Master-slave control and insulation interlocking protection system and method for live-line robot

Abstract

The application relates to the technical field of mechanical hands, and particularly discloses a master-slave control and insulation interlocking protection system and method for a live working robot. The system comprises a master hand control terminal, a slave end working robot, a high-speed communication link and an insulation safety interlocking center, realizes high-precision motion tracking, soft obstacle avoidance and emergency braking through integration of multi-sensor data, real-time safety distance calculation and multi-modal control strategy, and simultaneously improves operation safety and reliability in combination with bidirectional force feedback.

Description

Technical Field

[0001] This invention belongs to the field of robotic arm technology, specifically relating to a master-slave control and insulation interlocking protection system and method for a live-line working robot. Background Technology

[0002] In the field of live-line working in power systems, the application of robotics technology has become a key development direction for improving operational safety and efficiency. Live-line working robots, by replacing manual labor in direct contact with high-voltage live components, can reduce the risk of electric shock and the intensity of labor for workers. The core of this approach lies in enabling precise and reliable operation of the robot in high-voltage environments.

[0003] Master-slave control technology is an important way to realize remote operation of robots. Operators issue commands remotely through the master controller, and the slave robot arm replicates the operation in the high-voltage field. However, existing technologies face significant challenges in practical applications: traditional master-slave control architectures have inherent operation delays and trajectory jitter, especially when facing complex and unstructured live-line working environments, such as when overhead power lines are subject to continuous light wind vibrations or when the work space contains various obstacles, making it difficult for the slave robot arm to maintain a smooth and precise motion response.

[0004] The inaccuracy of existing motion control technology directly leads to deviations in the real-time calculation of the minimum safe distance between the robot and charged parts. This deviation is amplified dramatically in high-voltage environments, increasing the risk of accidental collisions between the robotic arm and charged components. The coupling effect of operational delay and trajectory jitter not only reduces the efficiency of complex processes but also poses a serious threat to the safety and reliability of the entire operation. Therefore, there is an urgent need for a master-slave control system and method that can achieve high precision, low latency, and has an insulation safety interlocking mechanism. Summary of the Invention

[0005] The present invention aims to provide a master-slave control and insulation interlock protection system and method for live-line working robots, so as to solve the problems of operation delay, trajectory jitter and the resulting deviation in safety distance calculation and insulation safety interlock failure in the existing master-slave control architecture.

[0006] This invention provides a master-slave control and insulation interlocking protection system for a live-line working robot. The system includes a master control terminal, a slave working robot, a high-speed communication link, and an insulation safety interlocking center. The master control terminal is deployed in a zero-potential safety area and has a built-in high-precision six-dimensional force feedback device and attitude sensing module for real-time acquisition of the operator's operating intentions and generation of master-end posture command streams.

[0007] The end-effector robot is deployed at the high-voltage electric field work site. Its robotic arm joints are equipped with multi-stage reduction mechanisms and absolute photoelectric encoders, and the end effector integrates a miniature lidar and an electric field strength sensor array.

[0008] The high-speed communication link uses a bidirectional fiber optic Ethernet with hardware time synchronization mechanism to establish a deterministic low-latency data channel between the master control terminal and the slave robot.

[0009] As the core decision-making unit of the system, the insulation safety interlocking center adopts a multi-core heterogeneous processor architecture on its hardware platform and runs a real-time deterministic task scheduling kernel on the software level. The center receives and processes instruction data from the master control terminal and status feedback data from the slave operation robot in real time, and generates the final motion control command and insulation interlocking signal based on the built-in multi-level safety strategy.

[0010] Furthermore, the specific composition of the master control terminal is as follows. The high-precision six-dimensional force feedback device is built based on a strain gauge torque sensor and an optical position tracker. It continuously measures the force and torque vector applied by the operator, as well as the spatial displacement and rotation of the handle, at a sampling frequency of 1KHz.

[0011] The attitude sensing module integrates a three-axis MEMS gyroscope and an accelerometer. It fuses inertial data through a complementary filtering algorithm and outputs the real-time attitude quaternion of the master controller at a frequency of 200Hz.

[0012] The master control terminal also has an instruction preprocessing unit. This unit performs moving average filtering and kinematic forward solution calculation on the collected raw data, encapsulates the force feedback data, position data and attitude data into a master pose instruction stream with timestamps, and sends it to the insulation safety interlocking center through a high-speed communication link.

[0013] Furthermore, the specific composition of the end-user robot is as follows: The robotic arm has a 6-DOF serial configuration, with each joint driven by a brushless DC servo motor. The motor output shaft is connected to the robotic arm linkage through a two-stage reduction mechanism consisting of a harmonic reducer and a hollow RV reducer.

