Four-legged robot collaborative handling system and method for power infrastructure special-shaped materials

Abstract

The application relates to a four-legged robot collaborative carrying system for power infrastructure special-shaped materials, which comprises a cloud collaborative scheduling module and a plurality of four-legged carrying robot units, the cloud collaborative scheduling module collaboratively controls and schedules the plurality of four-legged carrying robot units, the cloud collaborative scheduling module comprises a dynamic collaborative control module, a terrain posture correction module and a fault tolerance protection module, the four-legged carrying robot unit comprises an onboard sensing and positioning module and a self-adaptive clamping execution module, the cloud collaborative scheduling module is used for receiving a power infrastructure carrying task, inputting special-shaped material parameters, construction terrain data and carrying start and end coordinates, and completing material grading and robot grouping scheduling, and the structure provided by the embodiment has high terrain adaptability, can stably work on a mountain slope with a slope of less than or equal to 25 degrees, muddy road surfaces and concave-convex gravel road surfaces, and meets complex terrain working conditions of field power construction.

Description

Technical Field

[0001] This application relates to the field of quadruped robot technology, and in particular to a quadruped robot collaborative handling system and method for irregularly shaped materials in power infrastructure. Background Technology

[0002] In power infrastructure projects such as transmission line erection, substation renovation and expansion, and power grid construction, there are numerous non-standard and irregularly shaped material handling operations. Typical materials include curved power pipe bends, irregularly shaped tower connecting hardware, eccentric cable reels, irregular insulation accessories, bent steel structure parts, and irregularly shaped counterweights. These materials lack a uniform and regular shape, have a large center of gravity offset, and lack flat clamping surfaces. Moreover, power construction sites are often located in remote mountainous areas, hills, and muddy gravel roads, with large terrain undulations, narrow passages, and inaccessible large hoisting equipment.

[0003] Currently, the industry mainly relies on manual carrying or simple forklifts for handling irregularly shaped materials in power plants, which has many technical drawbacks: 1. High safety risks: Manual handling in complex terrain is prone to slipping, bumping, and personnel injuries. Safety accidents related to power infrastructure handling account for over 32% annually; 2. Low work efficiency: A single 3-person team can only handle irregularly shaped materials weighing up to 20kg at a time, and long-distance round trips are time-consuming; 3. Poor adaptability: Traditional forklifts and tracked handling equipment are only suitable for regular materials and cannot handle irregularly shaped objects. 4. Uneven load distribution leads to poor stability. Existing multi-robot handling solutions lack specific algorithms for adapting to irregularly shaped materials and only use a uniform load distribution mode. For eccentric irregularly shaped materials, problems such as single-machine overload, machine tipping, and material misalignment are prone to occur. The force deviation of multiple machines generally exceeds 12%, resulting in a high handling failure rate. 5. Weak terrain adaptability. Conventional quadruped robot collaborative solutions have not made specific posture optimizations for power infrastructure slopes within 25° and uneven road surfaces, resulting in large errors in synchronous operations on complex terrain.

[0004] In existing publicly available technologies, multi-quadruped robot collaborative handling solutions are mostly applied to standardized industrial materials and indoor delivery scenarios. They lack specific technical solutions for adaptive 3D contour matching of irregularly shaped power materials, dynamic balancing of eccentric loads, collaborative posture correction on sloping terrain, and real-time fault-tolerant reconfiguration. These solutions cannot meet the stable, accurate, and safe collaborative handling requirements for irregularly shaped materials under complex working conditions in power infrastructure projects. Based on the shortcomings of existing technologies, this paper proposes a quadruped robot collaborative handling system and method for irregularly shaped materials in power infrastructure projects. Summary of the Invention

[0005] This application provides a quadruped robot collaborative handling system and method for irregularly shaped materials in power infrastructure construction, in order to solve the problems mentioned above.

[0006] This application provides a quadruped robot collaborative handling system for irregularly shaped materials in power infrastructure construction, including:

[0007] A cloud-based collaborative scheduling module and multiple quadruped transport robot units; the cloud-based collaborative scheduling module coordinates and controls the scheduling of the multiple quadruped transport robot units.

[0008] The cloud-based collaborative scheduling module includes a dynamic collaborative control module, a terrain attitude correction module, and a fault tolerance and protection module.

