A pre-contact synchronous wire-attaching control method and system for UAV-mounted robots oriented towards split conductors
By constructing a virtual stationary coordinate system and multi-mode switching control, the problem of synchronous following and access before contact between the UAV-mounted robot and the split conductor was solved, thereby reducing relative motion deviation and improving safety.
Patent Information
- Application Number
- CN202611269657.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-20
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies lack effective control methods when drones deploy robots to split conductors, resulting in asynchrony between the position, speed, and attitude of the hanging end and the target sub-conductor. This may lead to problems such as excessive relative contact speed, excessive impact of hanging, accidental collision with adjacent sub-conductors, or drone attitude instability.
A virtual stationary coordinate system is constructed. The target and adjacent sub-traverses are identified through multi-source sensing data, a pre-aiming stationary point is generated, the synchronization error is calculated, the equivalent disturbance compensation is decomposed, and multi-mode switching control is implemented, including synchronous following, pre-contact compliance, hanging clamp, and abnormal withdrawal modes, to ensure that the hanging end and the swaying target sub-traverse synchronously follow and enter before contact.
It effectively reduces the relative motion deviation between the hanging end and the conductor, avoids collisions and impacts, improves approach safety and operational feasibility, enhances the pertinence and robustness of control, and ensures the safety of the entire process of pre-contact line-holding operations.
Smart Images

Figure CN123114015A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent maintenance operation technology for power transmission lines, specifically to a pre-contact synchronous wire-attaching control method and system for a drone-mounted robot for split conductors. Background Technology
[0002] With the continuous growth in the scale of ultra-high voltage transmission lines, cross-river and cross-sea lines, long-span lines in mountainous areas, and new energy transmission lines, the maintenance tasks of transmission lines are becoming increasingly high-altitude, complex, and risky. Using drones to deploy robots to the conductors can reduce the risks of personnel climbing towers and working along the lines at high altitudes, thereby improving maintenance efficiency.
[0003] Existing technologies include solutions that focus on disturbance rejection and sway reduction for UAV-borne load systems. These solutions improve stability during transport by establishing nonlinear dynamic models of the UAV-borne load system, using disturbance observers and robust controllers. Other solutions address the collaborative loading and unloading of UAVs and power line robots, employing UAV-load dynamic modeling, extended state observers, visual servoing, six-dimensional force sensors, and load trajectory tracking control to achieve attitude correction and trajectory tracking during UAV-borne robot transport.
[0004] The above-mentioned solutions are of positive significance for improving the stability of drones carrying robots during transportation. However, in the actual scenario of drones deploying robots to split conductors, the target conductor is not a static rigid target, but rather sways under the combined effects of wind disturbance, line tension, span vibration, drone downwash airflow, and robot approach disturbance. Especially in two-split, four-split, or six-split conductor scenarios, there are spatial gaps and no-entry constraints between the target sub-conductor and adjacent sub-conductors. If only ordinary visual tracking, load anti-swing, or trajectory tracking is performed before the robot's wire-hanging end contacts the target sub-conductor, problems such as excessive contact relative velocity, excessive wire-hanging impact, accidental collision with adjacent sub-conductors, clamping deviation, or drone attitude instability may still occur.
[0005] Therefore, existing technologies still lack a dedicated control method for scenarios where swaying split conductors are stuck before contact, which can incorporate the spatial relationship between the target sub-conductor and adjacent sub-conductors into the control constraints before the hanging end contacts the target sub-conductor, and reduce the risk of hanging by means of synchronous following and admission judgment before contact. Summary of the Invention
[0006] The purpose of this invention is to provide a pre-contact synchronous wire-attaching control method and system for a drone-mounted robot oriented towards split conductors, in order to solve the problem in the prior art that before the wire-attaching end contacts the target sub-conductor, the wire-attaching end and the swaying target sub-conductor are not synchronized in position, speed and attitude, and will accidentally collide with adjacent sub-conductors, resulting in large wire-attaching impact and low wire-holding success rate.
[0007] To achieve the above objectives, the technical solution provided by the present invention is as follows: A pre-contact synchronous wire-attaching control method for an unmanned aerial vehicle (UAV) mounted robot for split conductors includes the following steps: S1: Acquire the drone pose, the robot wire-attaching end pose, the split wire sensing data, and the robot wire-attaching end contact status data; S2: Identify the target sub-traverse and its adjacent sub-traverses using the split traverse sensing data, determine the reference point and direction of the target sub-traverse and the reference point of the adjacent sub-traverse, and construct a virtual stationary coordinate system; estimate the motion trend of the target sub-traverse and generate the pre-aiming stationary point in the virtual stationary coordinate system; calculate the position, velocity and attitude synchronization error based on the robot's hanging end pose and the pre-aiming stationary point; S3: Based on the robot's wire-hanging end pose, the split conductor perception data, and the deviation between the current control input of the UAV and their respective predicted values, obtain the observation residuals of the wire-hanging end, the conductor, and the control input, calculate the equivalent disturbance compensation amount, project it onto the virtual stationary coordinate system, and decompose it into components along the conductor direction, the radial approach direction, and the split conductor plane normal component. S4: Correct the position of the pre-aiming stationary point using the component along the traverse direction, and generate the synchronization stationary control quantity from the corrected pre-aiming stationary point, each synchronization error, the radial approach direction, and the normal component of the split traverse plane; calculate the UAV's desired motion command based on the synchronization stationary control quantity and the UAV's pose. S5: Perform multi-mode switching control based on synchronization error, line-holding feasibility, and robot wire-attaching end contact status data. The multi-mode includes synchronous following mode, pre-contact compliant mode, wire-attaching clamping mode, and abnormal retraction mode. Through multi-mode switching, synchronous following and wire-attaching access control between the wire-attaching end and the swaying target sub-conductor before contact are achieved.
[0008] To optimize the above technical solution, the specific measures also include: In step S2, the split traverse sensing data includes split traverse image features, laser ranging point cloud, and inertial measurement data; the inertial measurement data is used to compensate for motion distortion of the laser ranging point cloud, and each sampling point is transformed to a coordinate system of a unified reference time according to its timestamp and the UAV attitude and position changes; based on the compensated laser ranging point cloud and split traverse image features, the spatial distribution relationship of the target sub-traverse and its adjacent sub-traverses in the split traverse is identified, and the reference point of the target sub-traverse, the direction of the target sub-traverse, and the reference point of the adjacent sub-traverse are determined; Furthermore, a virtual stationary coordinate system, updated as the target sub-traverse moves, is constructed based on the direction of the target sub-traverse and the reference points of adjacent sub-traverses. The specific process is as follows: Obtain two fitting points on the center line of the target sub-traverse. and and the target sub-traverse reference point ; Obtain reference points for adjacent sub-traverses and By fitting points and Normalize the direction of the connecting line to obtain the unit vector of the target sub-traverse direction. ; Through the target sub-traverse reference point Pointing to the reference points of the adjacent sub-traverses on both sides and The cross product of two spatial vectors, followed by normalization, yields the unit vector normal to the plane of the split traverse. ; through the split traverse plane normal unit vector Unit vector of the target sub-traverse direction After performing the cross product and normalizing, we obtain the unit vector of the radial approach direction. .
[0009] When adjacent sub-traverses are obstructed or the plane normal of the split traverse cannot be stably calculated due to sparse point clouds, the alternative normal unit vector is determined according to the following priority: if the adjacent sub-traverse is temporarily obstructed, the historical normal unit vector of the previous moment is used; if the point cloud is sparse and the line design parameters are available, the theoretical normal unit vector of the sub-traverse layout determined according to the line design parameters is used; if a stable design direction cannot be obtained, the alternative normal unit vector determined according to the current safe approach direction of the UAV is used; the determined alternative normal unit vector is normalized and used as the plane normal unit vector of the split traverse at the current moment.