[0014] An absolute photoelectric encoder, with a resolution of 24 bits, is directly mounted on the rear end of the motor rotor and is used to provide feedback on the absolute rotation angle of the motor. Four miniature LiDAR probes are arranged in a ring on the end effector base. Each probe has a ranging range of 0.05m to 10m, a ranging accuracy of ±1mm, and a scanning frequency of 100Hz. The electric field strength sensor array consists of eight spherical electric field probes, uniformly embedded in the end effector housing. Its measurement range is 0.1kV / cm to 30kV / cm, and it is used to sense changes in the spatial electric field gradient when the robot's end effector is near a charged object.

[0015] Furthermore, the internal logic processing flow of the insulation safety interlocking center is as follows.

[0016] The center first receives the master-end pose command stream and slave-end status feedback data through a high-speed communication link. The status feedback data includes the absolute rotation angle of each joint, the real-time pose of the end effector, the lidar point cloud data, and the electric field strength sensor readings.

[0017] Subsequently, the dynamic inverse kinematics solution module in the center, based on the DH parameter model of the robotic arm, uses a numerical iterative method to calculate in real time the target angles of each joint required to achieve the master-end command pose.

[0018] Meanwhile, the safe distance calculation module processes the lidar point cloud data in real time, identifies the nearest obstacle point to the robotic arm using the kd-tree nearest neighbor search algorithm, and calculates the Euclidean distance between the robotic arm surface and the nearest obstacle by combining the current link model and envelope volume of the robotic arm.

[0019] The module also receives data from an electric field strength sensor. When the detected spatial electric field gradient is greater than the preset level 2 threshold, it automatically increases the conservative level of the safety distance calculation.

[0020] Furthermore, the insulation safety interlocking center is equipped with a multi-modal control strategy switching module. The inputs to this module include the joint target angle output by the dynamic inverse kinematics calculation module, the real-time minimum safe distance output by the safe distance calculation module, and preset safety thresholds at various levels.

[0021] The system has three preset safety thresholds, corresponding to normal operation, early warning intervention, and emergency braking, respectively.

[0022] When the real-time minimum safe distance is greater than the first-level threshold, the system operates in high-precision position control mode, and the joint controller uses a PID control algorithm based on feedforward and feedback to track the target angle.

[0023] When the real-time minimum safe distance is between the first and second level thresholds, the system automatically switches to impedance control mode. By introducing a virtual spring damping model into the position control loop, the robotic arm exhibits compliant characteristics when approaching obstacles.

[0024] When the real-time minimum safe distance is less than the second-level threshold or the electric field strength is greater than the highest alarm threshold, the multimodal control strategy switching module immediately sends the highest priority interrupt signal to all joint controllers, triggering dynamic braking of all joints of the robotic arm, and at the same time sends a tactile alarm signal to the master control terminal through the high-speed communication link.

[0025] Furthermore, the system also includes a delay compensation and trajectory smoothing subsystem, which operates as a software module of the insulation safety interlocking center.

[0026] This module first estimates the fixed delay of the current communication loop using the hardware timestamp of the high-speed communication link.

[0027] The module then buffers the received master pose command stream and uses a forward prediction mechanism based on the Kalman prediction algorithm to predict the master command state for the next control cycle.

[0028] For the trajectory fed back from the slave end, this module applies a 5th-order B-spline curve interpolation algorithm to replan the trajectory, in order to eliminate high-frequency noise in the trajectory caused by joint servo jitter and ensure that the movement of the slave robotic arm is smooth and continuous.

[0029] Furthermore, a two-way force feedback channel is established between the master control terminal and the insulation safety interlocking center.

[0030] Based on the results from the safety distance calculation module and the current operating mode of the system, the insulation safety interlock center generates corresponding force feedback signals. When the system is operating in impedance control mode, the center calculates the feedback force generated by the virtual spring damping model and maps it to the six-dimensional force feedback device of the master control terminal.

[0031] When the system triggers an early warning intervention, the force feedback device will generate a continuous low-frequency vibration signal.

[0032] When the system enters an emergency braking state, the force feedback device will simulate a huge reverse resistance to alert the operator.

[0033] This invention also provides a master-slave control and insulation interlocking protection method for a live-line working robot applied to the above-mentioned system. This method is executed by an insulation safety interlocking center and includes the following steps: Step 1: System Initialization. Establish communication connections between the master control terminal, the slave robot, and the insulation safety interlock center; perform sensor calibration and robotic arm zero-return operation; and load preset safety threshold parameters at various levels.

[0034] Step two, data synchronization acquisition and transmission. The master control terminal continuously acquires operation commands and sends master pose command streams, while the slave robot continuously acquires joint status and environmental perception data and sends status feedback data. All data is accompanied by high-precision hardware timestamps.

[0035] Step 3: Command Parsing and Motion Planning. The insulation safety interlocking center parses the master-end pose command flow, predicts commands using a communication delay compensation algorithm, and obtains the target trajectory in joint space through dynamic inverse kinematics calculation.

[0036] Step 4: Real-time safety situation assessment. The safety distance calculation module calculates the real-time minimum safe distance between the robotic arm and charged bodies or obstacles based on the latest lidar point cloud and electric field intensity data, and assesses the current electric field environment level.