[0009] The quadruped transport robot unit includes an onboard sensing and positioning module and an adaptive gripping and execution module;

[0010] The cloud-based collaborative scheduling module is used to receive power infrastructure transportation tasks, input parameters of irregularly shaped materials, construction terrain data, and transportation start and end coordinates, and complete material classification and robot grouping scheduling.

[0011] The dynamic collaborative control module has a built-in load balancing algorithm and synchronous motion control model, which are used for load distribution, speed synchronization and displacement matching among multiple robots.

[0012] The terrain posture correction module is used to correct the robot's quadruped gait and body balance posture in real time based on the slope and unevenness data of the construction site.

[0013] The fault-tolerant protection module is used to monitor the robot's communication, power, and force status in real time. When a single robot fails, it automatically reconstructs the collaborative group to ensure that the handling task is not interrupted.

[0014] The airborne perception and positioning module integrates a lidar, a binocular depth camera, a six-axis inertial sensor, and a UWB high-precision positioning unit, with a positioning accuracy of ±2cm and a frame rate of 30FPS. It is used to collect material three-dimensional contours, terrain undulation data, robot pose parameters, and body force data in real time.

[0015] The adaptive clamping execution module is located on the top of the quadruped handling robot's single body and includes a clamping manipulator, a plate clamp, a flexible anti-slip clamping pad, a pressure sensor array, and an angle fine-tuning servo motor. It can adaptively adjust the clamping distance and fitting angle according to the curved surface, edges, and irregular contours of the irregularly shaped material.

[0016] Preferably, the pressure sensor array of the adaptive clamping execution module is distributed in a matrix on the clamping surface, with a single-point pressure detection range of 0-500N and a detection accuracy of ±0.8N. It can collect the contact pressure distribution between irregular materials and the clamping surface in real time, so as to achieve uniform force application and fit of irregular contact surfaces.

[0017] Preferably, the quadruped transport robot is a high-load off-road quadruped robot with a maximum single-body load of 80kg, an adjustable walking stride of 0-35cm, and adaptive adjustment of body pitch angle ±30° and roll angle ±25°.

[0018] Preferably, the end of the gripping manipulator is connected to a plate clamp, which consists of two plates that can move relative to and opposite to each other. One of the plates is driven to move by an electric telescopic rod. Flexible anti-slip gripping pads are evenly connected to the opposing gripping surfaces of the two plates by angle fine-tuning servos. Pressure sensors are embedded inside the flexible anti-slip gripping pads, and all the pressure sensors form a pressure sensing array.

[0019] A method for collaborative handling of irregularly shaped materials in power infrastructure using a quadruped robot includes the following steps:

[0020] S1. Construction scenario and non-standard material parameter acquisition and modeling: The handling task is entered through the cloud scheduling module, and the airborne perception and positioning module scans the non-standard electrical materials to be handled, constructs a three-dimensional point cloud model, and extracts the material length, width, height, center of gravity offset, surface curvature, and weight parameters. The materials are divided into four categories: symmetrical center of gravity, unilateral eccentricity, curved surface non-standard, and irregular edge and corner type.

[0021] S2. Robot Adaptive Grouping and Position Layout: The number of robots is matched according to the size and weight of the material. For materials weighing 5-30kg, two robots work together; for materials weighing 30-80kg, three robots work together; and for materials weighing 80-150kg, four robots work together. The support points of each robot are automatically planned based on the coordinates of the material's center of gravity. For eccentric materials, an asymmetrical position layout is used to correct the support offset error.

[0022] S3. Dynamic Load Balancing Quantitative Allocation: Calculates the real-time load threshold of each robot using a preset load balancing model, and dynamically allocates the load based on feedback data from the robot's pressure sensors, controlling the force deviation among multiple robots to ≤3.5%.

[0023] S4. Terrain Adaptation Posture Pre-adjustment: Collect data on terrain slope, unevenness, and obstacle distribution at the construction site. Within the terrain slope range of 0°-25°, adjust the robot's roll and pitch angles according to the slope threshold and simultaneously optimize the four-legged support gait.

[0024] S5. Multi-machine synchronous clamping and lifting operation: Each robot moves and aligns according to the planned points. The adaptive clamping module fits the material contour to complete flexible clamping. The pressure sensor array detects the fit. When the pressure at a single point is less than 30N, the clamping angle is automatically fine-tuned. The lifting speed is uniformly set to 5cm / s to ensure lifting synchronization.