[0010] Further, in step S2, the generation of the pre-aiming stationary point in the virtual stationary coordinate system before contact specifically involves: ; in, To pre-aim at the position of the stationary point; This is the aiming distance along the direction of the guide wire; This is the safe radial offset before contact; To avoid the normal offset of adjacent sub-traverses; When the conductor sway intensity increases or the recognition confidence decreases, the safety radial offset or normal offset is adaptively increased; when the position synchronization error is continuously lower than the preset threshold, the safety radial offset is gradually reduced so that the hanging end enters the safe approach area corresponding to the pre-contact compliance mode.
[0011] In step S2, the calculation of position, velocity, and attitude synchronization errors based on the robot's wire-attaching end pose and the pre-aiming stationary point specifically involves: ; ; ; ; in, This refers to the position synchronization error; This refers to speed synchronization error; For attitude synchronization error, This indicates the location of the robot's cable attachment point; For the speed at the robot's wire-attaching end; The unit vector representing the opening direction of the wire-hanging mechanism; The attitude matrix for the virtual stationary coordinate system; This is the corrected rate of change of the pre-contact aiming stationary point.
[0012] Further, in step S3, the calculation of the equivalent disturbance compensation amount, projecting it onto the virtual stationary coordinate system, and decomposing it into components along the conductor direction, the radial approach direction, and the normal component of the split conductor plane, specifically: ; in, This is the equivalent disturbance compensation amount; To observe the residual at the hanging end, This represents the actual pose observation value of the robot's wire-attaching end. The model prediction value for the robot's wire-attaching end; For traverse observation residuals, The actual observed values of the motion state of the target sub-traverse. The predicted value of the motion state model of the target sub-traverse; To control the input residual, This is the current actual control input for the drone. The desired control input is generated based on the pre-aimed stationary point; , , This is the weight matrix; The component along the conductor direction is used to correct the position of the pre-aiming stationary point along the conductor: the component along the conductor direction is multiplied by a preset disturbance correction coefficient along the conductor direction to obtain the correction amount of the pre-aiming distance along the conductor direction, and the pre-aiming distance along the conductor direction is updated using this correction amount to obtain the corrected pre-aiming stationary point; the radial approach direction component is used to suppress the radial impact when the hanging end approaches the target sub-conductor; the split conductor plane normal component is used to suppress the hanging end from deviating from the split conductor plane where the target sub-conductor is located or entering the interference region of adjacent sub-conductors.
[0013] Further, in step S4, the generation of the synchronization control quantity from the corrected pre-aiming stationary point, each synchronization error, the radial approach direction, and the normal component of the split traverse plane specifically involves: ; in, Desired acceleration for the robot's wire-attached end; This is the corrected estimate of the acceleration at the pre-contact aiming stationary point; , , These are the position, velocity, and attitude feedback gains, respectively. The equivalent inertia of the drone and robot in the direction of the hanging end; The radial approach direction component; For the plane normal component of the split traverse; and These are the compensation weights for the radial approach direction component and the plane normal component of the split traverse, respectively.
[0014] In step S5, the feasibility of stationary operation determined based on the normal contact force in the robot's wire-attaching end contact state data specifically includes: The degree to which the normal contact force deviates from the safe range is calculated. The degree of deviation is mapped to the interval of 0 to 1 using the Sigmoid function to obtain the feasibility evaluation value for the suspension line. The degree of deviation of the normal contact force from the safe range is defined as follows: when the normal contact force is less than the minimum contact force required for reliable suspension, the deviation is the difference between the minimum contact force and the normal contact force; when the normal contact force is greater than the maximum contact force to prevent damage to the conductor or suspension mechanism, the deviation is the difference between the normal contact force and the maximum contact force; when the normal contact force is within the safe range between the minimum and maximum contact forces required for reliable suspension, the deviation is 0.
[0015] Further, in step S5, the multi-mode switching control based on synchronization error, line-holding feasibility, and robot wire-attaching end contact state data specifically includes: In the coarse approach mode, the UAV flies to the vicinity of the target conductor based on positioning. When the target sub-conductor is stably identified for a first preset number of consecutive frames and the distance to the UAV is less than a first distance threshold, it switches to the synchronous following mode; In the synchronous following mode, when the position synchronization error, velocity synchronization error, and attitude synchronization error meet the threshold requirements for a second preset number of consecutive frames, and the line-holding feasibility evaluation value is greater than a preset evaluation threshold, it switches to the pre-contact compliant mode; In the pre-contact compliant mode, the gain components of position control gain and velocity control gain in the radial approach direction are reduced. When the normal contact force and its rate of change meet the safety conditions for a third preset number of consecutive frames, it switches to the wire-attaching clamping mode; In the wire-attaching clamping mode, the wire-attaching mechanism is controlled to close, and the wire-attaching state is confirmed based on contact force feedback; In the abnormal retreat mode, when the abnormal triggering conditions are met, the UAV is controlled to retreat to a safe airspace along a direction away from the plane normal of the split conductor.
[0016] As another important technical solution, the present invention also provides a pre-contact synchronous wire-attaching control system for a UAV-mounted robot oriented towards split conductors, realizing the pre-contact synchronous wire-attaching control method for a UAV-mounted robot oriented towards split conductors as described above, including: The multi-source sensing and state acquisition module is used to acquire the pose of the UAV, the pose of the robot's wire-attached end, the sensing data of the split wire, and the contact state data of the robot's wire-attached end. The target recognition and synchronization error calculation module is used to identify the target sub-traverse and its adjacent sub-traverse from the split traverse sensing data, determine the reference point and direction of the target sub-traverse and the reference point of the adjacent sub-traverse, and construct a virtual stationary coordinate system; estimate the motion trend of the target sub-traverse, generate the pre-aiming stationary point in the virtual stationary coordinate system; and calculate the position, velocity and attitude synchronization error based on the robot's hanging end pose and the pre-aiming stationary point. The equivalent disturbance compensation and decomposition module is used to obtain the observation residuals of the hanging end, the wire and the control input based on the robot hanging end pose, the split wire perception data and the deviation between the current control input of the UAV and their respective predicted values, calculate the equivalent disturbance compensation amount, project it onto the virtual stationary coordinate system, and decompose it into components along the wire direction, the radial approach direction and the split wire plane normal component. The synchronization stationary control quantity generation module is used to correct the position of the pre-aiming stationary point using the component along the traverse direction. The synchronization stationary control quantity is generated from the corrected pre-aiming stationary point, each synchronization error, the radial approach direction, and the normal component of the split traverse plane. The desired motion command of the UAV is calculated based on the synchronization stationary control quantity and the UAV pose. The line-standing decision and mode management module is used to perform multi-mode switching control based on synchronization error, line-standing feasibility and robot wire-attaching end contact status data. The multi-mode includes synchronous following mode, pre-contact compliant mode, wire-attaching clamping mode and abnormal withdrawal mode. Through multi-mode switching, the synchronous following and wire-attaching access control between the wire-attaching end and the swaying target sub-lead before contact is realized.
[0017] The present invention also proposes an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements a pre-contact synchronous wire attachment control method for a UAV-mounted robot oriented to split conductors as described above.
[0018] The present invention also proposes a computer-readable storage medium storing a computer program that enables a computer to execute a pre-contact synchronous wire-attaching control method for a UAV-mounted robot oriented to split conductors, as described above.
