Air-ground twinborn cooperative mechanical arm power transmission line foreign matter picking system and method

Abstract

The invention discloses an air-ground twinborn cooperative mechanical arm power transmission line foreign matter picking system and method, and the system comprises an air-based intelligent operation terminal which comprises a large-load eight-rotor unmanned aerial vehicle platform, an unmanned aerial vehicle power and navigation module is arranged in the air-based intelligent operation terminal, and a mechanical arm is connected with the bottom of an unmanned aerial vehicle through a rigid connecting rod; a camera module, a dexterous hand pressure sensor and an insulating flexible dexterous hand are integrated at the tail end; the ground control and somatosensory interaction platform comprises a data glove, a PC terminal, a display screen and a battery; a forearm posture inclination angle detection module, a hand back posture angle sensor, a finger bending degree sensor and a force feedback module are arranged in the data glove; the ground control and somatosensory interaction platform interacts with the air-based intelligent operation terminal in real time through wireless unvarnished transmission. Hand actions drive the mechanical arm to accurately pick foreign matters after being coded. Risks of manual hot-line work are eliminated, zero damage of the power transmission line in the foreign matter stripping process is ensured, and finally an operator remotely controls the mechanical arm to accurately pick the foreign matter.

Description

Technical Field

[0001] This invention pertains to foreign object removal technology for power transmission lines, and particularly relates to a system and method for removing foreign objects from power transmission lines using a hybrid robotic arm with a ground-to-air twin mechanism. Background Technology

[0002] With the continuous expansion of the power grid, overhead transmission lines have become the main channel for power transmission. However, due to the long-term exposure of these lines to the outdoor environment, foreign objects such as kites, plastic films, and advertising banners are easily entangled in the conductors by the wind, creating safety hazards. These foreign objects can not only cause line discharges and short circuits, but also easily lead to a decline in insulation performance under rain, snow, and humid conditions, even inducing regional power outages. In addition, the removal of foreign objects from transmission lines in mountainous areas, river bends, and other complex terrains presents even greater challenges due to the difficulty in quickly reaching these areas.

[0003] Current conventional methods for removing foreign objects from power transmission lines have significant limitations. Manual live-line removal requires personnel to climb towers or use aerial work platforms to approach the conductors and manually remove the foreign object using insulated operating rods. This method is not only time-consuming and labor-intensive but also carries safety risks such as falls from heights and electric shocks, especially in complex terrain areas such as those crossing rivers or mountains, where personnel and equipment cannot quickly reach the work site. On the other hand, high-temperature fusing technology uses lasers or flame-emitting devices to non-contactly fuse foreign objects at high temperatures. While this avoids direct contact with live lines, the high temperatures during operation can damage the conductor insulation layer, and its effectiveness in removing non-flammable or high-melting-point foreign objects such as nylon ropes and composite fibers is limited. Furthermore, high-temperature operations pose a potential fire risk during dry seasons, further limiting its application scenarios.

[0004] Currently, manual live-line work requires workers to directly contact high-voltage wires, posing significant safety risks such as falls from heights and electric shocks. Furthermore, in complex terrain areas, the inability of equipment to reach these areas leads to low work efficiency. While high-temperature fusing technologies such as lasers or flamethrowers can be operated remotely, they are prone to damaging the wire insulation layer and lack the ability to remove non-flammable or high-melting-point foreign objects, posing a fire hazard. Existing drone-mounted mechanical structures or robotic arms lack dynamic adjustment capabilities and tactile feedback mechanisms, making them prone to jamming due to foreign object entanglement or wire swinging. Operation relies on remote control via a handle, demanding high skill levels from personnel, and uncontrolled clamping force can easily lead to broken wire strands or foreign object residue. Summary of the Invention

[0005] To address the above problems, this invention provides a system and method for removing foreign objects from power transmission lines using a hybrid aerial-ground twin robotic arm.

[0006] The present invention relates to an air-ground twin collaborative robotic arm system for removing foreign objects from power transmission lines, comprising an air-based intelligent operation terminal and a ground control and motion-sensing interaction platform.

[0007] The airborne intelligent operation terminal includes a heavy-duty octocopter UAV platform, a robotic arm, and a battery. The heavy-duty octocopter UAV platform is equipped with a UAV power and navigation module, which integrates a high-precision GPS positioning unit, a dual-antenna direction finder, and redundant IMU sensors. The robotic arm is connected to the bottom of the UAV via a rigid link. The robotic arm controller controls the motors of each joint through a motor drive unit to achieve the position adjustment and dynamic impedance matching of the end effector in three-dimensional space. A camera module is installed at the front end of the robotic arm to acquire high-definition images of the wire entanglement state in real time, and transmits the image data, robotic arm position information, and clamping force data synchronously to the ground control and motion-sensing interaction platform through a wireless transmission unit.