[0025] S6. Real-time closed-loop correction during handling: Real-time acquisition of robot pose, travel speed and material offset data during handling. Parameter correction is completed every 10ms. Secondary posture correction is triggered when the material offset exceeds 8mm.

[0026] S7. Fault-tolerant collaborative reconfiguration: Real-time monitoring of robot operating status. When a single-machine communication interruption, power failure, or overload failure is detected, the remaining robot load is redistributed and the path is reconfigured within 0.1 seconds to continue completing the handling task.

[0027] S8. Fixed-point material drop and task reset: After the robot reaches the target point, it drops the material at a constant speed, releases the clamping and reset structure, and returns to the standby point to complete a single collaborative handling task.

[0028] Preferably, the quantitative calculation formula for the dynamic load balancing model in S3 is:

[0029]

[0030] in, To ensure the i-th robot bears the pressure in real time, The weight is determined by the unilateral partition of the material. Let be the straight-line distance from the support point of the i-th robot to the center of gravity of the material. It is the sum of the distances from all support points to the center of gravity. The pressure value is the slope compensation value; for every increase in slope... , Increment by 2.5N to achieve dynamic compensation of slope and terrain load.

[0031] Preferably, the synchronous motion error correction formula in S6 is:

[0032]

[0033]

[0034] in, This is the speed deviation value. Set a standard travel speed for the system. For the robot's real-time movement speed, The gait correction angle is k, the correction coefficient is 0.85, and t is the correction period of 10ms, ensuring that the synchronization error of multiple machines is ≤0.02s.

[0035] Preferably, the formula for correcting the offset of the eccentric material point in S2 is:

[0036]

[0037] in, Correct the coordinates of the robot support points. The initial planning coordinates are: e is the material's center of gravity eccentricity, L is the material's support span, and W is the material's overall width. When the material's center of gravity eccentricity e is ≥ 15mm, point correction is forcibly triggered.

[0038] The technical solutions provided in this application have the following advantages compared with the prior art:

[0039] The structure provided in this application embodiment has a wide range of applications: it can be adapted to all four types of irregular materials in power infrastructure, including curved surfaces, eccentricities, sharp corners, and irregular shapes, with a size range of 10cm-120cm and a weight range of 5kg-150kg, covering more than 95% of non-standard handling scenarios in power infrastructure.

[0040] High operational accuracy: multi-machine force deviation ≤ ±3.5%, material handling offset ≤ 8mm, and travel synchronization error ≤ 0.02s, far superior to existing conventional collaborative handling technologies;

[0041] Strong terrain adaptability: It can operate stably on mountain slopes with a gradient of ≤25°, muddy roads, and uneven gravel roads, meeting the complex terrain conditions of field power construction;

[0042] Significantly improved efficiency and safety: No manual handling is required, eliminating personnel safety accidents; handling efficiency is 180% higher than traditional manual handling and 65% higher than conventional robot collaborative solutions.

[0043] High fault tolerance: It has the ability to reconstruct single-machine failures in real time, and the success rate of operation in complex field conditions has been increased from the traditional 68% to 99.2%, ensuring the continuity of construction. Attached Figure Description

[0044] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0045] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram of the principle of the present invention;

[0047] Figure 2 This is a flowchart of the method steps of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] Various embodiments of this application may exist in the form of a range. It should be understood that the description in the form of a range is merely for convenience and brevity and should not be construed as a rigid limitation on the scope of this application. Therefore, it should be considered that the range description has specifically disclosed all possible sub-ranges and single numerical values ​​within that range. For example, it should be considered that the range description from 1 to 6 has specifically disclosed sub-ranges, such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6, etc., and single numbers within the range, such as 1, 2, 3, 4, 5, and 6, regardless of the range. In addition, whenever a numerical range is indicated in this application, it means including any referenced number (fraction or integer) within the indicated range. Unless otherwise specified, all raw materials, reagents, instruments, and equipment used in this application can be purchased commercially or prepared using existing equipment.