[0019] Compared with the prior art, the beneficial effects of the present invention are: This invention constructs a virtual stationary coordinate system that is updated in real time with the swaying of the target sub-conductor and generates a pre-aiming stationary point before contact. This allows the hanging end to continuously track the movement trend of the conductor during the approach process, thereby maintaining synchronization of position, speed and attitude at the moment of contact. This effectively reduces the relative motion deviation between the hanging end and the conductor and avoids excessive collision impact caused by speed or position mismatch.
[0020] This invention identifies the target sub-traverse while obtaining the spatial position of adjacent sub-traverses, uses their geometric relationship to determine the normal constraint of the split traverse plane, and introduces normal offset in the aiming point and control quantity to ensure that the hanging end is always restricted within the correct approach path, effectively avoiding the risk of the hanging end accidentally entering the interference area of adjacent sub-traverses, and improving the approach safety and operational feasibility in split traverse operation scenarios.
[0021] This invention unifies the observation residual and control input residual into an equivalent disturbance, and decomposes it along the traverse direction, the radial approach direction, and the normal direction of the split traverse plane. Compensation is applied to the impact on the safety of the stationary line in different directions, which not only corrects the pre-aiming deviation of the position along the line, but also suppresses the impact during radial approach, while maintaining stable isolation in the normal direction. This makes the system's anti-disturbance behavior match the working geometry characteristics, and improves the pertinence and robustness of pre-contact control.
[0022] This invention features a multi-mode switching mechanism, including synchronous following, pre-contact compliance, wire clamping, and abnormal retraction. Each mode progresses step-by-step based on synchronization accuracy and contact status, ensuring that the next stage can only be entered when conditions are met. This avoids attitude instability or contact loss of control due to blind feeding. Meanwhile, the abnormal retraction mode can actively control the UAV to move away from the wire when identification is lost or the status exceeds limits, which helps ensure the safety of the entire pre-contact line-holding operation and also leaves room for fault handling in subsequent engineering applications. Attached Figure Description
[0023] Figure 1 : A schematic diagram of the overall process of the pre-contact synchronous stationary line control method in this embodiment of the invention.
[0024] Figure 2 : Schematic diagram of the construction of the virtual stationary coordinate system of the split conductor in this embodiment of the invention.
[0025] Figure 3 : A schematic diagram of the equivalent disturbance decomposition and compensation control process in an embodiment of the present invention.
[0026] Figure 4 : A schematic diagram of the multi-mode switching mechanism in this embodiment of the invention.
[0027] Figure 5 This is a schematic diagram of the process for evaluating the feasibility of stationing and determining the admission criteria for line hanging in this embodiment of the invention. Detailed Implementation
[0028] The present invention will be further described in detail below through specific embodiments, but it should not be construed as limiting the scope of the subject matter of the present invention to the following embodiments. All technologies implemented based on the above content of the present invention fall within the scope of the present invention.
[0029] In some implementations, such as Figure 1 As shown, this invention provides a pre-contact synchronous wire attachment control method for a drone-mounted robot oriented towards split conductors, comprising the following steps: S1: Acquire the UAV pose, the robot's wire-attaching end pose, the split wire sensing data, and the robot's wire-attaching end contact status data; the split wire sensing data includes split wire image features, laser ranging point cloud, and inertial measurement data. In some implementations, after the drone carrying the robot flies to the vicinity of the power transmission line, it uses an airborne camera to acquire images of the conductors, including each sub-conductor in the split conductor. Line features are extracted from the conductor images to obtain the features of the split conductor images, including the position of the image center line corresponding to each sub-conductor, the direction of the image center line, and the relative positional relationship between each sub-conductor. These features are used to distinguish each sub-conductor in the split conductor and determine their arrangement relationship in the image.
[0030] The split traverse is sampled by a spatial ranging unit to obtain multiple spatial ranging points. Based on the ranging direction, ranging distance and installation pose of the spatial ranging unit corresponding to each spatial ranging point, the spatial ranging points are transformed into a preset coordinate system to form a laser ranging point cloud, which includes multiple three-dimensional spatial sampling points on the surface or near the center line of each sub-traverse.
[0031] Based on the image features of the split traverse, the sub-traverse points in the laser ranging point cloud are associated or grouped, and spatial straight line fitting is performed on the point cloud corresponding to each sub-traverse to obtain the spatial centerline of each sub-traverse in the split traverse: ; in, Indicates the first Root wire in parameters Spatial location; Indicates the first Reference point on the center line of the root conductor; Indicates the first The unit vector of the direction of the centerline of the root conductor; This represents a parameter along the direction of the conductor's centerline.
[0032] In some implementations, the UAV pose includes the UAV's three-dimensional position in a preset world coordinate system, as well as roll, pitch, and yaw angles; the robot wire-attaching end pose includes the robot wire-attaching end's three-dimensional position in the world coordinate system and the attitude corresponding to the opening direction of the wire-attaching mechanism, calculated based on the UAV pose, the geometric parameters of the mounting connection mechanism, and the geometric parameters of the robot wire-attaching mechanism; the split conductor image features include the image line features corresponding to each sub-conductor, the centerline pixel position, the image direction, and the sub-conductor spacing features; the laser ranging point cloud includes three-dimensional spatial sampling points on the surface of the split conductor or near the centerline collected by the laser ranging device; the inertial measurement data includes the attitude angles, angular velocities, and accelerations collected by the IMU; the robot wire-attaching end contact state data includes the contact markers between the wire-attaching end and the target sub-conductor, the normal contact force, the rate of change of contact force, and the force or displacement state of the wire-attaching mechanism.
[0033] The inertial measurement unit includes an IMU, which is used to acquire the UAV's attitude angles, angular velocities, and acceleration information.
[0034] The robot is equipped with a wire-hanging end or a clamping end. The wire-hanging end is equipped with a force sensor, a current feedback unit, or a displacement feedback unit to obtain the normal contact force between the wire-hanging end and the target sub-conductor, or the force state and displacement state of the wire-hanging mechanism.
[0035] The normal contact force is measured directly by a force sensor or estimated based on feedback from the drive current or displacement of the robot actuator.
[0036] The rate of change of contact force is determined based on the normal contact force obtained at continuous sampling times. The rate of change of normal contact force is used to characterize the degree of change of contact force over time during the contact process between the hanging end and the target sub-conductor.
[0037] In some implementations, the flight controller is used to generate UAV flight control commands based on pre-contact synchronous standby control quantities; the robot controller is deployed on the robot to control the actions of the robot's wire-attaching mechanism, gripping mechanism, release mechanism, or retraction mechanism.
[0038] S2: Identify the target sub-traverse and its adjacent sub-traverses using the split traverse sensing data, determine the reference point and direction of the target sub-traverse and the reference point of the adjacent sub-traverse, and construct a virtual stationary coordinate system; estimate the motion trend of the target sub-traverse and generate the pre-aiming stationary point in the virtual stationary coordinate system; calculate the position, velocity and attitude synchronization error based on the robot's hanging end pose and the pre-aiming stationary point; Motion distortion compensation is performed on the laser ranging point cloud using inertial measurement data. Each sampling point is transformed to a coordinate system based on its timestamp and the UAV's attitude and position changes. Based on the compensated laser ranging point cloud and the features of the split traverse image, the spatial distribution relationship of the target sub-traverse and its adjacent sub-traverses in the split traverse is identified. Based on the maintenance task, the opening direction of the wire-hanging mechanism, the safe distance between sub-conductors, and the approach direction of the UAV, select the target sub-conductor from among the sub-conductors: ; in, For the target sub-wire; For the hanging end to the first The distance of the root wire, The angle between the opening direction of the wire-hanging mechanism and the direction of the sub-conductor. The risk of interference with adjacent sub-traverses when approaching this sub-traverse is determined based on the safe distance between sub-traverses and the approach direction of the UAV. j For each candidate sub-conductor, calculate the minimum distance between the sweep envelope of the hanging end and the adjacent sub-conductor along the preset proximity path; increase the minimum distance if this minimum distance is less than the preset safety distance. Simultaneously, calculate the angle between the UAV's current approach direction and the radial approach direction of the candidate sub-trajectory; the larger the angle, the greater the increase. ; The task priority cost corresponding to this sub-conductor is determined by the maintenance task, and the candidate sub-conductor with the smaller cost is selected for priority processing. ; to These are the weighting coefficients.