[0037] Step 5: Multimodal Control Decision and Execution. Based on the evaluation results from Step 4, the multimodal control strategy switching module selects and activates the corresponding control mode, generating specific joint control commands. In normal mode, it accurately tracks the target trajectory; in early warning mode, impedance control is introduced to ensure compliant operation; in emergency mode, braking logic is immediately triggered.

[0038] Step Six: Force Feedback and Status Synchronization. Based on control decisions and safety status, generate corresponding force signals and send them to the main control terminal. Simultaneously, update system status information to the human-machine interface in real time.

[0039] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention systematically solves the problems of operation delay and trajectory jitter in master-slave control by integrating a high-speed communication link with hardware time synchronization mechanism, a delay compensation and trajectory smoothing subsystem located in the insulation safety interlock center, and high-precision control of the slave-end robotic arm joint servo system. The delay compensation algorithm offsets the effects of fixed and variable delays through predictive processing of the command stream; the trajectory smoothing algorithm filters out high-frequency noise during motion through advanced interpolation techniques, thereby ensuring that the slave-end robotic arm can still achieve stable, accurate, and real-time motion reproduction under complex high-pressure environments, improving the execution efficiency and success rate of complex tasks.

[0040] 2. This invention constructs a multi-layered, proactive insulation safety interlocking mechanism based on multi-sensor information fusion. This system does not solely rely on traditional distance detection, but combines precise lidar ranging with electric field intensity gradient sensing, enabling a more proactive assessment of potential discharge risks. By pre-setting multiple safety thresholds and tightly coupling them with a multi-modal control strategy, the system can seamlessly and intelligently switch between high-precision operation, compliant obstacle avoidance, and emergency braking modes based on real-time safety conditions. This design transforms safety protection from a passive, reactive response into an active, adjustable constraint throughout the entire operation process, fundamentally preventing major safety accidents caused by the accumulation of control errors leading to accidental collisions between the robotic arm and live conductors.

[0041] 3. This invention achieves a deep closed-loop integration of master-slave control and safety interlocking at the information and execution levels. The force feedback signal received by the operator through the master terminal directly reflects the safety status of the slave-end work site and the interaction force between the robot and the environment, enabling the operator to obtain an immersive remote operation experience and a clear perception of the safety status. The decision-making of the insulation safety interlocking center not only controls the actions of the slave robot but also synchronously guides the behavior of the master operator, forming a complete closed loop from perception, decision-making, control to feedback. This enhanced human-machine collaborative interaction mode reduces the cognitive load and probability of error for operators, further consolidating the safety and reliability of the entire live-line work process. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the overall technical architecture of the master-slave control and insulation interlocking protection system for live-line working robots proposed in this invention; Figure 2 This is a schematic diagram of the core principle framework of the insulation safety interlocking center in this invention; Figure 3 This is a logical flow diagram of the multimodal control strategy switching module in this invention; Figure 4 This is a schematic diagram of the multi-level interaction relationship and data flow between the master control terminal and the slave robot in this invention; Figure 5 This is a schematic diagram of the principle framework of the delay compensation and trajectory smoothing subsystem in this invention; Detailed Implementation

[0043] Please refer to the attached document. Figure 1 This embodiment details the specific technical implementation of a master-slave control and insulation interlocking protection system for a live-line working robot. The overall system architecture consists of four core parts: a master control terminal deployed in a zero-potential safety area, a slave working robot deployed at the high-voltage electric field work site, a high-speed communication link connecting the two ends, and an insulation safety interlocking center serving as the system decision-making hub.

[0044] The master control terminal is responsible for accurately capturing the operator's intentions and generating a flow of control commands. The slave robots are responsible for executing specific tasks under high-pressure environments.

[0045] High-speed communication links ensure reliable, low-latency transmission of commands and status data between the two ends.

[0046] The insulation safety interlocking center integrates and processes data from the entire system, performs motion planning, safety assessment and control decisions, and generates insulation interlocking protection signals.

[0047] The specific hardware configuration of the master control terminal is as follows. Its core input device is a high-precision six-dimensional force feedback device, which is based on a strain gauge torque sensor and an optical position tracker.

[0048] The torque sensor uses a Wheatstone bridge structure to directly measure the force and torque applied by the operator to the handle in three directions. Its analog signal is digitized by a 24-bit analog-to-digital converter, and the sampling frequency is fixed at 1KHz.

[0049] The optical position tracker uses a built-in infrared LED and an externally arranged infrared camera array to calculate the three-dimensional displacement and three-dimensional rotation of the handle in space in real time through the principle of triangulation. The output data format is a homogeneous transformation matrix containing the position vector and the rotation matrix.

[0050] The attitude sensing module is tightly integrated inside the handle and includes a set of three-axis MEMS gyroscopes and three-axis MEMS accelerometers.