[0050] In this application, unless otherwise stated, directional terms such as "upper" and "lower" specifically refer to the drawing directions in the accompanying drawings. Furthermore, in this application, the terms "comprising," "including," etc., mean "including but not limited to." In this application, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. In this application, "and / or" describes the relationship between related objects, indicating that three relationships may exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. A and B can be singular or plural. In this application, "at least one" means one or more, and "more than one" means two or more. "At least one," "at least one of the following," or similar expressions refer to any combination of these items, including any combination of a single item or a plural item. For example, "at least one of a, b, or c" or "at least one of a, b, and c" can both mean: a, b, c, ab (i.e., a and b), ac, bc, or abc, where a, b, and c can be a single or multiple.

[0051] like Figures 1-2As shown, this application provides a quadruped robot collaborative handling system for irregularly shaped materials in power infrastructure, comprising:

[0052] A cloud-based collaborative scheduling module and multiple quadruped transport robot units; the cloud-based collaborative scheduling module coordinates and controls the scheduling of the multiple quadruped transport robot units.

[0053] The cloud-based collaborative scheduling module includes a dynamic collaborative control module, a terrain attitude correction module, and a fault tolerance and protection module.

[0054] The quadruped transport robot unit includes an onboard sensing and positioning module and an adaptive gripping and execution module;

[0055] The cloud-based collaborative scheduling module is used to receive power infrastructure transportation tasks, input parameters of irregularly shaped materials, construction terrain data, and transportation start and end coordinates, and complete material classification and robot grouping scheduling.

[0056] The dynamic collaborative control module has a built-in load balancing algorithm and synchronous motion control model, which are used for load distribution, speed synchronization and displacement matching among multiple robots.

[0057] The terrain posture correction module is used to correct the robot's quadruped gait and body balance posture in real time based on the slope and unevenness data of the construction site.

[0058] The fault-tolerant protection module is used to monitor the robot's communication, power, and force status in real time. When a single robot fails, it automatically reconstructs the collaborative group to ensure that the handling task is not interrupted.

[0059] The airborne perception and positioning module integrates a lidar, a binocular depth camera, a six-axis inertial sensor, and a UWB high-precision positioning unit, with a positioning accuracy of ±2cm and a frame rate of 30FPS. It is used to collect material three-dimensional contours, terrain undulation data, robot pose parameters, and body force data in real time.

[0060] The adaptive clamping execution module is located on the top of the quadruped handling robot's single body and includes a clamping manipulator, a plate clamp, a flexible anti-slip clamping pad, a pressure sensor array, and an angle fine-tuning servo motor. It can adaptively adjust the clamping distance and fitting angle according to the curved surface, edges, and irregular contours of the irregularly shaped material.

[0061] The pressure sensor array of the adaptive clamping execution module is distributed in a matrix on the clamping surface. The single-point pressure detection range is 0-500N, and the detection accuracy is ±0.8N. It can collect the contact pressure distribution between irregular materials and the clamping surface in real time, so as to achieve uniform force and fit on irregular contact surfaces.

[0062] The quadruped transport robot is a high-load off-road quadruped robot with a maximum single-body load of 80kg, an adjustable walking stride of 0-35cm, and adaptive adjustment of body pitch angle ±30° and roll angle ±25°.

[0063] The end of the gripping manipulator is connected to a plate clamp, which consists of two plates that can move relative to each other and in opposite directions. One of the plates is driven to move by an electric telescopic rod. Flexible anti-slip gripping pads are evenly connected to the opposing gripping surfaces of the two plates by angle fine-tuning servos. Pressure sensors are embedded inside the flexible anti-slip gripping pads, and all the pressure sensors form a pressure sensing array.

[0064] A method for collaborative handling of irregularly shaped materials in power infrastructure using a quadruped robot includes the following steps:

[0065] S1. Construction scenario and non-standard material parameter acquisition and modeling: The handling task is entered through the cloud scheduling module, and the airborne perception and positioning module scans the non-standard electrical materials to be handled, constructs a three-dimensional point cloud model, and extracts the material length, width, height, center of gravity offset, surface curvature, and weight parameters. The materials are divided into four categories: symmetrical center of gravity, unilateral eccentricity, curved surface non-standard, and irregular edge and corner type.