[0039] A virtual stationary coordinate system is constructed based on the direction of the target sub-traverse and the reference points of adjacent sub-traverses, and is updated as the target sub-traverse moves. In this system, the first coordinate axis represents the direction of the target sub-traverse, the second coordinate axis represents the radial approach direction, and the third coordinate axis represents the normal or approximate normal of the split traverse plane. The specific process is as follows: In some implementations, such as Figure 2 As shown, two fitting points are obtained on the center line of the target sub-traverse. and and the target sub-traverse reference point ; Obtain reference points for adjacent sub-traverses and Unit vector of the target sub-traverse direction for: ; Unit vector of the plane normal of the split traverse for: ; Radial approach direction unit vector for: ; The unit vector normal to the plane of the split traverse Unit vector of the target sub-traverse direction After performing the cross product and normalizing, we obtain the unit vector of the radial approach direction. .
[0040] The positions of the target sub-traverse reference points obtained at different times are transformed to a unified reference coordinate system based on the corresponding timestamps and UAV pose changes. The lateral displacement of the target sub-traverse reference points is extracted along the radial approach direction of the virtual stationary line coordinate system to obtain the time series of the lateral displacement of the target sub-traverse: ; in, For the first Lateral displacement of the target sub-traverse reference point at each sampling time; For the first The position of the target sub-traverse reference point at each sampling time; The average position of the target sub-traverse reference points within the preset observation time window; It is the transpose of the unit vector in the radial approach direction.
[0041] The time intervals corresponding to adjacent lateral displacement peaks in the same direction are determined based on the lateral displacement time series of the target sub-conductor, and the lateral swaying period of the target sub-conductor is determined based on the time intervals between multiple adjacent lateral displacement peaks in the same direction. Target sub-guide lateral sway frequency for: ; The lateral sway frequency of the target sub-traverse is not measured directly by an independent frequency sensor, but is estimated based on the continuous position observation results of the target sub-traverse reference point obtained by fusion of visual and laser ranging.
[0042] In some implementations, when adjacent sub-traverses are occluded or the plane normal of the split traverse cannot be stably calculated due to sparse point clouds, the alternative normal unit vector is determined according to the following priority: : in, The unit vector of the split traverse plane normal obtained from the stable calculation of the previous sampling period; The design normal unit vector is determined based on the digital three-dimensional model of the transmission line, which includes the number of split conductors, the spacing between sub-conductors, the spatial arrangement of sub-conductors, and the conductor structure layout parameters. This is the normal unit vector corresponding to the safe approach direction determined based on the spatial relationship between the current position of the UAV and the target sub-traverse.
[0043] when or If the condition is not met, it is determined that a valid point cloud that meets the accuracy requirements cannot be obtained, and it is considered that the current point cloud cannot stably support the calculation of the plane normal of the split traverse.
[0044] in, This represents the number of effective point clouds used for plane fitting of the split conductor in the current sampling period. To ensure the minimum effective point cloud quantity threshold required for plane normal calculation; The threshold for the change in normal angle between adjacent sampling periods.
[0045] Unlike methods that rely solely on real-time vision or point cloud detection to obtain conductor attitude, this implementation utilizes historical spatial states, prior knowledge of line design, and UAV safe proximity constraints to form a multi-level normal recovery mechanism, ensuring that the virtual stationary coordinate system maintains continuity even under conditions of partial occlusion and perception degradation of the split conductor.
[0046] When a line is accepted and completed, the digital 3D model is considered unusable. This means that the number of split conductors, the spacing between sub-conductors, the spatial arrangement, and the conductor structure layout parameters that match the current tower span can be obtained from the line completion data or digital ledger, and can be converted to the current working coordinate system. When the above data is missing, cannot be matched with the current span, or the coordinate conversion error exceeds the preset tolerance, the digital 3D model of the transmission line is considered unusable.
[0047] The determined alternative normal unit vector is normalized to become the current normal unit vector of the split traverse plane: ; Since the amplitude and velocity of conductor sway in actual transmission lines are difficult to measure directly, this embodiment uses a continuous observation sequence to estimate the motion state of the target sub-conductor. : ; in, The rate of change of the reference point position. This represents the rate of change of direction in the virtual stationary coordinate system.
[0048] The following prediction and update methods are adopted: ; ; in, To predict the state, This is the updated state. The conductor observations are obtained by fusing visual and laser ranging. For filter gain, For the observation matrix, This is process noise.
[0049] In this way, the present invention does not require direct measurement of the amplitude or speed of the conductor swing, but improves the feasibility of the project by continuously observing and calculating the movement trend of the target sub-conductor.
[0050] Generate the pre-contact stationary points in the virtual stationary coordinate system, specifically as follows: ; in, To pre-aim at the position of the stationary point; This is the aiming distance along the direction of the guide wire; This is the safe radial offset before contact; To avoid the normal offset of adjacent sub-traverses; In some implementations... The value ranges from 0.05m to 0.50m. The value ranges from 0.05m to 0.30m. The value ranges from 0.05m to 0.40m, and the specific value is calibrated based on the UAV response delay, conductor sway intensity, split conductor spacing, and hanging mechanism dimensions.
[0051] When the conductor sway intensity increases or the recognition confidence decreases, the safety radial offset or normal offset is adaptively increased; when the position synchronization error is continuously lower than the preset threshold, the safety radial offset is gradually reduced so that the hanging end enters the safe approach area corresponding to the pre-contact compliance mode.
[0052] In some implementations, the conductor sway intensity is determined based on the rate of change of position, rate of change of velocity, or root mean square index of the target sub-conductor reference point within the sliding time window; the recognition confidence is determined jointly based on the image line feature matching confidence and the point cloud fitting residual. When the sway intensity exceeds a preset sway threshold or the recognition confidence is lower than a preset confidence threshold, the safety radial offset is increased by a preset step size. or normal bias And not exceeding their respective preset upper limits; when the position synchronization error is below the threshold for a consecutive preset number of frames and the recognition confidence is restored, the safety radial offset is reduced according to the preset step size.
[0053] The safe approach zone corresponding to the pre-contact compliant mode refers to the spatial range within which the distance between the hanging end and the target sub-conductor has entered the preset contact preparation range, while still meeting the safe distance constraint with adjacent sub-conductors. Entering this zone only indicates that the spatial conditions for compliant approach are met; actually switching to the pre-contact compliant mode also requires meeting the synchronization error and stationary feasibility conditions simultaneously.