[0051] The gyroscope measures angular velocity, and the accelerometer measures linear acceleration. The raw data from both are fused using a built-in complementary filtering algorithm.

[0052] The algorithm first performs low-pass filtering on the accelerometer data to suppress high-frequency vibration noise, then integrates the gyroscope data to obtain angle changes, and finally fuses the two through an adjustable gain coefficient, outputting a stable and drift-free attitude quaternion at a frequency of 200Hz.

[0053] The main control terminal also has an instruction preprocessing unit, which is implemented by an embedded microprocessor.

[0054] The unit receives raw data streams from the six-dimensional force feedback device and attitude sensing module. It first applies a moving average filter with a window size of 10 to the force, torque, and position data to eliminate high-frequency measurement noise.

[0055] The unit then performs forward kinematics calculations, encapsulating the filtered force vector, position vector, and attitude quaternion into a complete data frame according to a predefined communication protocol.

[0056] The data frame contains a frame header, a data payload, and a cyclic redundancy check code. The data payload contains all sensor data and a high-precision timestamp generated by a hardware clock. This data frame constitutes the master pose command stream.

[0057] After encapsulation, the instruction preprocessing unit injects the master pose instruction stream into the high-speed communication link through its Ethernet physical interface and sends it to the insulation safety interlocking center.

[0058] The high-speed communication link uses bidirectional fiber optic Ethernet with hardware time synchronization mechanism.

[0059] Its physical medium is single-mode optical fiber, and its transmission protocol is Gigabit Ethernet based on the IEEE 802.3 standard.

[0060] Both communication ends, namely the network interface cards of the master control terminal and the insulation safety interlocking center, support the IEEE 1588 precision time protocol.

[0061] During the system initialization phase, the precision time protocol master clock is located at the insulation safety interlocking center. It periodically sends synchronization messages and follow messages to the master control terminal. The master control terminal calibrates its local clock according to the time information in the messages, so that all data timestamps of the entire system are unified on a global time reference.

[0062] This hardware-level time synchronization mechanism controls network transmission latency jitter to the microsecond level, thereby establishing a deterministic low-latency data channel between the master control terminal and the insulation safety interlocking center.

[0063] All data packets transmitted through this link carry precise send and receive timestamps, providing crucial input for subsequent delay compensation algorithms.

[0064] The mechanical body of the end-effector robot is a 6-DOF serial robotic arm.

[0065] The drive system for each joint consists of a brushless DC servo motor, a harmonic reducer, a hollow rotary vector reducer, and an absolute photoelectric encoder.

[0066] The brushless DC servo motor serves as the power source, with its stator windings connected in a star configuration and its rotor being a permanent magnet.

[0067] The motor output shaft is connected to the wave generator of the harmonic reducer via an interference fit.

[0068] Harmonic reducers achieve first-stage speed reduction due to their high transmission ratio and zero backlash characteristics.

[0069] The output end of the harmonic reducer is connected to the input shaft of the hollow rotary vector reducer via a flange. The hollow rotary vector reducer provides a second stage with a higher reduction ratio and greater output torque, and its output flange directly drives the connecting rod of the robotic arm.

[0070] This two-stage reduction mechanism allows the robotic arm to achieve large output torque while maintaining a compact structure and high motion accuracy.

[0071] The absolute photoelectric encoder is directly mounted on the rear shaft of the servo motor rotor, and its code disk is a 24-bit absolute grating.

[0072] The encoder uses a photodiode array to read the code disk position. Even if the system is powered off and then powered on again, it can immediately provide feedback on the absolute mechanical angle of the motor without the need for zeroing operation.

[0073] Encoder data is transmitted to the joint controller in real time via a serial communication interface, such as a bidirectional serial synchronous interface.

[0074] The end effector of a robot is a key component for its interaction with the environment.

[0075] Its base is made of high-strength engineering plastic, and four miniature lidar probes are evenly distributed in a ring around the circumference of the base.

[0076] Each miniature lidar probe operates based on the time-of-flight ranging principle, and its core consists of a vertical cavity surface-emitting laser and a single-photon avalanche diode detector.

[0077] A vertical-cavity surface-emitting laser emits modulated laser pulses with a wavelength of 905 nm, and a single-photon avalanche diode detector receives the light pulses reflected back from the target object.

[0078] The internal timing circuit precisely measures the round-trip time of the laser pulse and calculates the distance based on the speed of light.

[0079] Each probe has an effective ranging range of 0.05m to 10m, with a ranging accuracy of ±1mm across the entire range.

[0080] Four probes scan synchronously at a frequency of 100Hz, jointly generating a three-dimensional point cloud data field around the end effector.

[0081] In addition, eight spherical electric field probes are uniformly embedded inside the end effector housing, forming an electric field strength sensor array.

[0082] Each spherical electric field probe consists of two symmetrical hemispherical electrodes separated by a high-impedance material.

[0083] When placed in an electric field, the probe induces a displacement current that is proportional to the field strength.