[0066] S2. Robot Adaptive Grouping and Position Layout: The number of robots is matched according to the size and weight of the material. For materials weighing 5-30kg, two robots work together; for materials weighing 30-80kg, three robots work together; and for materials weighing 80-150kg, four robots work together. The support points of each robot are automatically planned based on the coordinates of the material's center of gravity. For eccentric materials, an asymmetrical position layout is used to correct the support offset error.

[0067] S3. Dynamic Load Balancing Quantitative Allocation: Calculates the real-time load threshold of each robot using a preset load balancing model, and dynamically allocates the load based on feedback data from the robot's pressure sensors, controlling the force deviation among multiple robots to ≤3.5%.

[0068] S4. Terrain Adaptation Posture Pre-adjustment: Collect data on terrain slope, unevenness, and obstacle distribution at the construction site. Within the terrain slope range of 0°-25°, adjust the robot's roll and pitch angles according to the slope threshold and simultaneously optimize the four-legged support gait.

[0069] S5. Multi-machine synchronous clamping and lifting operation: Each robot moves and aligns according to the planned points. The adaptive clamping module fits the material contour to complete flexible clamping. The pressure sensor array detects the fit. When the pressure at a single point is less than 30N, the clamping angle is automatically fine-tuned. The lifting speed is uniformly set to 5cm / s to ensure lifting synchronization.

[0070] S6. Real-time closed-loop correction during handling: Real-time acquisition of robot pose, travel speed and material offset data during handling. Parameter correction is completed every 10ms. Secondary posture correction is triggered when the material offset exceeds 8mm.

[0071] S7. Fault-tolerant collaborative reconfiguration: Real-time monitoring of robot operating status. When a single-machine communication interruption, power failure, or overload failure is detected, the remaining robot load is redistributed and the path is reconfigured within 0.1 seconds to continue completing the handling task.

[0072] S8. Fixed-point material drop and task reset: After the robot reaches the target point, it drops the material at a constant speed, releases the clamping and reset structure, and returns to the standby point to complete a single collaborative handling task.

[0073] The quantitative calculation formula for the dynamic load balancing model in S3 is as follows:

[0074]

[0075] in, To ensure the i-th robot bears the pressure in real time, The weight is determined by the unilateral partition of the material. Let be the straight-line distance from the support point of the i-th robot to the center of gravity of the material. It is the sum of the distances from all support points to the center of gravity. This is the pressure value to compensate for terrain slope; for every 5° increase in slope, Increment by 2.5N to achieve dynamic compensation of slope and terrain load.

[0076] The synchronous motion error correction formula in S6 is as follows:

[0077]

[0078]

[0079] in, This is the speed deviation value. Set a standard travel speed for the system. For the robot's real-time movement speed, The gait correction angle is k, the correction coefficient is 0.85, and t is the correction period of 10ms, ensuring that the synchronization error of multiple machines is ≤0.02s.

[0080] The formula for correcting the offset of eccentric irregular material points in S2 is as follows:

[0081]

[0082] in, Correct the coordinates of the robot support points. The initial planning coordinates are: e is the material's center of gravity eccentricity, L is the material's support span, and W is the material's overall width. When the material's center of gravity eccentricity e is ≥ 15mm, point correction is forcibly triggered.

[0083] Example 1:

[0084] Dual-machine collaborative handling of curved, irregularly shaped power pipes (for small-weight curved materials)

[0085] 1. Operating conditions

[0086] At a 110kV transmission line infrastructure construction site, the material to be transported is an arc-shaped anti-corrosion bend for power transmission. The material parameters are: weight 22kg, length 85cm, maximum radius of curvature 42cm, no flat clamping surface, center of gravity in the middle, construction terrain is a gravel slope with an 8° gradient, and the transport distance is 50m.

[0087] 2. Implementation Steps

[0088] S1. 3D modeling classification: The laser radar scan of the bent pipe generates a 3D point cloud model, which is determined to be a curved symmetrical irregular material.

[0089] S2. Robot grouping: Two quadruped robots work together. Based on the symmetrical characteristics of the center of gravity, the support points are planned symmetrically. The distance between the left and right support points and the center of gravity is 42.5cm.

[0090] S3. Load Balancing Calculation: Substitute into the load formula , The slope is 8°, and ΔF = 4N; the calculation yields... , The force deviation was 0.18%, which meets the preset standard;

[0091] S4. Posture pre-adjustment: Based on the 8° slope parameter, adjust the robot's body roll angle to 8° and the quadruped gait stride to 22cm.