[0054] The synchronization errors of position, velocity, and attitude are calculated based on the robot's wire-attaching end pose and the pre-aimed stationary point, specifically: ; ; ; ; in, This refers to the position synchronization error; This refers to speed synchronization error; The attitude synchronization error is used to characterize the attitude deviation between the opening direction of the wire-hanging mechanism and the direction of the target sub-guide wire. and The closer to parallel, The smaller; This indicates the location of the robot's cable attachment point; The speed at the end of the robot that attaches the cable is obtained from position difference, inertial measurement data, or filtered estimation. The unit vector representing the opening direction of the wire-hanging mechanism; This is the attitude matrix of the virtual stationary coordinate system.
[0055] S3: Based on the robot's wire-hanging end pose, the split conductor perception data, and the deviation between the current control input of the UAV and their respective predicted values, obtain the observation residuals of the wire-hanging end, the conductor, and the control input, calculate the equivalent disturbance compensation amount, project it onto the virtual stationary coordinate system, and decompose it into components along the conductor direction, the radial approach direction, and the split conductor plane normal component. In some implementations, when the drone-deployed robot approaches the guide wire, it is affected by various factors such as wind disturbance, drone downwash airflow, robot off-center loading, micro-oscillation of the mounting mechanism, and guide wire observation errors. To avoid limiting this embodiment to a single extended state observer, such as... Figure 3 As shown, this embodiment converts the above-mentioned effects into equivalent disturbance compensation amounts. The calculation method is as follows: ; in, To observe the residual at the hanging end, This represents the actual pose observation value of the robot's wire-attaching end. The model prediction value for the robot's wire-attaching end; For traverse observation residuals, The actual observed values of the motion state of the target sub-traverse. The predicted value of the motion state model of the target sub-traverse; To control the input residual, This is the current actual control input for the drone. The desired control input is generated based on the pre-aimed stationary point; ; ; ; ; in, This is the discrete state transition matrix of the hanging end state prediction model; This is the control input matrix for the wire-end state prediction model; This is the discrete state transition matrix of the target sub-traverse motion state prediction model.
[0056] because , and u Since the units of each component are different, before calculating the equivalent disturbance compensation, the residuals are normalized or the unit conversion factor and scale normalization factor are incorporated into the weight matrix. , and Specifically: ; in, , and The weights correspond to the position residual, velocity residual, and attitude residual of the hanging end, respectively. and Adopted and The same scale normalization principle is set, where =diag( , , ), which correspond to the weights of the target sub-traverse reference point position, the target sub-traverse direction, and the residual of the rate of change of motion, respectively; =diag( , ..., ), which correspond to the weights of the residuals of each control input component of the UAV; each weight is determined by calibration experiments, offline identification or online adaptive methods.
[0057] Project the equivalent disturbance compensation onto the virtual stationary coordinate system: ; ; ; in, The component along the conductor direction, For the radial approach direction component, This represents the plane normal component of the split traverse. This represents the transpose of the unit vector representing the direction of the target sub-traverse; This is the equivalent disturbance compensation amount; This represents the transpose of the unit vector of the radial approach direction; This represents the transpose of the unit vector normal to the plane of the split traverse.
[0058] In some implementations, the component along the conductor direction It is not directly used as a mandatory compensation term for approaching the conductor, but rather to correct the position of the pre-aiming stationary point along the line. : ; ; in, The corrected aiming distance along the guide line; The disturbance correction factor along the line, ranging from 0.01 to 0.20, is calibrated based on the UAV flight control response frequency and the target sub-traverse lateral sway frequency. The UAV flight control response frequency characterizes the speed at which the UAV flight control system responds to desired motion commands, while the target sub-traverse lateral sway frequency is estimated based on continuous position observations of the target sub-traverse reference point. The radial approach direction component... Used to suppress radial impact when the hanging end approaches the target sub-conductor; split conductor plane normal component It is used to suppress the hanging end from deviating from the plane of the split conductor where the target sub-conductor is located or from entering the interference area of the adjacent sub-conductor.
[0059] S4: Correct the position of the pre-aiming stationary point using the component along the traverse direction, and generate the synchronization stationary control quantity from the corrected pre-aiming stationary point, each synchronization error, the radial approach direction, and the normal component of the split traverse plane; calculate the UAV's desired motion command based on the synchronization stationary control quantity and the UAV's pose. The formula for calculating the synchronous stationary control quantity is: ; in, The desired acceleration for the robot's wire-attached end. This is the corrected estimate of the acceleration at the pre-contact aiming stationary point; , and These are the position, velocity, and attitude feedback gains, respectively, and can be positive definite matrices or positive gain numbers. The equivalent inertia of the drone and robot in the direction of the hanging end is obtained through calibration tests, or approximately taken as the equivalent mass matrix corresponding to the total mass of the drone and robot. , These are the compensation weights for the radial approach direction component and the plane normal component of the split traverse, respectively.
[0060] In some implementations, position control gain Select a speed control gain of 0.5 to 5.0. The attitude control gain is set to 0.2 to 3.0. The value can be set from 0.1 to 2.0, depending on the specific payload mass of the UAV, the flight control response frequency, and the size of the cable-attaching mechanism.
[0061] To enable those skilled in the art to implement the control process according to the present invention, this embodiment provides a method for converting the desired acceleration at the hanging end into UAV flight control commands, specifically: Acceleration compensation based on the mechanical motion of the robot's wire-attached end relative to the drone body The expected acceleration of the drone is obtained: ; The total thrust is generated based on the expected acceleration of the drone: ; in, The values are calculated from the length of the mounting connection mechanism, the geometric offset of the robot's cable attachment end relative to the drone, and the angular velocity and angular acceleration of the mounting mechanism; if the mounting mechanism remains rigidly fixed before contact, Set to zero or a calibration constant; Let g be the equivalent mass of the UAV and its payload, and g be the acceleration due to gravity. It is a unit vector in the vertical direction.
[0062] Setting the desired yaw angle Then, the expected roll angle and desired pitch angle Calculate as follows: ; ; in, , , These represent the components of the expected acceleration of the UAV in the three coordinate axes.
[0063] The flight controller determines the desired yaw angle based on the settings. Expected roll angle Desired pitch angle and total thrust TL It outputs speed commands for each rotor motor, thereby controlling the drone to carry the robot to perform pre-contact synchronization and parking.
[0064] The system can selectively compensate for the impact of disturbances in different directions on the safety of the stationary line, avoiding the mixing of disturbances along the conductor direction, radial approach disturbances and normal disturbances, thereby reducing the risk of pre-contact control coupling and wire-hanging impact.
[0065] S5: Perform multi-mode switching control based on synchronization error, line-holding feasibility, and robot wire-attaching end contact status data. The multi-mode includes synchronous following mode, pre-contact compliant mode, wire-attaching clamping mode, and abnormal retraction mode. Through multi-mode switching, synchronous following and wire-attaching access control between the wire-attaching end and the swaying target sub-conductor before contact are achieved.
[0066] The degree to which the normal contact force deviates from the safe range is calculated. The degree of deviation is then mapped to the interval between 0 and 1 using the Sigmoid function to obtain the stationary line feasibility evaluation value. Specifically: ; in, For the Sigmoid function, i.e. ; The baseline bias term weighting coefficient for stationary line feasibility; , and These are the weighting coefficients for position synchronization error, velocity synchronization error, and attitude synchronization error, respectively. and These are the weighting coefficients for the radial approach direction disturbance component and the plane normal disturbance component of the split traverse, respectively. This is a weighting coefficient for the degree to which the normal contact force deviates from the safe range. To avoid the direct addition of different dimensions, each error, disturbance component, and contact force deviation is normalized according to its corresponding safety threshold before being input into the Sigmoid function, making them dimensionless quantities.