[0084] The current signal is converted into a digital electric field strength value by a high input impedance preamplifier and a 16-bit analog-to-digital converter.

[0085] The array has a measurement range of 0.1 kV / cm to 30 kV / cm, and can accurately sense subtle changes in the electric field gradient of the surrounding space when the robot's end effector approaches a charged body.

[0086] All end-sensor data, including lidar point cloud and electric field strength readings, are aggregated and timestamped by a data acquisition unit inside the end effector. The data is then uploaded to the robot's main controller via a fieldbus inside the robotic arm, such as Ethernet power supply. Subsequently, it is sent to the insulation and safety interlocking center via a high-speed communication link as part of the slave-end status feedback data.

[0087] Please refer to the attached document. Figure 2 The insulation safety interlocking center is the core of the intelligent decision-making of the entire system.

[0088] Its hardware platform adopts a multi-core heterogeneous processor architecture, which typically includes multiple high-performance ARM architecture processor cores and a field-programmable gate array.

[0089] At the software level, a real-time deterministic task scheduling kernel runs, which ensures that critical tasks, such as safe distance calculation and control decisions, can be completed within strict deadlines.

[0090] After the insulation safety interlocking center is activated, it first receives the master end pose command stream from the master control terminal and the status feedback data from the slave end operation robot synchronously through the high-speed communication link.

[0091] The status feedback data is comprehensive, including 24-bit absolute rotation angles of all 6 joints, real-time pose of the end effector calculated by the forward kinematics of the robotic arm, raw point cloud data generated by 4 miniature lidars, and instantaneous readings from 8 electric field strength sensors.

[0092] The primary processing module within the center is the dynamic inverse kinematics solution module.

[0093] This module is built based on the Danavitt Hartenberg parametric model of the slave robotic arm.

[0094] The Danawirt-Hartenberg parametric model defines four parameters for each link: link length, link torsion angle, link offset, and joint angle. The module receives the desired end effector pose parsed from the master end effector pose command stream.

[0095] Since the analytical solution to the inverse kinematics of a 6-DOF serial manipulator may not exist or may have multiple solutions, this module uses a numerical iterative method, specifically the Newton-Raphson method, to solve the problem in real time.

[0096] The solution process uses the current joint angles of the robotic arm as the initial values ​​for iteration, calculates the error between the actual pose of the current end effector and the desired pose, and then obtains the incremental adjustment of the joint angles by calculating the pseudo-inverse of the Jacobian matrix. The process iterates until the pose error is less than a preset threshold, such as a position error of 0.001m and an attitude error of 0.01rad. Finally, it outputs the six target joint angles required to achieve the desired pose.

[0097] The safety distance calculation module is executed in parallel with the inverse kinematics solution.

[0098] This module is responsible for real-time assessment of the safety status of the robotic arm and its surrounding environment, especially charged objects.

[0099] The module first processes the lidar point cloud data.

[0100] The raw point cloud data it receives from the robot's main controller typically contains tens of thousands of three-dimensional point coordinates.

[0101] For efficient processing, the module first performs voxel mesh downsampling on the point cloud, reducing the point cloud density to a manageable level.

[0102] Subsequently, a three-dimensional kd-tree data structure is constructed to spatially index the downsampled point cloud.

[0103] For each link of the robotic arm, the module simplifies it into a cylindrical envelope model.

[0104] For each link model surface with several sampling points, the module uses a kd-tree to perform a nearest neighbor search, quickly finding the point in the point cloud closest to that sampling point and calculating the Euclidean distance between them. After traversing all sampling points of all links, the minimum value is taken as the real-time minimum safe distance at the current moment.

[0105] Meanwhile, the safe distance calculation module continuously monitors the readings of the electric field strength sensor array. The system has preset multiple levels of electric field strength thresholds.

[0106] When the module detects that any electric field probe reading is greater than the preset Level 2 threshold, such as 15kV / cm, it will automatically trigger an upgrade of the safety policy.

[0107] Specifically, after calculating the real-time minimum safe distance, the value is not used directly, but multiplied by a safety factor less than 1, such as 0.8, to obtain a more conservative effective safe distance for decision-making, thus providing early warning of potential discharge risks.

[0108] The most critical decision-making unit within the insulation safety interlocking center is the multi-modal control strategy switching module.

[0109] Please refer to the attached document. Figure 3The input sources for this module include three aspects: the target angles of the six joints output by the dynamic inverse kinematics solution module, the real-time minimum safe distance output by the safe distance calculation module, and the system's preset safety thresholds at various levels. The system explicitly presets three levels of safety thresholds. The first level threshold corresponds to the safety boundary of normal operation, the second level threshold corresponds to the critical point for early warning intervention, and the third level threshold corresponds to the final line of defense for emergency braking.

[0110] The core logic of the multimodal control strategy switching module is a finite state machine.

[0111] When the real-time minimum safe distance is greater than the first-level threshold, the system is in high-precision position control mode.