[0008] The ground control and motion-sensing interaction platform includes a single-handed wearable data glove, a PC, a display screen, and a battery 1. The data glove is designed based on the human right hand and incorporates a forearm tilt angle detection module, a back-of-hand posture angle sensor, a finger flexion sensor, and a force feedback module. The forearm tilt angle detection module is fixed near the wrist of the operator's arm and has a built-in six-axis inertial measurement unit (IMU) to detect the tilt angle of the forearm relative to the horizontal plane in real time, providing an attitude reference for the vertical lifting and lowering motion of the robotic arm. The back-of-hand posture angle sensor is integrated into the back of the glove and uses a fusion algorithm of a three-axis gyroscope and accelerometer to measure the pitch, yaw, and roll angles of the back of the hand, driving the three-dimensional spatial posture adjustment of the robotic arm's end effector. The finger flexion sensor is also included. At the finger joints, flexible resistance strain gauges detect the bending angle of each finger, mapping it to the opening and closing control commands of the robotic arm's dexterous hand. The linear resonant motor array of the force feedback module is distributed in the palm and fingertip areas of the glove, dynamically adjusting the vibration mode based on the data returned by the dexterous hand pressure sensor. The finger limiting mechanism is located at the index finger joint, limiting excessive finger bending through a damping pivot. When the bending angle exceeds a preset safety threshold, a mechanical lock is triggered to prevent overload of the dexterous hand closure. The PC receives and parses the posture data, mapping the hand movements to the six-degree-of-freedom joint angle commands of the robotic arm and the opening and closing control signals of the dexterous hand through a preset algorithm. The display screen shows the wire and foreign object images and dynamic values ​​of the clamping force returned by the camera module in real time.

[0009] The ground control and somatosensory interaction platform interacts with the airborne intelligent operation terminal in real time via wireless transmission: hand movements are encoded to drive the robotic arm to precisely pick up foreign objects, and the dexterous hand pressure and environmental data are transmitted back to the ground, triggering the tactile response of the force feedback module and updating the interface status; if a communication interruption or abnormal posture is detected, the system immediately freezes the robotic arm's movements and activates the autonomous protection protocol to ensure that the entire live-line operation is safe and controllable.

[0010] Furthermore, the robotic arm end effector integrates a dexterous hand pressure sensor and an insulated flexible dexterous hand, with the inside of the dexterous hand covered with an anti-slip silicone layer to prevent foreign objects from slipping off.

[0011] Furthermore, the linear resonant motor array dynamically adjusts the vibration mode as follows: when in contact with a foreign object, the fingertip motor uses low-frequency micro-vibration to simulate the tactile sensation; when the clamping force approaches the safety threshold of the wire, the palm motor switches to high-frequency strong vibration, simultaneously triggering the warning indicator on the display screen.

[0012] The present invention provides a method for removing foreign objects from power transmission lines using an air-to-ground twin collaborative robotic arm, employing the aforementioned air-to-ground twin collaborative robotic arm system for removing foreign objects from power transmission lines, specifically including the following steps:

[0013] Step 1: The drone approaches and hovers.

[0014] Operators send takeoff commands to the heavy-duty octocopter UAV via the LoRa communication module to achieve centimeter-level hovering positioning. The UAV flies smoothly to the top of the target tower. When the horizontal distance between the UAV and the work point is reduced to 0.5 meters, the lidar ranging module triggers the automatic hovering program. The PID control algorithm stabilizes the body attitude, providing a stable platform for the robotic arm to operate.

[0015] Step 2: Image transmission of the work environment.

[0016] The binocular camera module installed at the front end of the six-degree-of-freedom robotic arm is activated. The image data is compressed by H.265 encoding and transmitted to the ground PC through the wireless transmission unit. The display screen simultaneously displays two images: the main image is a 1080P@30fps high-definition video stream, and the auxiliary image is a projection of the robotic arm's kinematic model.

[0017] Step 3: Acquisition and fusion processing of motion-sensing interaction data.

[0018] Operator 2 puts on a single right-hand data glove and starts the motion-sensing interaction system for initial calibration.

[0019] Finger motion acquisition: Flexible resistance strain gauges at the finger joints monitor the bending angle of each finger joint in real time: 0-90°. The bending angle is mapped to the dexterity hand opening and closing amplitude control command through normalization processing.

[0020] Hand posture analysis: The MPU6050 chip in the back of the hand posture acquisition module acquires raw data from the three-axis accelerometer and gyroscope at a sampling rate of 200Hz. After fusion using the Kalman filter algorithm, the back of the hand pitch angle, roll angle and yaw angle are output.

[0021] Forearm posture baseline establishment: The forearm posture acquisition unit uses a six-axis IMU module to continuously detect the tilt angle between the forearm and the horizontal plane.

[0022] Step 4: Kinematics calculation and instruction generation.