[0067] The degree to which the normal contact force deviates from the safe range Defined as: ; in, The minimum normal contact force required for reliable wire hanging. The maximum normal contact force to prevent damage to the conductor or hanging mechanism.
[0068] In some implementations, the position synchronization error threshold is set to 0.03m to 0.08m, the speed synchronization error threshold to 0.05m / s to 0.20m / s, the attitude error threshold to 3° to 8°, the line-holding feasibility threshold to 0.75 to 0.90, the number of consecutive stable frames to 5 to 15, and the contact force safety range to 5N to 30N. These thresholds can be calibrated and adjusted according to the conductor type, the structure of the hanging mechanism, the UAV's payload capacity, and the operating wind speed conditions.
[0069] In some implementations, such as Figure 4As shown, multi-mode switching control is performed based on synchronization error, stationary feasibility, and robot wire-attaching end contact status data. Specifically, in coarse approach mode, the UAV flies to the vicinity of the target wire based on positioning, and when the target sub-wire is continuous... When a frame is stably identified and the distance to the drone is less than the first distance threshold, switch to synchronous follow mode; In synchronous follow mode, when position synchronization error, velocity synchronization error, and attitude synchronization error are continuous When the frame meets the threshold requirement and the stationary feasibility evaluation value is greater than the preset evaluation threshold, switch to the pre-contact compliance mode: In the pre-contact compliance mode, reduce the gain components of position control gain and velocity control gain in the radial approach direction, so that the hanging end gradually approaches the target sub-conductor with a lower radial relative velocity; when the hanging end is detected to be in contact with the target sub-conductor, limit the control amount in the radial approach direction according to the normal contact force and its rate of change, so as to reduce the initial contact impact.
[0070] If any of the following criteria for wire insertion is deemed negative: stationary feasibility, synchronization error, normal contact force, or rate of change of normal contact force, the system will not proceed to the next mode. If the deviation is recoverable, the system will maintain or return to the synchronous following mode, and will regenerate the pre-aiming stationary point based on the latest target sub-traverse movement state before continuing synchronous following. If the deviation is unrecoverable or continues to exceed the preset recovery time, the system will switch to the abnormal retreat mode. A recoverable deviation means that the target sub-traverse can still be stably identified, the wire insertion end has not entered the prohibited area of an adjacent sub-traverse, the normal contact force has not exceeded the maximum safe value, and the temporarily excessive synchronization error or disturbance shows a convergence trend within the preset recovery time. If the above conditions are not met, the deviation is deemed unrecoverable.
[0071] When the normal contact force satisfies: And its rate of change is continuous Frame satisfies Switch to wire-attaching and clamping mode: Control the robot's wire-attaching or clamping mechanism to close, and confirm whether the wire attachment or clamping is successful based on the feedback of the normal contact force. If clamping is completed and the drone unloading conditions are met, the mounting connection mechanism can be released or the subsequent maintenance process can begin; In actual mode switching and wire insertion admission judgment, discrete sampling values are uniformly used, and the normal contact force change rate is determined according to the difference in normal contact force at adjacent sampling times: ; in, To preset the threshold for the rate of change of normal contact force, calibration is performed based on the conductor type, hanging mechanism structure, sampling period, and allowable contact impact. The time interval between adjacent sampling moments; In abnormal rollback mode, when the abnormal triggering condition is met... At that time, the drone is controlled to retreat to a safe airspace along a direction away from the plane normal of the split conductor, specifically as follows: ; The direction of the pullback is: ; in, This indicates the number of frames lost during continuous identification of the target sub-trajectory. Indicates the safety threshold for disturbance. Indicates the attitude angle of the drone. Indicates the safe threshold for the drone's attitude angle. Indicates the no-entry zone for adjacent sub-traverses. Indicates the location of the drone. Indicates the speed of retracement. This indicates the pullback speed coefficient.
[0072] By employing the above methods, when identification is lost, disturbance exceeds limits, normal contact force exceeds limits, attitude is abnormal, or the hanging end enters the restricted area of an adjacent sub-conductor, the UAV will retreat along the direction away from the plane normal of the split conductor, thereby improving the safety of pre-contact line-holding operations.
[0073] In some implementations, such as Figure 5 As shown, after the UAV retreats to a safe airspace, if the target sub-traverse line is re-identified stably for a preset number of consecutive frames, the abnormal triggering amount returns to within the corresponding safe threshold, and the safety distance condition between the hanging end and the adjacent sub-traverse line is met again, then the abnormal state is cleared and the UAV returns to the coarse approach mode, and the target identification, pre-aiming stationary point generation, synchronous following, and hanging access judgment are re-executed; if the re-entry conditions cannot be met within the preset waiting time, then the UAV will maintain a safe hover or end the current hanging task.
[0074] Example 1 To verify the effectiveness of the method of the present invention, a simulation environment was established for the standby of the UAV, robot, and split conductor before contact. The simulation object is a four-split conductor, the target sub-conductor is one of the outer sub-conductors in the four-split conductor, the spacing between the sub-conductors is set to 450mm, the lateral swaying frequency of the target sub-conductor is set to 0.2Hz to 1.0Hz, the external disturbance wind speed is set to 3m / s to 8m / s, and the robot mass is set to 5kg to 12kg.
[0075] The simulation parameters described above are used to construct different target sub-conductor motions, external disturbances, and load conditions, and are not used as independent control inputs for the pre-contact synchronous wire-hanging control law. To clarify the relationship between the lateral sway frequency of the target sub-conductor and the time-varying trajectory of the target sub-conductor reference point, this embodiment uses the lateral sway frequency of the target sub-conductor... f LDefined as the dominant frequency when the target sub-traverse reference point undergoes periodic motion in a transverse direction perpendicular to the target sub-traverse direction. Let the equilibrium position of the target sub-traverse reference point be... p L0 The lateral sway amplitude is A L The initial phase is φ L The unit vector of the lateral sway direction is e L ,and e L Unit vector of the initial direction of the target sub-traverse t L0 Vertical; the time-varying trajectory of the target sub-traverse reference point refers to the position of the target sub-traverse reference point changing with time. t The trajectory formed by the change is expressed as: ; in, p L sim ( t ( ) represents the simulation time. t The position of the target sub-traverse reference point. From the above time-varying motion trajectory, the rate of change of the target sub-traverse reference point's position can be obtained as: ; thus, f L This directly determines the position of the reference point of the simulated target sub-traverse and the rate of change of its position over time. During the simulation, [the following is used:] p L sim ( t Generate the corresponding split traverse sensing data, and obtain the target sub-traverse reference point position through continuous observation and state estimation in step S2. p L ( t ) and the motion state of the target sub-traverse z L ( t The formula for calculating the pre-aiming stationary point in step S2 is as follows: p L ( t As the reference point for the target sub-traverse, the lateral sway frequency of the target sub-traverse is determined by... p L ( t )and z L ( t The time variation affects the pre-aiming stationary point, but is not used as an additional independent input for calculating the pre-aiming stationary point or the synchronous stationary control quantity. The equivalent disturbance compensation quantity in step S3 is still determined by the observation residual at the hanging end and the observation residual of the traverse.z L - L And the calculation of control input residuals, where the traverse observation residuals are used to characterize the deviation between the actual motion state and the model's predicted state, and not to... f L Add the residual or equivalent perturbation formula separately.
[0076] External wind disturbance, as a simulated environmental disturbance, acts on the UAV, the hanging end, and the target sub-traverse. Its effect manifests as a deviation between the actual observed state and the model's predicted state, and participates in equivalent disturbance compensation through the corresponding residuals in step S3. The robot mass is used to determine the UAV-robot equivalent inertia in step S4. M e and the equivalent mass of the drone and its payload m u Therefore, wind speed and robot mass are not directly added to the control law as additional independent control variables.