[0112] In this mode, the module directly sends the target angle of the joint to the servo controller of each joint.

[0113] The joint controller adopts a proportional-integral-derivative control algorithm based on a combination of feedforward and feedback.

[0114] The feedforward term is calculated based on the acceleration of the target trajectory and the system dynamics model to counteract the effects of nonlinear dynamics; the feedback term performs proportional, integral, and differential operations based on the deviation between the actual angle fed back by the absolute photoelectric encoder and the target angle, and its output controls the current of the servo motor, thereby achieving high-precision trajectory tracking.

[0115] When the real-time minimum safe distance drops to between the first and second level thresholds, the multi-modal control strategy switching module immediately switches the system state to impedance control mode.

[0116] In this mode, the control objective changes from pure position tracking to coordinated control of position and force.

[0117] The module achieves its impedance characteristics by introducing a virtual spring-damped model into the position control loop.

[0118] This model equates the interaction between the robotic arm's end effector and the environment to a parallel system of springs and dampers.

[0119] Its dynamic behavior can be described by the following formula: ; Represents virtual interactivity. Represents the virtual stiffness matrix. Represents the virtual damping matrix. Indicates the desired pose. Indicates the actual pose. Indicates the desired speed. Indicates the actual speed.

[0120] When the robotic arm approaches an obstacle, even without physical contact, the virtual model generates a virtual force pointing in the opposite direction to the obstacle.

[0121] The virtual force is mapped to the joint space through the Jacobian matrix of the robotic arm, and converted into an additional joint torque command.

[0122] This torque command, combined with the existing position control command, causes the robotic arm to automatically decelerate and exhibit compliant retraction behavior when approaching obstacles, thus avoiding rigid collisions.

[0123] When the real-time minimum safe distance further decreases to below the Level 2 threshold, or when the reading of any electric field strength sensor exceeds the Level 3 threshold of the highest alarm, the multi-modal control strategy switching module immediately enters the emergency braking state.

[0124] At this point, the module sends an interrupt signal with the highest priority to the servo controllers of all six joints.

[0125] The interrupt signal triggers the preset dynamic braking logic inside the servo driver.

[0126] For brushless DC servo motors, dynamic braking is typically achieved by short-circuiting the three-phase windings of the motor. The back electromotive force generated by the motor's rotation creates a braking torque, causing the robotic arm to stop rapidly. The braking process is completed within milliseconds.

[0127] At the same time, the module sends a specific tactile alarm signal to the main control terminal via a high-speed communication link.

[0128] Please refer to the attached document. Figure 5 To further improve the system's operational performance and safety, a delay compensation and trajectory smoothing subsystem is integrated into the system. This subsystem operates as an independent software module of the insulation safety interlocking center.

[0129] The primary task of this module is to accurately estimate communication latency. It utilizes the hardware timestamps provided by the high-speed communication link. For each received data packet, it compares its sending timestamp with its receiving timestamp, and the difference between the two is the one-way latency of that communication.

[0130] By performing statistical filtering on the delay of multiple consecutive data packets, such as taking the median, the module can estimate the fixed delay and jitter range of the current communication loop relatively accurately.

[0131] Based on this estimated delay, the delay compensation module processes the received master pose command stream.

[0132] It maintains an instruction buffer that stores historical instruction data for a recent period. The module then employs a forward prediction mechanism based on the Kalman prediction algorithm.

[0133] The Kalman prediction algorithm is an optimal recursive state estimation algorithm.

[0134] It models the motion state of the main hand as a state vector containing position, velocity, and acceleration, and establishes the system's state equation and observation equation.

[0135] The algorithm predicts the master instruction state for the next control cycle based on historical instruction data, thereby offsetting the operational lag caused by communication delays and enabling the movement of the slave robotic arm to be synchronized with the operator's current intention.

[0136] In terms of slave-end trajectory processing, the trajectory smoothing subsystem mainly targets the high-frequency components of the joint trajectory fed back from the slave end, which may be caused by servo system noise or control jitter.

[0137] The system uses a 5th-order B-spline curve interpolation algorithm to reprogram the original joint angle sequence.

[0138] The 5th-order B-spline curve has the characteristic of continuous second derivative, which can ensure that the velocity and acceleration curves of the generated trajectory are smooth.

[0139] The algorithm uses timestamps as parameters and the original joint angles as shape points to determine the control vertices of the B-spline curve by solving a system of linear equations.

[0140] The redesigned trajectory not only eliminates high-frequency noise but also ensures continuous and smooth motion, improving the stability and accuracy of the slave robotic arm during precision operations.

[0141] Please refer to the attached document. Figure 4 The system also constructs a two-way force feedback channel to achieve deep information fusion and interaction between the master and slave ends.

[0142] Based on the results of the safety distance calculation module and the current operating mode of the system, the insulation safety interlocking center generates corresponding force feedback signals in real time and sends them to the master control terminal via a high-speed communication link.