[0092] S5. Synchronous clamping and lifting: Two robots adaptively adjust the clamping angle to fit the curved surface of the pipe. The pressure sensor array maintains a stable single-point pressure of 85-92N, ensuring uniform fit. The material is simultaneously lifted 15cm off the ground at a speed of 5cm / s.

[0093] S6. Travel correction: Closed-loop correction of speed and attitude throughout the entire process with a cycle of 10ms, travel speed of 0.3m / s, maximum material offset of 3.2mm throughout the entire process, and multi-machine synchronization error of 0.012s;

[0094] S7. Material Drop and Reset: Accurately reach the target point, drop the material at a uniform speed, clamp and reset, and complete the operation.

[0095] 3. Implementation Results

[0096] The entire handling process is free from slippage, deviation, and overturning, taking only 92 seconds, compared to 240 seconds for traditional manual handling of the same material, resulting in a 161% increase in efficiency. The uniformity of force distribution is far superior to existing technologies.

[0097] Example 2: Three-machine collaborative handling of eccentric irregular-shaped cable reel (eccentric heavy-duty materials)

[0098] 1. Operating conditions

[0099] At the construction site of a substation renovation and expansion project, the material to be transported is an eccentric irregular-shaped cable reel. The material parameters are: weight 65kg, diameter 78cm, thickness 32cm, center of gravity eccentricity 22mm (uneven weight distribution on one side), construction terrain is a muddy and uneven road surface with a slope of 15°, and the transport distance is 35m.

[0100] 2. Implementation Steps

[0101] S1. 3D modeling and classification: Scanning and identifying the eccentric features of the cable reel, determining it to be a unilaterally eccentric irregular material, and extracting the eccentricity of 22mm;

[0102] S2. Robot grouping and point correction: Three robots work together. The initial symmetrical points are calculated using the eccentricity correction formula X_correction = X_original + (e×L) / W to correct the offset of the single-sided support point by 18mm, forming an asymmetrical triangular support layout.

[0103] S3. Dynamic load distribution: With a slope of 15° and ΔF = 7.5N, the load of the three robots is distributed based on the center of gravity distance weight. The calculated bearing pressures are 228.5N, 219.2N, and 223.8N, respectively, with a maximum force deviation of 3.2%, which meets the accuracy requirements.

[0104] S4. Terrain and attitude adaptation: For 15° slope conditions, the aircraft pitch and roll angles are adjusted in stages to optimize the gait of the four-legged landing and compensate for the unevenness of the road surface.

[0105] S5. Flexible clamping operation: The matrix pressure sensor array adaptively adjusts the clamping force, and the eccentric pressure is automatically compensated to avoid one-sided suspension.

[0106] S6. Full-process deviation correction handling: The posture is corrected in real time during the movement, with a maximum material offset of 6.5mm and a synchronization error of 0.018s;

[0107] S7. Fault-tolerant monitoring: The entire process is fault-free and the handling operation is completed stably.

[0108] 3. Implementation Results

[0109] It perfectly solves the problem of uneven load for eccentric and irregularly shaped materials, with no one-sided overload or material tilting. It has good stability in complex muddy slopes and a 100% success rate in handling. Compared with the traditional load-sharing scheme, the force deviation is reduced by 9.8%.

[0110] Example 3: Four-machine collaborative operation of irregularly shaped iron tower fittings + fault-tolerant handling (large-sized irregular materials)

[0111] 1. Operating conditions

[0112] At a construction site of a mountainous power transmission line, there is a 125kg irregularly shaped tower connecting hardware to be transported. It has an irregular shape, multiple angles, and no symmetrical structure, with a maximum size of 115cm. The construction terrain is a gravelly and uneven terrain with a slope of 22°. The transport distance is 60m. A low-voltage fault test is set for the single unit during the journey.

[0113] 2. Implementation Steps

[0114] S1. Modeling and Classification: Identified as irregularly shaped materials with sharp edges, and adopted a four-machine rectangular support layout;

[0115] S2. Dynamic load distribution: 22° slope, ΔF=11N, the load of the four machines is quantitatively distributed according to the distance between their centers of gravity, and the initial force deviation is controlled within 3.1%;

[0116] S3, Extreme Attitude Adaptation: For terrain with a steep slope of 22°, the aircraft attitude is adjusted in multiple levels, and the four-legged differentiated gait compensates for terrain differences.