[0023] The ground PC receives the three-dimensional attitude data stream of the data glove, constructs the kinematic model of the robotic arm based on the improved DH parameter method, and converts the hand movements into a six-degree-of-freedom joint angle command sequence through inverse kinematics solution. During the solution process, the Jacobian matrix iteration method is used to optimize the trajectory smoothness and avoid the singular configuration of the robotic arm. The calculated control commands are encrypted and sent to the robotic arm controller of the air-based terminal through the wireless transparent transmission unit.

[0024] Step 5: Precise control and dynamic impedance adjustment of the robotic arm.

[0025] After parsing the instructions, the robotic arm controller drives the brushless servo motors to perform multi-axis linkage control:

[0026] The boom slewing joint and elbow pitch joint adopt position-velocity dual closed-loop PID control to achieve coarse positioning of the end effector.

[0027] The wrist yaw and pitch joints dynamically adjust their rotational inertia based on a fuzzy adaptive algorithm to compensate for drone hovering jitter.

[0028] The dexterous hand opening and closing mechanism receives finger bending degree mapping commands and drives the flexible dexterous hand to adjust the clamping force with a step accuracy of 0.1N through a harmonic reducer, while simultaneously triggering the contact force feedback of the pressure sensor.

[0029] Step 6: Tactile feedback and safety interlock mechanism.

[0030] The dexterous hand pressure sensor collects gripping force data at a frequency of 60Hz. After normalization processing by the robotic arm controller, the data is transmitted back to the ground platform wirelessly.

[0031] Tactile simulation: When the contact force is within a safe range, the linear resonant motor at the fingertips of the data glove will gradually increase the vibration amplitude and energy density as the applied force increases, simulating the sensation of a foreign object; while the vibration module in the palm area adopts a mode that combines short-term high-frequency impact with intermittent vibration to form a progressive tactile cues from weak to strong.

[0032] Over-limit warning: When the contact force reaches 90% of the wire safety threshold, the palm motor switches to high-frequency strong vibration mode, and the display screen simultaneously displays a red warning box and a buzzer alarm.

[0033] Emergency Lockout: If the pressure exceeds the safety threshold of the wire or a stall is detected in the joint motor, the main control chip immediately triggers the electromagnetic lock of the finger limit mechanism, forcing the dexterous hand to maintain its current state.

[0034] Step 7: Safe evacuation and verification of operational effectiveness.

[0035] After the foreign object is removed, the operator sends an evacuation command via remote control, and the drone evacuates to an observation point 10 meters away along a preset safe path; the camera module starts a second scan to verify that there is no residual damage on the surface of the wire; the ground platform generates an operation report, recording the peak clamping force, operation time, and comparison images before and after the foreign object removal.

[0036] Furthermore, constructing the kinematic model of the robotic arm includes the following steps:

[0037] S1: Obtain the mapping relationship between the data glove and the robotic arm, measure the range of values ​​of the end position of the data glove in the XYZ directions, and the range of values ​​of the end position of the robotic arm in the XYZ directions, so as to establish the mapping relationship between the two.

[0038] S2: Based on the DH parameter method, a forward kinematics model of the robotic arm is established, and the transformation matrix between the base coordinate system and the end-joint coordinate system is obtained by matrix transformation formula:

[0039]

[0040] The T matrix between two joints is:

[0041]

[0042] in, Indicates the length of the rod. For the twist angle, It is the horizontal spacing. Joint angle, This represents the transformation matrix from coordinate system j to coordinate system i.

[0043] S3: Based on the algebraic method and the actual operating range of the robotic arm, inverse kinematics is solved to obtain the transformation relationship between the body coordinate system and the end effector of the robotic arm, thereby establishing the correspondence between the end position of the robotic arm and the ideal servo motor rotation angle.

[0044] The advantages of the right-hand technology of this invention are as follows:

[0045] This invention utilizes a drone equipped with a high-degree-of-freedom robotic arm to approach the work site. Combined with motion-sensing interaction and force feedback technology, it eliminates the risks of manual live-line work and avoids secondary damage to power lines caused by high-temperature processes. Wearable devices are mapped in real-time to the robotic arm, enabling intuitive and simplified six-degree-of-freedom control. A dynamic contact force adjustment mechanism ensures zero damage to power lines during foreign object removal. Ultimately, this allows operators to remotely control the robotic arm to precisely remove foreign objects. Attached Figure Description

[0046] Figure 1 This is a structural block diagram of the air-ground twin collaborative robotic arm system for removing foreign objects from power transmission lines according to the present invention.

[0047] In the diagram: 1-Human arm; 2-Data glove; 3-Linear resonant motor array; 4-Finger limiting mechanism; 5-Force feedback module; 6-Voltage control; 7-Motor control; 8-Forearm tilt angle detection module; 9-Hand back posture angle sensor; 10-Finger flexure sensor; 11-Main control chip; 12-Wireless transmission; 13-PC terminal; 14-Display screen; 15-Battery 1; 16-Large payload octocopter UAV platform; 17-UAV power and navigation module; 18-Robotic arm controller; 19-Wireless transmission unit; 20-Robotic arm; 21-Camera module; 22-Battery 2; 23-External environment; 24-Dexterous hand pressure sensor; 25-Motor.