[0077] The system without pre-contact synchronous line-holding access control was used as the control group, while the system using the method of this invention was used as the experimental group. Comparative indicators included the maximum relative position error between the hanging end and the target sub-conductor, the maximum relative velocity error, the peak impact force at the moment of contact, and the line-holding success rate.
[0078] Under simulation conditions of 5 m / s wind speed, 0.5 Hz lateral sway frequency of the target sub-guide wire, and 8 kg robot mass, 50 consecutive line-stopping simulations were performed. The results are shown in Table 1. Table 1 Comparison of Synchronous Stationary Line Performance Before Contact in Operating Condition 1
[0079] Under another simulation condition, with a wind speed of 8 m / s, a target sub-guide lateral sway frequency of 0.8 Hz, and a robot mass of 10 kg, 50 consecutive stationary simulations were performed. The results are as follows: Table 2 Comparison of Synchronous Stationary Line Performance Before Contact in Operating Condition 2
[0080] The above results show that the present invention, through virtual stationary coordinate system, pre-aiming stationary point before contact, equivalent disturbance direction decomposition and multi-mode access control, can effectively reduce the relative position error and relative velocity error between the hanging end and the swaying conductor, reduce the peak impact force at the moment of contact, and improve the stationary success rate.
[0081] In another embodiment of the present invention, a pre-contact synchronous wire-attaching control system for a UAV-mounted robot oriented towards split conductors is proposed, realizing the pre-contact synchronous wire-attaching control method for a UAV-mounted robot oriented towards split conductors as described above, comprising: The multi-source sensing and state acquisition module is used to acquire the pose of the UAV, the pose of the robot's wire-attached end, the sensing data of the split wire, and the contact state data of the robot's wire-attached end. The target recognition and synchronization error calculation module is used to identify the target sub-traverse and its adjacent sub-traverse from the split traverse sensing data, determine the reference point and direction of the target sub-traverse and the reference point of the adjacent sub-traverse, and construct a virtual stationary coordinate system; estimate the motion trend of the target sub-traverse, generate the pre-aiming stationary point in the virtual stationary coordinate system; and calculate the position, velocity and attitude synchronization error based on the robot's hanging end pose and the pre-aiming stationary point. The equivalent disturbance compensation and decomposition module is used to obtain the observation residuals of the hanging end, the wire and the control input based on the robot hanging end pose, the split wire perception data and the deviation between the current control input of the UAV and their respective predicted values, calculate the equivalent disturbance compensation amount, project it onto the virtual stationary coordinate system, and decompose it into components along the wire direction, the radial approach direction and the split wire plane normal component. The synchronization stationary control quantity generation module is used to correct the position of the pre-aiming stationary point using the component along the traverse direction. The synchronization stationary control quantity is generated from the corrected pre-aiming stationary point, each synchronization error, the radial approach direction, and the normal component of the split traverse plane. The desired motion command of the UAV is calculated based on the synchronization stationary control quantity and the UAV pose. The line-standing decision and mode management module is used to perform multi-mode switching control based on synchronization error, line-standing feasibility and robot wire-attaching end contact status data. The multi-mode includes synchronous following mode, pre-contact compliant mode, wire-attaching clamping mode and abnormal withdrawal mode. Through multi-mode switching, the synchronous following and wire-attaching access control between the wire-attaching end and the swaying target sub-lead before contact is realized.
[0082] In another embodiment of the present invention, an electronic device is proposed, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements a pre-contact synchronous wire attachment control method for a UAV-mounted robot oriented to split conductors as described above.
[0083] In another embodiment of the present invention, a computer-readable storage medium is provided storing a computer program that causes a computer to execute a pre-contact synchronous wire-attaching control method for a UAV-mounted robot oriented to split conductors, as described above.
[0084] In the embodiments disclosed in this application, a computer storage medium may be a tangible medium that may contain or store programs for use by or in conjunction with an instruction execution system, apparatus, or device. The computer storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of computer storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0085] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications, equivalent substitutions, and improvements made by those skilled in the art to the above embodiments without departing from the scope of the technical solution of the present invention, based on the technical essence of the present invention, shall still fall within the protection scope of the technical solution of the present invention.
Claims
1. A pre-contact synchronous wire-attaching control method for an unmanned aerial vehicle (UAV) mounted robot oriented towards split conductors, characterized in that, Includes the following steps: S1: Acquire the drone pose, the robot wire-attaching end pose, the split wire sensing data, and the robot wire-attaching end contact status data; S2: Identify the target sub-traverse and its adjacent sub-traverses using the split traverse sensing data, determine the reference point and direction of the target sub-traverse and the reference point of the adjacent sub-traverse, and construct a virtual stationary coordinate system; estimate the motion trend of the target sub-traverse and generate the pre-aiming stationary point in the virtual stationary coordinate system; calculate the position, velocity and attitude synchronization error based on the robot's hanging end pose and the pre-aiming stationary point; S3: Based on the robot's wire-hanging end pose, the split conductor perception data, and the deviation between the current control input of the UAV and their respective predicted values, obtain the observation residuals of the wire-hanging end, the conductor, and the control input, calculate the equivalent disturbance compensation amount, project it onto the virtual stationary coordinate system, and decompose it into components along the conductor direction, the radial approach direction, and the split conductor plane normal component. S4: Correct the position of the pre-aiming stationary point using the component along the traverse direction, and generate the synchronization stationary control quantity from the corrected pre-aiming stationary point, each synchronization error, the radial approach direction, and the normal component of the split traverse plane; calculate the UAV's desired motion command based on the synchronization stationary control quantity and the UAV's pose. S5: Perform multi-mode switching control based on synchronization error, line-holding feasibility, and robot wire-attaching end contact status data. The multi-mode includes synchronous following mode, pre-contact compliant mode, wire-attaching clamping mode, and abnormal retraction mode. Through multi-mode switching, synchronous following and wire-attaching access control between the wire-attaching end and the swaying target sub-conductor before contact are achieved. 2.The contact pre-sync line hanging control method of the UAV-mounted robot facing the split conductor according to claim 1, wherein: In step S2, the split traverse sensing data includes split traverse image features, laser ranging point cloud, and inertial measurement data; the inertial measurement data is used to compensate for motion distortion of the laser ranging point cloud, and each sampling point is transformed to a coordinate system of a unified reference time according to its timestamp and the UAV attitude and position changes; based on the compensated laser ranging point cloud and split traverse image features, the spatial distribution relationship of the target sub-traverse and its adjacent sub-traverses in the split traverse is identified, and the reference point of the target sub-traverse, the direction of the target sub-traverse, and the reference point of the adjacent sub-traverse are determined; A virtual stationary coordinate system is constructed based on the direction of the target sub-traverse and the reference points of adjacent sub-traverses, and is updated as the target sub-traverse moves. The specific process is as follows: Obtain two fitting points on the center line of the target sub-traverse. and and the target sub-traverse reference point ; Obtain reference points for adjacent sub-traverses and By fitting points and Normalize the direction of the connecting line to obtain the unit vector of the target sub-traverse direction. ; Through the target sub-traverse reference point Pointing to the reference points of the adjacent sub-traverses on both sides and The cross product of two spatial vectors, followed by normalization, yields the unit vector normal to the plane of the split traverse. ; through the split traverse plane normal unit vector Unit vector of the target sub-traverse direction After performing the cross product and normalizing, we obtain the unit vector of the radial approach direction. .