[0143] When the system is operating in impedance control mode, the center calculates the feedback force F generated by the aforementioned virtual spring damping model.

[0144] The force vector is transformed from the task space of the slave robot to the coordinate system of the master control terminal handle through a force mapping algorithm, and drives the motor of its six-dimensional force feedback device, so that the operator's hand can feel a real force that prevents him from moving in a dangerous direction.

[0145] When the system is in a warning intervention state, that is, when the safe distance is between the first and second level thresholds, the center will generate a continuous low-frequency vibration signal, such as a 5Hz square wave. This signal controls the force feedback device to produce a slight but continuous tactile reminder.

[0146] When the system enters emergency braking mode, the center generates a force signal simulating huge reverse resistance. This signal is usually a force vector with saturated amplitude, and its direction is opposite to the operator's current operating direction, making the operator feel a strong sense of obstruction and thus immediately realize that a dangerous situation has occurred.

[0147] This two-way force feedback mechanism directly and intuitively transmits the safety status from the remote end to the operator, forming a complete human-machine collaborative closed loop from perception, decision-making, control to feedback.

[0148] The system described in this embodiment achieves high precision, low latency, and high reliability of master-slave control through the detailed hardware configuration and software algorithms described above. It also constructs a multi-level, proactive insulation safety interlocking protection mechanism to ensure that the live-line working robot can safely and efficiently complete various tasks in complex high-voltage environments.

[0149] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0150] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A master-slave control and insulation interlocking protection system for a live-line working robot, characterized in that, include: The master control terminal is deployed in a zero-potential safe area and has a built-in high-precision six-dimensional force feedback device and attitude sensing module to collect the operator's operation intentions in real time and generate master position and posture command streams. The end-effector robot is deployed at the high-voltage electric field work site. Its robotic arm joints are equipped with multi-stage reduction mechanisms and absolute photoelectric encoders, and the end effector integrates a miniature lidar and an electric field strength sensor array. The high-speed communication link adopts a bidirectional fiber optic Ethernet with hardware time synchronization mechanism to establish a deterministic low-latency data channel between the master control terminal and the slave operation robot. The insulation safety interlocking center, as the core decision-making unit of the system, adopts a multi-core heterogeneous processor architecture on its hardware platform and runs a real-time deterministic task scheduling kernel on the software level. The insulation safety interlocking center receives and processes instruction data from the master control terminal and status feedback data from the slave robot in real time, and generates the final motion control commands and insulation interlocking signals based on the built-in multi-level safety strategy. The insulation safety interlocking center includes a dynamic inverse kinematics calculation module and a safety distance calculation module; The dynamic inverse kinematics solution module is based on the DH parameter model of the robotic arm and uses a numerical iterative method to calculate in real time the target angles of each joint required to achieve the master-end command pose. The safe distance calculation module processes the lidar point cloud data in real time, identifies the nearest obstacle point to the robotic arm through the kd-tree nearest neighbor search algorithm, and calculates the Euclidean distance between the surface of the robotic arm and the nearest obstacle by combining the current link model and envelope volume of the robotic arm. The safe distance calculation module also receives data from the electric field strength sensor. When the detected spatial electric field gradient is greater than the preset second-level threshold, it automatically increases the conservative level of the safe distance calculation. The insulation safety interlocking center also includes a multi-mode control strategy switching module; The inputs to the multimodal control strategy switching module include the joint target angle output by the dynamic inverse kinematics solution module, the real-time minimum safe distance output by the safe distance calculation module, and preset safety thresholds at various levels. The system has three preset safety thresholds, corresponding to normal operation, early warning intervention, and emergency braking, respectively. When the real-time minimum safe distance is greater than the first-level threshold, the system operates in high-precision position control mode, and the joint controller uses a PID control algorithm based on feedforward and feedback to track the target angle. When the real-time minimum safe distance is between the first and second level thresholds, the system automatically switches to impedance control mode. By introducing a virtual spring damping model into the position control loop, the robotic arm exhibits compliant characteristics when approaching obstacles. When the real-time minimum safe distance is less than the second-level threshold or the electric field strength is greater than the highest alarm threshold, the multi-modal control strategy switching module immediately sends the highest priority interrupt signal to all joint controllers, triggering the full-joint dynamic braking of the robotic arm.

2. The master-slave control and insulation interlocking protection system for a live-line working robot according to claim 1, characterized in that, The high-precision six-dimensional force feedback device is built on strain gauge torque sensor and optical position tracker, and continuously measures the force and torque vector applied by the operator and the spatial displacement and rotation of the handle at a fixed sampling frequency. The attitude sensing module integrates a three-axis MEMS gyroscope and an accelerometer, and fuses inertial data through a complementary filtering algorithm to output the real-time attitude quaternion of the main hand controller at a fixed frequency. The main hand control terminal is also equipped with an instruction preprocessing unit, which performs moving average filtering and kinematic forward solution calculation on the collected raw data, and encapsulates the force feedback data, position data and attitude data into a main end pose instruction stream with timestamps; The robotic arm is a 6-DOF serial configuration, with each joint driven by a brushless DC servo motor. The motor output shaft is connected to the robotic arm link through a two-stage reduction mechanism consisting of a harmonic reducer and a hollow RV reducer. The absolute photoelectric encoder is directly mounted on the rear end of the motor rotor to provide feedback on the absolute rotation angle of the motor; multiple miniature lidar probes are arranged in a ring on the end effector base. The electric field strength sensor array consists of multiple spherical electric field probes, which are uniformly embedded in the end effector housing.