[0048] Figure 2 This is a schematic diagram of a drone performing a biomimetic retrieval operation to remove a kite entangled in a high-altitude power transmission line.

[0049] In the diagram: a- Operator 1; b- Operator 2; c- Tower; d- Heavy-duty octocopter UAV; e- Six-DOF robotic arm; f- Foreign object. Detailed Implementation

[0050] The present invention will now be described in further detail with reference to the accompanying drawings and specific implementation methods.

[0051] This invention provides a foreign object removal system for power transmission lines using an air-to-ground twin-assisted robotic arm, such as... Figure 1 As shown, it includes an airborne intelligent operation terminal and a ground control and motion-sensing interaction platform.

[0052] like Figure 1 As shown, the airborne intelligent operation terminal includes a heavy-duty octocopter UAV platform 16, a robotic arm 20, and a battery 2 (22). The heavy-duty octocopter UAV platform 16 is equipped with a UAV power and navigation module 17, which integrates a high-precision GPS positioning unit, a dual-antenna direction finder, and redundant IMU sensors. The robotic arm 20 is connected to the bottom of the UAV via a rigid linkage. The end of the robotic arm 20 integrates a dexterous hand pressure sensor 24 and an insulated flexible dexterous hand. The inside of the dexterous hand is covered with an anti-slip silicone layer to prevent foreign objects from slipping off. The robotic arm controller 18 controls the motors 25 of each joint through the motor drive unit to realize the position adjustment and dynamic impedance matching of the end effector in three-dimensional space. The camera module 21 is installed at the front end of the robotic arm 20 to collect high-definition images of the wire foreign object entanglement in real time, and transmits the image data, robotic arm position information, and clamping force data synchronously to the ground control and somatosensory interaction platform through the wireless transmission unit 19.

[0053] like Figure 1As shown, the ground control and somatosensory interaction platform includes a single right-hand wearable data glove 2, a PC terminal 13, a display screen 14, and a battery 1 (15). The data glove 2 is designed based on the human right hand and includes a forearm tilt angle detection module 8, a back-of-hand posture angle sensor 9, a finger flexure sensor 10, and a force feedback module 5. The forearm tilt angle detection module 8 is fixed near the wrist of the operator's arm 1 and has a built-in six-axis inertial measurement unit (IMU) to detect the tilt angle of the forearm relative to the horizontal plane in real time, providing a posture reference for the vertical lifting and lowering motion of the robotic arm. The back-of-hand posture angle sensor 9 is integrated on the back of the glove and uses a fusion algorithm of a three-axis gyroscope and an accelerometer to measure the pitch angle, yaw angle, and roll angle of the back of the hand, driving the three-dimensional spatial posture adjustment of the robotic arm's end effector. The finger flexure sensor 10 is placed at the finger joints and detects the bending angle of each finger through flexible resistance strain gauges, mapping it to the opening and closing control commands of the robotic arm's dexterous hand. The linear resonant motor array 3 of the force feedback module 5 is distributed in the palm and fingertip areas of the glove. Based on data from the dexterous hand pressure sensor 24, it dynamically adjusts the vibration mode: when in contact with a foreign object, the fingertip motors simulate tactile sensation with low-frequency, micro-amplitude vibrations; when the clamping force approaches the safety threshold of the wire, the palm motors switch to high-frequency, strong vibrations, simultaneously triggering a warning indicator on the display screen. The finger limiting mechanism 4 is located at the index finger joint, limiting excessive finger bending via a damping pivot. When the bending angle exceeds a preset safety threshold, a mechanical lock is triggered to prevent overload of the dexterous hand closure. The PC terminal 13 receives and parses the posture data, mapping the hand movements to six-degree-of-freedom joint angle commands of the robotic arm and opening / closing control signals of the dexterous hand through a preset algorithm. The display screen 14 displays real-time images of the wire / foreign object and dynamic values ​​of the clamping force transmitted by the camera module 21.

[0054] The ground control and motion-sensing interaction platform interacts in real time with the airborne intelligent operation terminal via wireless transmission: hand movements are encoded to drive the robotic arm 20 to precisely remove foreign objects, and the dexterous hand pressure and environmental data are transmitted back to the ground, triggering the tactile response of the force feedback module 5 and updating the interface status. If a communication interruption or abnormal posture is detected, the system immediately freezes the robotic arm's movements and activates the autonomous protection protocol to ensure that the entire live-line operation is safe and controllable.

[0055] The present invention provides a method for removing foreign objects from power transmission lines using an air-to-ground twin collaborative robotic arm, employing the aforementioned air-to-ground twin collaborative robotic arm system for removing foreign objects from power transmission lines, specifically including the following steps:

[0056] Step 1: The drone approaches and hovers.