[0117] S4. Normal handling: Stable synchronous handling for the first 40m, with material deviation always ≤7mm;

[0118] S5. Fault-tolerant reconfiguration: If the power voltage of robot No. 1 is found to be lower than the threshold (fault state) during the handling process, the system will trigger the fault tolerance mechanism within 0.1s, automatically remove the faulty robot, distribute the load borne by robot No. 1 to the remaining 3 robots according to the weight, and re-plan the support force parameters and travel path.

[0119] S6. Secondary correction operation: After reconstruction, the force deviation of the multiple machines is re-corrected to 3.3%, and the remaining 20m of material handling is completed, and finally the material is accurately dropped.

[0120] 3. Implementation Results

[0121] The system successfully achieved uninterrupted material handling under fault conditions, and the system stability remained unchanged after reconstruction. No materials were tilted or shifted, which verified the practicality of the fault-tolerant reconstruction mechanism of this invention and solved the industry problem of task interruption caused by the failure of field construction robots.

[0122] Specific parameter comparison table

[0123]

[0124] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed in this application.

Claims

1. A quadruped robot collaborative handling system for irregularly shaped materials in power infrastructure construction, characterized in that, include: A cloud-based collaborative scheduling module and multiple quadruped transport robot units; the cloud-based collaborative scheduling module coordinates and controls the scheduling of the multiple quadruped transport robot units. The cloud-based collaborative scheduling module includes a dynamic collaborative control module, a terrain attitude correction module, and a fault tolerance and protection module. The quadruped transport robot unit includes an onboard sensing and positioning module and an adaptive gripping and execution module; The cloud-based collaborative scheduling module is used to receive power infrastructure transportation tasks, input parameters of irregularly shaped materials, construction terrain data, and transportation start and end coordinates, and complete material classification and robot grouping scheduling. The dynamic collaborative control module has a built-in load balancing algorithm and synchronous motion control model, which are used for load distribution, speed synchronization and displacement matching among multiple robots. The terrain posture correction module is used to correct the robot's quadruped gait and body balance posture in real time based on the slope and unevenness data of the construction site. The fault-tolerant protection module is used to monitor the robot's communication, power, and force status in real time. When a single robot fails, it automatically reconstructs the collaborative group to ensure that the handling task is not interrupted. The airborne perception and positioning module integrates a lidar, a binocular depth camera, a six-axis inertial sensor, and a UWB high-precision positioning unit, with a positioning accuracy of ±2cm and a frame rate of 30FPS. It is used to collect material three-dimensional contours, terrain undulation data, robot pose parameters, and body force data in real time. The adaptive clamping execution module is located on the top of the quadruped handling robot's single body and includes a clamping manipulator, a plate clamp, a flexible anti-slip clamping pad, a pressure sensor array, and an angle fine-tuning servo motor. It can adaptively adjust the clamping distance and fitting angle according to the curved surface, edges, and irregular contours of the irregularly shaped material.

2. The quadruped robot collaborative handling system for irregularly shaped materials in power infrastructure as described in claim 1, characterized in that: The pressure sensor array of the adaptive clamping execution module is distributed in a matrix on the clamping surface. The single-point pressure detection range is 0-500N, and the detection accuracy is ±0.8N. It can collect the contact pressure distribution between irregular materials and the clamping surface in real time, so as to achieve uniform force and fit on irregular contact surfaces.

3. The quadruped robot collaborative handling system for irregularly shaped materials in power infrastructure as described in claim 2, characterized in that: The quadruped transport robot is a high-load off-road quadruped robot with a maximum single-body load of 80kg, an adjustable walking stride of 0-35cm, and adaptive adjustment of body pitch angle ±30° and roll angle ±25°.

4. The quadruped robot collaborative handling system for irregularly shaped materials in power infrastructure as described in claim 1, characterized in that: The end of the gripping manipulator is connected to a plate clamp, which consists of two plates that can move relative to each other and in opposite directions. One of the plates is driven to move by an electric telescopic rod. Flexible anti-slip gripping pads are evenly connected to the opposing gripping surfaces of the two plates by angle fine-tuning servos. Pressure sensors are embedded inside the flexible anti-slip gripping pads, and all the pressure sensors form a pressure sensing array.