3. The master-slave control and insulation interlocking protection system for a live-line working robot according to claim 1, characterized in that, The dynamic behavior of the virtual spring-damped model in the impedance control mode is described by the following formula: the interaction force is equal to the virtual stiffness matrix multiplied by the difference between the desired pose and the actual pose, plus the virtual damping matrix multiplied by the difference between the desired velocity and the actual velocity; the virtual stiffness matrix and the virtual damping matrix are preset parameter matrices.

4. The master-slave control and insulation interlocking protection system for a live-line working robot according to claim 1, characterized in that, It also includes a delay compensation and trajectory smoothing subsystem; The delay compensation and trajectory smoothing subsystem operates as a software module of the insulation safety interlocking center; The module first estimates the fixed delay of the current communication loop using the hardware timestamp of the high-speed communication link; then it buffers the received master pose command stream and uses a forward prediction mechanism based on the Kalman prediction algorithm to predict the master command state for the next control cycle. For the trajectory fed back from the end, this module applies a 5th-order B-spline curve interpolation algorithm to replan the trajectory in order to eliminate high-frequency noise in the trajectory caused by joint servo jitter. The Kalman prediction algorithm models the motion state of the master hand as a state vector containing position, velocity, and acceleration, and establishes the state equation and observation equation of the system; the algorithm predicts the master hand command state for the next control cycle based on historical command data.

5. The master-slave control and insulation interlocking protection system for a live-line working robot according to claim 1, characterized in that, A bidirectional force feedback channel is also established between the master control terminal and the insulation safety interlocking center; Based on the results of the safety distance calculation module and the current operating mode of the system, the insulation safety interlock center generates corresponding force feedback signals; When the system is running in impedance control mode, the center calculates the feedback force generated by the virtual spring damping model and maps it to the six-dimensional force feedback device of the master control terminal. When the system triggers an early warning intervention, the force feedback device generates a continuous low-frequency vibration signal; When the system enters an emergency braking state, the force feedback device simulates a huge reverse resistance.

6. The master-slave control and insulation interlocking protection system for a live-line working robot according to claim 2, characterized in that, The instruction preprocessing unit applies a moving average filter to the force, torque, and position data to eliminate high-frequency measurement noise; the forward kinematics calculation encapsulates the filtered force vector, position vector, and attitude quaternion into a complete data frame according to a predefined communication protocol.

7. The master-slave control and insulation interlocking protection system for a live-line working robot according to claim 1, characterized in that, The safe distance calculation module performs voxel grid downsampling processing on the lidar point cloud data to reduce the point cloud density to a manageable level; Subsequently, a three-dimensional kd-tree data structure was constructed to spatially index the downsampled point cloud; For each link of the robotic arm, it is simplified into a cylindrical envelope model. For several sampling points on the surface of each link model, a nearest neighbor search is performed using a kd-tree.

8. A master-slave control and insulation interlocking protection method for a live-line working robot, characterized in that, The master-slave control and insulation interlocking protection system for the live-line working robot as described in any one of claims 1-7 is used to implement master-slave control and insulation interlocking protection, which is executed by the insulation safety interlocking center, and includes the following steps: The system initializes, establishes communication connections between the master control terminal, the slave robot, and the insulation safety interlock center, performs sensor calibration and robotic arm zero-return operation, and loads preset safety threshold parameters at all levels. Data is collected and transmitted synchronously. The master control terminal continuously collects operation commands and sends master pose command streams, while the slave robot continuously collects joint status and environmental perception data and sends status feedback data. All data is accompanied by high-precision hardware timestamps. Command parsing and motion planning: The insulation safety interlocking center parses the master end pose command flow, combines the communication delay compensation algorithm to predict the command, and obtains the target trajectory in the joint space through dynamic inverse kinematics calculation; Real-time safety situation assessment: The safety distance calculation module calculates the real-time minimum safety distance between the robotic arm and charged bodies or obstacles based on the latest lidar point cloud and electric field intensity data, and assesses the current electric field environment level. Multimodal control decision-making and execution: The multimodal control strategy switching module selects and activates the corresponding control mode based on the evaluation results and generates specific joint control commands. Force feedback and status synchronization: Based on control decisions and safety status, corresponding force signals are generated and sent to the master control terminal, while system status information is updated to the human-machine interface in real time.

Images