[0057] The operator sends a takeoff command to the heavy-duty octocopter d via the LoRa communication module to achieve centimeter-level hovering positioning; the drone flies smoothly to the target tower c, and when the horizontal distance between the drone and the work point is reduced to 0.5 meters, the lidar ranging module triggers the automatic hovering program, and stabilizes the body attitude through the PID control algorithm, providing a stable platform for the robotic arm to work.

[0058] Step 2: Image transmission of the work environment.

[0059] The binocular camera module installed at the front end of the six-degree-of-freedom robotic arm is activated. The image data is compressed using H.265 encoding and transmitted to the ground PC via a wireless transmission unit. The display screen simultaneously shows two images: the main image is a 1080P@30fps high-definition video stream, and the auxiliary image is a projection of the robotic arm's kinematic model.

[0060] Step 3: Acquisition and fusion processing of motion-sensing interaction data.

[0061] Operator 2b, wearing a single-right-hand data glove, initiates the motion-sensing interaction system for initial calibration.

[0062] Finger motion acquisition: Flexible resistance strain gauges at the finger joints monitor the bending angle of each finger joint in real time: 0-90°. The bending angle is mapped to the dexterity hand opening and closing amplitude control command through normalization processing.

[0063] Hand posture analysis: The MPU6050 chip in the back of the hand posture acquisition module acquires raw data from the three-axis accelerometer and gyroscope at a sampling rate of 200Hz. After fusion using the Kalman filter algorithm, the back of the hand pitch angle, roll angle and yaw angle are output.

[0064] Forearm posture baseline establishment: The forearm posture acquisition unit uses a six-axis IMU module to continuously detect the tilt angle between the forearm and the horizontal plane.

[0065] Step 4: Kinematics calculation and instruction generation.

[0066] The ground PC receives the three-dimensional attitude data stream of the data glove, constructs the kinematic model of the robotic arm based on the improved DH parameter method, and converts the hand movements into a six-degree-of-freedom joint angle command sequence through inverse kinematics solution. During the solution process, the Jacobian matrix iteration method is used to optimize the trajectory smoothness and avoid the singular configuration of the robotic arm. The calculated control commands are encrypted and sent to the robotic arm controller of the air-based terminal through the wireless transparent transmission unit.

[0067] Constructing a kinematic model of a robotic arm involves the following steps:

[0068] S1: Obtain the mapping relationship between the data glove and the robotic arm, measure the range of values ​​of the end position of the data glove in the XYZ directions, and the range of values ​​of the end position of the robotic arm in the XYZ directions, so as to establish the mapping relationship between the two.

[0069] S2: Based on the DH parameter method, a forward kinematics model of the robotic arm is established, and the transformation matrix between the base coordinate system and the end-joint coordinate system is obtained by matrix transformation formula:

[0070]

[0071] The T matrix between two joints is:

[0072]

[0073] in, Indicates the length of the rod. For the twist angle, It is the horizontal spacing. Joint angle, This represents the transformation matrix from coordinate system j to coordinate system i.

[0074] S3: Based on the algebraic method and the actual operating range of the robotic arm, inverse kinematics is solved to obtain the transformation relationship between the body coordinate system and the end effector of the robotic arm, thereby establishing the correspondence between the end position of the robotic arm and the ideal servo motor rotation angle.

[0075] Step 5: Precise control and dynamic impedance adjustment of the robotic arm.

[0076] After parsing the instructions, the robotic arm controller drives the brushless servo motors to perform multi-axis linkage control:

[0077] The boom slewing joint and elbow pitch joint adopt position-velocity dual closed-loop PID control to achieve coarse positioning of the end effector.

[0078] The wrist yaw and pitch joints dynamically adjust their rotational inertia based on a fuzzy adaptive algorithm to compensate for drone hovering jitter.

[0079] The dexterous hand opening and closing mechanism receives finger bending degree mapping commands and drives the flexible dexterous hand to adjust the clamping force with a step accuracy of 0.1N through a harmonic reducer, while simultaneously triggering the contact force feedback of the pressure sensor.

[0080] Step 6: Tactile feedback and safety interlock mechanism.

[0081] The dexterous hand pressure sensor collects gripping force data at a frequency of 60Hz. After normalization processing by the robotic arm controller, the data is transmitted back to the ground platform wirelessly.

[0082] Tactile simulation: When the contact force is within a safe range, the linear resonant motor at the fingertips of the data glove will gradually increase the vibration amplitude and energy density as the applied force increases, simulating the sensation of a foreign object; while the vibration module in the palm area adopts a mode that combines short-term high-frequency impact with intermittent vibration to form a progressive tactile cues from weak to strong.

[0083] Over-limit warning: When the contact force reaches 90% of the wire safety threshold, the palm motor switches to high-frequency strong vibration mode, and the display screen simultaneously displays a red warning box and a buzzer alarm.

[0084] Emergency Lockout: If the pressure exceeds the safety threshold of the wire or a stall is detected in the joint motor, the main control chip immediately triggers the electromagnetic lock of the finger limit mechanism, forcing the dexterous hand to maintain its current state.

