elastic soft body; the dielectric elastic soft body simulates the muscle activation of the target object and drives each joint in the soft body to output a power assist torque to the target object.

[0107] Optionally, the device can be equipped with a variety of sensing units, such as a flexible electronic skin array, a bionic compound eye vision system, and a nine-axis inertial measurement unit (IMU).

[0108] Optionally, the device can be equipped with multiple bionic drive units, for example: the hip, knee, and ankle joints are integrated with dielectric elastomer drive units and friction nanogenerators (TENGs), and the supporting skeleton adopts a topologically optimized magnesium alloy hollow structure, etc.

[0109] Optionally, the device can be equipped with a decision control unit, for example: an FPGA and a real-time operating system (RTOS) at the edge, a quantum annealing coprocessor connected to the cloud, etc.

[0110] Optionally, the device can be equipped with a safety protection unit, such as an integrated magnetorheological adhesive pad, a surface acoustic wave sensor, a tactile feedback array, and the like.

[0111] In an embodiment of the present invention, through the acquisition module, multiple modal signals of the target object are obtained to analyze the climbing intention and climbing stage; through the generation model, the expected joint trajectory and joint stiffness damping coefficient are dynamically adjusted; through the power assist module, the target object is flexibly assisted, providing a more intelligent and safe solution for high-altitude operations.

[0112] Example 3

[0113] Figure 3 1 is a schematic diagram of the structure of an electronic device that implements the method for assisting tower climbing according to an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0114] like Figure 3 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0115] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0116] Processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any other suitable processor, controller, microcontroller, etc. Processor 11 executes the various methods and processes described above, such as the power-assisted method for assisting tower climbing.

[0117] In some embodiments, the method for assisting tower climbing can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for assisting tower climbing described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to execute the method for assisting tower climbing in any other suitable manner (e.g., via firmware).

[0118] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0119] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0120] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0121] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0122] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0123] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0124] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0125] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for assisting tower climbing, characterized in that: include: S101, collecting lower limb muscle deformation signals and plantar pressure distribution signals of a target subject; and predicting the target subject's climbing intention based on the lower limb muscle deformation signals; determining the climbing stage of the target object according to the plantar pressure signal; S102, generating an initial expected joint trajectory according to the climbing intention and the climbing stage; and setting initial joint stiffness and damping parameters according to the initial expected joint trajectory; Real-time perception of the environmental information around the tower, dynamic adjustment of the desired joint trajectory and joint stiffness and damping coefficient; S103, inputting the joint stiffness damping coefficient into the dielectric elastic soft body; The dielectric elastic soft body simulates the muscle activation of the target object and drives each joint in the soft body to output a power-assisting torque to the target object.

2. The method according to claim 1, characterized in that S101 specifically includes: S201, capturing lower limb muscle deformation signals of the quadriceps femoris and calf muscles of the target subject in real time through a flexible electronic skin array; extracting intrinsic mode function features of the lower limb muscle deformation signals by performing CEEMDAN signal decomposition on the lower limb muscle deformation signals; S202, inputting the intrinsic mode function into a preset time-dynamic model, wherein the output of the time-dynamic model is an estimated result of muscle activation; and obtaining muscle activation information based on the estimated result of muscle activation information; S203, collecting pressure distribution information through a plantar pressure sensor; inputting the muscle activation information and the pressure distribution information into a preset dual-channel LSTM neural network, the output of which is the target object's climbing intention and climbing stage.

3. The method according to claim 1, characterized in that S102 specifically includes: S301: Formulate an initial climbing strategy based on the climbing intention and the climbing stage; input the initial climbing strategy into a two-layer optimization model, wherein the upper layer of the model aims to generate initial expected joint trajectories based on the initial climbing strategy, and the lower layer of the model aims to set initial joint stiffness and damping coefficients based on the initial expected joint trajectories; S302: Establishing a bionic compound eye vision network and a spiking neural network; obtaining multispectral image information of the iron tower through the bionic compound eye vision network; and identifying structural hidden danger information of the iron tower through the spiking neural network; S303, based on the multispectral image information of the tower and the structural hidden danger information of the tower, the climbing strategy is corrected in real time, and the corrected climbing strategy is input into the two-layer optimization model to dynamically adjust the expected joint trajectory and the joint stiffness damping coefficient.

4. The method according to claim 3, characterized in that S303 specifically includes: S401, adjusting the adhesion between the target object and the tower according to the structural hidden danger information of the tower, and generating a new climbing strategy; S402: Identify, using a surface acoustic wave sensor, whether the new climbing strategy poses a structural resonance risk; if not, use the new climbing strategy as the current climbing strategy; if so, trigger tactile feedback and transcranial electrical stimulation to modify the new climbing strategy, and use the modified climbing strategy as the current climbing strategy; S403: Input the current climbing strategy into the two-layer optimization model to dynamically adjust the expected joint trajectory and joint stiffness damping coefficient.

5. The method according to claim 4, characterized in that The determination condition for the structural resonance risk is: detecting the resonance frequency offset Δf by a surface acoustic wave sensor, and when Δf>5%, there is a structural resonance risk.

6. The method according to claim 4, characterized in that Also includes: The climbing strategy is divided into a real-time control subgroup and a path optimization subgroup by using a swarm optimization algorithm, and different motion modes are set for different target objects; The real-time control subgroup is used to optimize the assist torque of the target object, and the path optimization subgroup is used to optimize the joint trajectory of the target object.

7. The method according to claim 6, characterized in that Also includes: Adapt the motion patterns of different target objects through transfer learning methods, and balance the relationship between work efficiency, work safety, and work energy consumption through the reward function; The reward function is defined as: R = w1×efficiency+w2×safety-w3×energy consumption Among them, w1, w2, and w3 are dynamic weight coefficients.

8. A power assist device for assisting tower climbing, characterized in that: The device is configured to implement the method according to any one of claims 1 to 7, and the device includes: The acquisition module is used to collect the target object's lower limb muscle deformation signals and plantar pressure distribution signals; predict the target object's climbing intention through the lower limb muscle deformation signals; and determine the target object's climbing stage through the plantar pressure signals. The generation module is used to generate an initial expected joint trajectory according to the climbing intention and the climbing stage; set the initial joint stiffness and damping parameters according to the initial expected joint trajectory; and perceive the environmental information around the tower in real time to dynamically adjust the expected joint trajectory and the joint stiffness and damping coefficient. The power assist module is used to input the joint stiffness damping coefficient into the dielectric elastic soft body; the dielectric elastic soft body simulates the muscle activation of the target object and drives each joint in the soft body to output a power assist torque to the target object.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the steps of the assisting method for assisting iron tower climbing according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the steps of the assisting method for assisting tower climbing according to any one of claims 1 to 7 when executed.

Images