5. A method for collaborative handling of irregularly shaped materials in power infrastructure using a quadruped robot, implemented based on the collaborative handling system for irregularly shaped materials in power infrastructure as described in any one of claims 1-4, characterized in that... Includes the following steps: S1. Construction scenario and non-standard material parameter acquisition and modeling: The handling task is entered through the cloud scheduling module, and the airborne perception and positioning module scans the non-standard electrical materials to be handled, constructs a three-dimensional point cloud model, and extracts the material length, width, height, center of gravity offset, surface curvature, and weight parameters. The materials are divided into four categories: symmetrical center of gravity, unilateral eccentricity, curved surface non-standard, and irregular edge and corner type. S2. Robot Adaptive Grouping and Position Layout: The number of robots is matched according to the size and weight of the material. For materials weighing 5-30kg, two robots work together; for materials weighing 30-80kg, three robots work together; and for materials weighing 80-150kg, four robots work together. The support points of each robot are automatically planned based on the coordinates of the material's center of gravity. For eccentric materials, an asymmetrical position layout is used to correct the support offset error. S3. Dynamic Load Balancing Quantitative Allocation: Calculates the real-time load threshold of each robot using a preset load balancing model, and dynamically allocates the load based on feedback data from the robot's pressure sensors, controlling the force deviation among multiple robots to ≤3.5%. S4. Terrain Adaptation Posture Pre-adjustment: Collect data on terrain slope, unevenness, and obstacle distribution at the construction site. Within the terrain slope range of 0°-25°, adjust the robot's roll and pitch angles according to the slope threshold and simultaneously optimize the four-legged support gait. S5. Multi-machine synchronous clamping and lifting operation: Each robot moves and aligns according to the planned points. The adaptive clamping module fits the material contour to complete flexible clamping. The pressure sensor array detects the fit. When the pressure at a single point is less than 30N, the clamping angle is automatically fine-tuned. The lifting speed is uniformly set to 5cm / s to ensure lifting synchronization. S6. Real-time closed-loop correction during handling: Real-time acquisition of robot pose, travel speed and material offset data during handling. Parameter correction is completed every 10ms. Secondary posture correction is triggered when the material offset exceeds 8mm. S7. Fault-tolerant collaborative reconfiguration: Real-time monitoring of robot operating status. When a single-machine communication interruption, power failure, or overload failure is detected, the remaining robot load is redistributed and the path is reconfigured within 0.1 seconds to continue completing the handling task. S8. Fixed-point material drop and task reset: After the robot reaches the target point, it drops the material at a constant speed, releases the clamping and reset structure, and returns to the standby point to complete a single collaborative handling task.

6. A method for collaborative handling of irregularly shaped materials in power infrastructure using a quadruped robot according to claim 5, characterized in that: The quantitative calculation formula for the dynamic load balancing model in S3 is as follows: ； in, To ensure the i-th robot bears the pressure in real time, The weight is determined by the unilateral partition of the material. Let be the straight-line distance from the support point of the i-th robot to the center of gravity of the material. It is the sum of the distances from all support points to the center of gravity. The pressure value is the compensation value for terrain slope; for every increase in slope... , Increment by 2.5N to achieve dynamic compensation of slope and terrain load.

7. A method for collaborative handling of irregularly shaped materials in power infrastructure using a quadruped robot according to claim 5, characterized in that: The synchronous motion error correction formula in S6 is as follows: ； ； in, This is the speed deviation value. The system is preset with a standard travel speed. For the robot's real-time movement speed, The gait correction angle is k, the correction coefficient is 0.85, and t is the correction period of 10ms, ensuring that the synchronization error of multiple machines is ≤0.02s.

8. A method for collaborative handling of irregularly shaped materials in power infrastructure using a quadruped robot according to claim 5, characterized in that: The formula for correcting the offset of eccentric irregular material points in S2 is as follows: ； in, Correct the coordinates of the robot support points. The initial planning coordinates are: e is the material's center of gravity eccentricity, L is the material's support span, and W is the material's overall width. When the material's center of gravity eccentricity e is ≥ 15mm, point correction is forcibly triggered.

Images