[0085] For example, when the dexterous hand experiences abnormal gripping force due to an angle deviation caused by foreign object entanglement, the thin-film pressure sensor integrated at its end collects contact force data in real time and transmits the encrypted data to a ground PC via a wireless transmission unit. The PC makes dynamic decisions based on a preset wire safety threshold model. If the contact force is detected to reach 90% of the wire damage warning value within a short period of time, a multi-level response is immediately triggered—the main control chip of the data glove sends pulse commands to the linear resonant motor array to generate high-frequency vibrations in the fingertip area to simulate the mechanical overload sensation, while the palm motor forms a gradient warning with intermittent strong vibrations at 120Hz. If the operator does not respond within 2 seconds to reduce the pressure below the wire damage warning value, the finger joint motors of the data glove are forcibly locked and the robotic arm is switched to constant torque mode.

[0086] Step 7: Safe evacuation and verification of operational effectiveness.

[0087] After the foreign object f is removed, the operator sends an evacuation command via remote control, and the drone evacuates to an observation point 10 meters away along a preset safe path; the camera module starts a second scan to verify that there is no residual damage on the surface of the wire; the ground platform generates an operation report, recording the peak clamping force, operation time, and comparison images before and after the foreign object removal.

[0088] The operating procedure of this invention system is as follows:

[0089] like Figure 2 As shown, the system of this invention achieves foreign object removal from power transmission lines through air-ground collaborative operation, specifically as follows: Operator A controls a drone D equipped with a six-DOF robotic arm E via a handheld remote control to fly to the area where the foreign object F is entangled near the power transmission tower C. The drone hovers stably using a high-precision navigation module, maintaining a safe working distance from the power line. Operator B wears data gloves, and the hand posture movements are captured by sensors to generate control commands, including finger bending to drive the robotic arm's dexterous hand opening and closing, and hand back rotation angle to adjust the spatial posture of the robotic arm's end effector. The commands are encoded on the PC and sent to the air-based intelligent operation terminal via a low-latency wireless transparent transmission link, driving the robotic arm to perform foreign object grasping and peeling actions.

[0090] The robotic arm's end effector, a dexterous hand, incorporates pressure sensors to collect gripping force data in real time. This data, combined with high-definition video streams captured by a camera module, is transmitted back to the ground platform via multi-channel concurrent technology. The PC analyzes the dexterous hand's pressure data and dynamically adjusts the vibration intensity and frequency of the force feedback module's motor: a low-frequency, micro-vibration is triggered in the fingertip area when lightly touching a foreign object, while a high-frequency, strong vibration is switched in the palm area when removing a high-resistance object. Simultaneously, the gripping force value and a safety threshold warning box are overlaid on the display screen. Operators can precisely control the dexterous hand's movements through a combined visual and tactile feedback system.

[0091] The system adopts a dual-power independent power supply design: the drone is powered by a built-in high-capacity battery 2, ensuring a continuous flight time of ≥45 minutes; the ground control platform is powered by an independent battery 1, supporting all-weather operation in outdoor mobile scenarios. After the robotic arm completes the removal of the foreign object, the drone autonomously returns to the take-off and landing point, and the operation data is uploaded to the cloud platform for archiving, achieving full traceability.

[0092] This invention, through the deep integration of air-to-ground twin collaborative control, haptic interaction, and dynamic force feedback technology, offers significant advantages over traditional methods for removing foreign objects from power transmission lines. For high-risk live-line work scenarios, traditional drone-mounted removal tools often suffer from incomplete object removal due to insufficient degrees of freedom and reliance on human experience. This invention, through seamless mapping of a six-degree-of-freedom robotic arm with haptic control, simplifies complex robotic arm operation to natural hand movements, significantly lowering the technical barrier for operators while avoiding the safety risks of manual climbing or high-temperature melting processes.

[0093] Existing drone robotic arm systems mostly rely on handles or remote control panels for operation, requiring operators to repeatedly switch perspectives to adjust the robotic arm's posture. This results in inherent drawbacks such as blind spots and a lack of tactile feedback. This invention innovatively introduces wearable data gloves and an augmented reality interface. Operators directly drive the robotic arm's three-dimensional movement through hand gestures, combined with real-time video feeds from a camera, achieving intuitive "what you see is what you control" interaction. Simultaneously, the force feedback module simulates changes in gripping force through tactile vibrations, triggering warning signals when foreign objects become stuck or contact force exceeds limits. This allows operators to adjust their force application strategy promptly, preventing wire damage or secondary entanglement due to uncontrolled force.

[0094] Traditional removal devices are limited by the rigidity of their mechanical structure, making them difficult to adapt to dynamic scenarios such as high-voltage wire swinging and complex foreign object entanglement (such as kite lines intertwined with plastic film). The robotic arm of this invention employs impedance adaptive adjustment technology at its end effector, combined with multi-modal sensor data to correct joint stiffness in real time. While ensuring stripping efficiency, it flexibly adapts to the vibration of wires and the deformation characteristics of foreign objects, significantly improving all-terrain operation capabilities.