3. The contact pre-sync line hanging control method of the unmanned aerial vehicle mounted robot facing split conductors according to claim 1, characterized in that: In step S2, generating the pre-aiming stationary point in the virtual stationary coordinate system before contact specifically involves: ; wherein, is a pre-look waypoint position; is a target sub-conductor reference point position; is a pre-look distance along the conductor direction; is a target sub-conductor direction unit vector; is a radial approach direction unit vector; is a pre-contact safety radial offset; is a split conductor plane normal unit vector; is a normal offset to avoid adjacent sub-conductors; When the conductor sway intensity increases or the recognition confidence decreases, the safety radial offset or normal offset is adaptively increased; when the position synchronization error is continuously lower than the preset threshold, the safety radial offset is gradually reduced so that the hanging end enters the pre-contact compliant region.
4. The pre-contact synchronous wire-attaching control method for a drone-mounted robot oriented towards split conductors according to claim 1, characterized in that: In step S3, the equivalent disturbance compensation is calculated, projected onto the virtual stationary coordinate system, and decomposed into components along the traverse direction, the radial approach direction, and the normal component of the split traverse plane. Specifically: ; in, This is the equivalent disturbance compensation amount; To observe the residual at the hanging end, This represents the actual pose observation value of the robot's wire-attaching end. The model prediction value for the robot's wire-attaching end; For traverse observation residuals, The actual observed values of the motion state of the target sub-traverse. The predicted value of the motion state model of the target sub-traverse; To control the input residual, This is the current actual control input for the drone. The desired control input is generated based on the pre-aimed stationary point; , , This is the weight matrix; The component along the conductor direction is used to correct the position of the pre-aiming stationary point along the conductor: the component along the conductor direction is multiplied by a preset disturbance correction coefficient along the conductor direction to obtain the correction amount of the pre-aiming distance along the conductor direction, and the pre-aiming distance along the conductor direction is updated using this correction amount to obtain the corrected pre-aiming stationary point; the radial approach direction component is used to suppress the radial impact when the hanging end approaches the target sub-conductor; the split conductor plane normal component is used to suppress the hanging end from deviating from the split conductor plane where the target sub-conductor is located or entering the interference region of adjacent sub-conductors.
5. The pre-contact synchronous wire-attaching control method for a drone-mounted robot oriented towards split conductors according to claim 1, characterized in that: In step S4, the generation of the synchronization control quantity from the corrected pre-aiming stationary point, each synchronization error, the radial approach direction, and the normal component of the split traverse plane specifically involves: ; in, Desired acceleration for the robot's wire-attached end; This is the corrected estimate of the acceleration at the pre-contact aiming stationary point; , , These are the position, velocity, and attitude feedback gains, respectively. This refers to the position synchronization error; For speed synchronization error; This refers to the attitude synchronization error; The equivalent inertia of the drone and robot in the direction of the hanging end; The radial approach direction component; It is the unit vector in the radial approach direction; For the plane normal component of the split traverse; The unit vector is the normal vector of the split traverse plane; , These are the compensation weights for the radial approach direction component and the plane normal component of the split traverse, respectively.
6. The pre-contact synchronous wire-attaching control method for a drone-mounted robot oriented towards split conductors according to claim 1, characterized in that: In step S5, the feasibility of stationary operation determined based on the normal contact force in the robot's wire-attaching end contact state data specifically includes: The degree to which the normal contact force deviates from the safe range is calculated. The degree of deviation is mapped to the interval of 0 to 1 using the Sigmoid function to obtain the feasibility evaluation value for the suspension line. The degree of deviation of the normal contact force from the safe range is defined as follows: when the normal contact force is less than the minimum contact force required for reliable suspension, the deviation is the difference between the minimum contact force and the normal contact force; when the normal contact force is greater than the maximum contact force to prevent damage to the conductor or suspension mechanism, the deviation is the difference between the normal contact force and the maximum contact force; when the normal contact force is within the safe range between the minimum and maximum contact forces required for reliable suspension, the deviation is 0.
7. The pre-contact synchronous wire-attaching control method for a drone-mounted robot oriented towards split conductors according to claim 1, characterized in that: In step S5, the multi-mode switching control based on synchronization error, line-standing feasibility, and robot wire-attaching end contact status data is specifically as follows: In coarse approach mode, the UAV flies to the vicinity of the target wire according to the positioning. When the target sub-wire is stably identified for a first preset number of consecutive frames and the distance to the UAV is less than a first distance threshold, it switches to synchronous following mode; In synchronous following mode, when the position synchronization error, velocity synchronization error, and attitude synchronization error meet the threshold requirements for a second preset number of consecutive frames, and the line-standing feasibility evaluation value is greater than a preset evaluation threshold, it switches to pre-contact compliant mode. In the pre-contact compliance mode, the gain components of position control gain and speed control gain in the radial approach direction are reduced. When the normal contact force and its rate of change meet the safety conditions for three consecutive preset frames, the system switches to the wire clamping mode. In the wire clamping mode, the wire clamping mechanism is controlled to close, and the wire clamping status is confirmed based on the contact force feedback. In abnormal retreat mode, when the abnormal triggering conditions are met, the UAV is controlled to retreat to a safe airspace in a direction away from the plane normal of the split conductor.
8. A pre-contact synchronous wire-attaching control system for a UAV-mounted robot oriented towards split conductors, used to implement the pre-contact synchronous wire-attaching control method for a UAV-mounted robot oriented towards split conductors as described in any one of claims 1 to 7, characterized in that, include: The multi-source sensing and state acquisition module is used to acquire the pose of the UAV, the pose of the robot's wire-attached end, the sensing data of the split wire, and the contact state data of the robot's wire-attached end. The target recognition and synchronization error calculation module is used to identify the target sub-traverse and its adjacent sub-traverse from the split traverse sensing data, determine the reference point and direction of the target sub-traverse and the reference point of the adjacent sub-traverse, and construct a virtual stationary coordinate system; estimate the motion trend of the target sub-traverse, generate the pre-aiming stationary point in the virtual stationary coordinate system; and calculate the position, velocity and attitude synchronization error based on the robot's hanging end pose and the pre-aiming stationary point. The equivalent disturbance compensation and decomposition module is used to obtain the observation residuals of the hanging end, the wire and the control input based on the robot hanging end pose, the split wire perception data and the deviation between the current control input of the UAV and their respective predicted values, calculate the equivalent disturbance compensation amount, project it onto the virtual stationary coordinate system, and decompose it into components along the wire direction, the radial approach direction and the split wire plane normal component. The synchronization stationary control quantity generation module is used to correct the position of the pre-aiming stationary point using the component along the traverse direction. The synchronization stationary control quantity is generated from the corrected pre-aiming stationary point, each synchronization error, the radial approach direction, and the normal component of the split traverse plane. The desired motion command of the UAV is calculated based on the synchronization stationary control quantity and the UAV pose. The line-standing decision and mode management module is used to perform multi-mode switching control based on synchronization error, line-standing feasibility and robot wire-attaching end contact status data. The multi-mode includes synchronous following mode, pre-contact compliant mode, wire-attaching clamping mode and abnormal withdrawal mode. Through multi-mode switching, the synchronous following and wire-attaching access control between the wire-attaching end and the swaying target sub-lead before contact is realized.
9. An electronic device, characterized in that, include: The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a pre-contact synchronous wire-attaching control method for a drone-mounted robot oriented towards split conductors, as described in any one of claims 1 to 7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program causes the computer to execute a pre-contact synchronous wire-attaching control method for a drone-mounted robot oriented towards split conductors, as described in any one of claims 1 to 7.