Claims

1. A foreign object removal system for power transmission lines using a hybrid aerial-ground robotic arm, characterized in that, This includes airborne intelligent operation terminals and ground control and motion-sensing interaction platforms; The airborne intelligent operation terminal includes a heavy-duty octocopter UAV platform (16), a robotic arm (20), and a battery 2 (22). The heavy-duty octocopter UAV platform (16) is equipped with a UAV power and navigation module (17), which integrates a high-precision GPS positioning unit, a dual-antenna direction finder, and redundant IMU sensors. The robotic arm (20) is connected to the bottom of the UAV through a rigid link. The robotic arm controller (18) controls the motors (25) of each joint through the motor drive unit to realize the position adjustment and dynamic impedance matching of the end effector in three-dimensional space. The camera module (21) is installed at the front end of the robotic arm (20) to collect high-definition images of the wire foreign object entanglement in real time, and transmits the image data, robotic arm position information, and clamping force data to the ground control and somatosensory interaction platform through the wireless transmission unit (19). The ground control and somatosensory interaction platform includes a single right-hand wearable data glove (2), a PC (13), a display screen (14), and a battery 1 (15). The data glove (2) is designed based on the human right hand and includes a forearm tilt angle detection module (8), a back-of-hand posture angle sensor (9), a finger bending angle sensor (10), and a force feedback module (5). The forearm tilt angle detection module (8) is fixed to the operator's arm (1) near the wrist and has a built-in six-axis inertial measurement unit (IMU) to detect the tilt angle between the forearm and the horizontal plane in real time, providing a posture reference for the vertical lifting and lowering motion of the robotic arm. The back-of-hand posture angle sensor (9) is integrated on the back of the glove and uses a three-axis gyroscope and accelerometer fusion algorithm to measure the pitch angle, yaw angle, and roll angle of the back of the hand, driving the three-dimensional spatial posture adjustment of the robotic arm end. Curvature sensors (10) are installed at the finger joints. Flexible resistance strain gauges detect the bending angle of each finger and map it to the opening and closing control command of the dexterous hand of the robotic arm. The linear resonant motor array (3) of the force feedback module (5) is distributed in the palm and fingertip areas of the glove. The vibration mode is dynamically adjusted according to the data returned by the dexterous hand pressure sensor (24). The finger limiting mechanism (4) is installed at the index finger joint. The damping shaft limits the excessive bending of the finger. When the bending angle exceeds the preset safety threshold, the mechanical lock is triggered to prevent the dexterous hand from closing overload. The PC (13) receives and parses the posture data and maps the hand movements to the six-degree-of-freedom joint angle command of the robotic arm and the opening and closing control signal of the dexterous hand through the preset algorithm. The display screen (14) displays the wire foreign object image and the dynamic value of the clamping force returned by the camera module (21) in real time. The ground control and somatosensory interaction platform interacts with the airborne intelligent operation terminal in real time via wireless transmission: after the hand movements are encoded, the robotic arm (20) is driven to accurately perform the foreign object retrieval, and the dexterous hand pressure and environmental data are transmitted back to the ground, triggering the tactile response and interface status update of the force feedback module (5); if a communication interruption or abnormal posture is detected, the system immediately freezes the robotic arm movement and starts the autonomous protection protocol to ensure that the live operation is safe and controllable throughout.

2. The air-to-ground twin collaborative robotic arm foreign object removal system for power transmission lines according to claim 1, characterized in that, The robotic arm (20) integrates a dexterous hand pressure sensor (24) and an insulated flexible dexterous hand at its end. The inside of the dexterous hand is covered with an anti-slip silicone layer to prevent foreign objects from slipping off.

3. The air-to-ground twin collaborative robotic arm system for removing foreign objects from power transmission lines according to claim 1, characterized in that, The linear resonant motor array (3) dynamically adjusts the vibration mode as follows: when in contact with foreign objects, the fingertip motor simulates the touch with low-frequency micro-vibration; when the clamping force approaches the safety threshold of the wire, the palm motor switches to high-frequency strong vibration and simultaneously triggers the warning sign on the display screen.

4. A method for removing foreign objects from power transmission lines using an air-to-ground twin-assisted robotic arm, employing the air-to-ground twin-assisted robotic arm foreign object removal system for power transmission lines as described in any one of claims 1-3, characterized in that... Includes the following steps: Step 1: The drone approaches and hovers; The operator sends a takeoff command to the heavy-duty octocopter UAV (d) via the LoRa communication module to achieve centimeter-level hovering positioning; the UAV flies smoothly to the top of the target tower (c), and when the horizontal distance between the UAV and the work point is reduced to 0.5 meters, the lidar ranging module triggers the automatic hovering program, and stabilizes the body attitude through the PID control algorithm, providing a stable platform for the robotic arm to work; Step 2: Image transmission of the work environment; The binocular camera module installed at the front end of the six-degree-of-freedom robotic arm (e) is activated. The image data is compressed by H.265 encoding and transmitted to the ground PC through the wireless transparent transmission unit. The display screen simultaneously displays two images: the main image is a 1080P@30fps high-definition video stream, and the auxiliary image is a projection of the robotic arm's kinematic model. Step 3: Acquisition and fusion processing of motion-sensing interaction data; Operator 2 (b) puts on a single right-hand data glove and starts the motion-sensing interaction system to initialize calibration; Finger motion acquisition: Flexible resistance strain gauges at the finger joints monitor the bending angle of each finger joint in real time: 0-90°. The bending angle is mapped to the dexterity hand opening and closing amplitude control command through normalization processing. Hand posture analysis: The MPU6050 chip in the back of the hand posture acquisition module acquires raw data from the three-axis accelerometer and gyroscope at a sampling rate of 200Hz. After fusion using the Kalman filter algorithm, the back of the hand pitch angle, roll angle and yaw angle are output. Forearm posture baseline establishment: The forearm posture acquisition unit uses a six-axis IMU module to continuously detect the tilt angle of the forearm relative to the horizontal plane; Step 4: Kinematics calculation and instruction generation; The ground PC receives the three-dimensional attitude data stream of the data glove, constructs the kinematic model of the robotic arm based on the improved DH parameter method, and converts the hand movements into a six-degree-of-freedom joint angle command sequence through inverse kinematics solution; during the solution process, the Jacobian matrix iteration method is used to optimize the trajectory smoothness and avoid the singular configuration of the robotic arm; after the solution is encrypted, the control commands are sent to the robotic arm controller of the air-based terminal through the wireless transparent transmission unit. Step 5: Precise control and dynamic impedance adjustment of the robotic arm; After parsing the instructions, the robotic arm controller drives the brushless servo motors to perform multi-axis linkage control: The boom slewing joint and elbow pitch joint adopt position-velocity dual closed-loop PID control to achieve coarse positioning of the end effector; The wrist yaw and pitch joints dynamically adjust their rotational inertia based on a fuzzy adaptive algorithm to compensate for drone hovering jitter. The dexterous hand opening and closing mechanism receives finger bending degree mapping instructions and drives the flexible dexterous hand to adjust the clamping force with a step accuracy of 0.1N through a harmonic reducer, while simultaneously triggering the contact force feedback of the pressure sensor; Step 6: Tactile feedback and safety interlock mechanism; The dexterous hand pressure sensor collects gripping force data at a frequency of 60Hz. After normalization processing by the robotic arm controller, the data is transmitted back to the ground platform wirelessly. Tactile simulation: When the contact force is within a safe range, the linear resonant motor at the fingertips of the data glove will gradually increase the vibration amplitude and energy density as the applied force increases, simulating the feeling of a foreign object; while the vibration module in the palm area adopts a mode that combines short-term high-frequency impact with intermittent vibration to form a progressive tactile cues from weak to strong. Over-limit warning: When the contact force reaches 90% of the wire safety threshold, the palm motor switches to high-frequency strong vibration mode, and the display screen simultaneously displays a red warning box and a buzzer alarm. Emergency Lockout: If the pressure exceeds the safety threshold of the wire or a stall of the joint motor is detected, the main control chip immediately triggers the electromagnetic lock of the finger limit mechanism, forcing the dexterous hand to maintain its current state; Step 7: Safe evacuation and verification of operational effectiveness; After the foreign object (f) is removed, operator 1 (a) sends an evacuation command via remote control, and the drone evacuates to an observation point 10 meters away along a preset safe path; the camera module starts a secondary scan to verify that there is no residual damage on the surface of the wire; the ground platform generates an operation report, recording the peak clamping force, operation time, and comparison images before and after the foreign object removal.

5. The method for removing foreign objects from power transmission lines using a ground-to-air twin-assisted robotic arm according to claim 4, characterized in that, The construction of the kinematic model of the robotic arm includes the following steps: S1: Obtain the mapping relationship between the data glove and the robotic arm, measure the range of values ​​of the end position of the data glove in the XYZ directions, and the range of values ​​of the end position of the robotic arm in the XYZ directions, so as to establish the mapping relationship between the two. S2: Based on the DH parameter method, a forward kinematic model of the robotic arm is established, and the transformation matrix between the base coordinate system and the end-joint coordinate system is obtained by the matrix transformation formula: ； The T matrix between two joints is: ； in, Indicates the length of the rod. For the twist angle, Horizontal spacing Joint angle, Represents the transformation matrix from coordinate system j to coordinate system i; S3: Based on the algebraic method and the actual operating range of the robotic arm, inverse kinematics is solved to obtain the transformation relationship between the body coordinate system and the end effector of the robotic arm, thereby establishing the correspondence between the end position of the robotic arm and the ideal servo motor rotation angle.

Images