A networked multi-machine linkage wire-laying robot control system

Abstract

The present invention relates to the field of robot control technology, and in particular to a networked multi-machine linkage wire-laying robot control system. It comprises a robot collaboration module, a positioning and tracking module, a collision avoidance planning module, a synchronous deviation correction module and a communication and coordination module. The robot collaboration module is used for the collaborative work between multiple wire-laying robots and the dynamic allocation of tasks; the positioning and tracking module tracks and predicts the robot's motion state in real time through high-precision positioning and causal convolutional neural network algorithms; the collision avoidance planning module plans the optimal collision avoidance path through dynamic obstacle avoidance algorithms and path planning algorithms; the synchronous deviation correction module uses fuzzy control and Bezier curve fitting to achieve precise trajectory correction; the communication and coordination module builds a high-speed wireless communication network to support distributed control and data transmission. The present invention greatly improves the collaborative operation capability of multiple wire-laying robots in complex environments, and realizes efficient and reliable wire-laying task execution.

Description

Technical Field

[0001] The invention relates to the technical field of robot control, in particular to a networked multi-machine linkage wire-laying robot control system. Background Art

[0002] With the continuous increase of construction projects, road construction, airport planning and other large-scale facility construction projects, line laying, as an important step in the early stage of construction, directly affects the quality and progress of the entire project. Traditional line laying work mainly relies on manual measurement. Although some automated equipment has been used to improve the accuracy and efficiency of line laying, the existing automatic line laying equipment is usually a single-machine working mode, which cannot meet the needs of efficient and multi-area synchronous operation in large-scale complex scenes. When facing large commercial buildings, airports, exhibition halls and other environments with complex structures or large areas, this type of equipment still has the following problems: In the traditional multi-machine line laying system, each robot usually works independently and lacks an effective coordination mechanism, resulting in unreasonable task allocation, low resource utilization, and failure to give full play to the collaborative advantages of multiple robots; the traditional line laying robot system lacks the ability to intelligently avoid collisions and optimize paths when working, which may cause collisions between robots or unreasonable path planning, further affecting work efficiency and safety; in the existing technology, when multiple line laying robots draw the same task synchronously, trajectory deviations and synchronization errors are prone to occur, especially when two robots lay out lines from different areas at the same time, it is difficult to maintain consistency. Summary of the invention

[0003] To solve the above problems, the present invention provides a networked multi-machine linkage wire-laying robot control system, which solves the problem of how to use multiple wire-laying robots in large-scale, complex terrain cable laying without effective coordination mechanism, intelligent collision avoidance capability and synchronization control accuracy, thereby improving the collaborative operation capability of multiple wire-laying robots in complex environments and achieving efficient and reliable execution of wire-laying tasks.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] A networked multi-machine linkage wire-laying robot control system, comprising a robot collaboration module, a positioning and tracking module, a collision avoidance planning module, a synchronous deviation correction module and a communication coordination module connected in sequence;

[0006] The robot collaboration module includes multiple wire-laying robots, which are used to establish a collaborative working mechanism between the multiple wire-laying robots and dynamically allocate wire-laying tasks based on the real-time position information, task load and working area of ​​each robot;

[0007] The positioning and tracking module is used to perform high-precision positioning and real-time tracking of the position and posture of each wire-laying robot through an absolute three-coordinate measuring device, obtain its parameter information, and use a causal convolutional neural network algorithm to establish a motion state model of the wire-laying robot, perform real-time positioning of the robot's current position and predict its motion trend;

[0008] The collision avoidance planning module is used to plan the motion path of each wire-laying robot based on the parameter information provided by the positioning and tracking module, using a dynamic obstacle avoidance algorithm and a path planning algorithm, and generate an optimal collision avoidance path;

[0009] The synchronous deviation correction module is used to perform synchronous control and trajectory correction based on real-time communication and position feedback when two or more wire-laying robots simultaneously draw the same straight line from both ends in different areas, and calculate the line deviation through the least squares method, and use Bezier curve fitting to achieve trajectory smoothing, so as to correct the deviation in the drawing process in real time;

[0010] The communication coordination module is used to build a high-speed data wireless communication network between robots and between robots and absolute three-dimensional coordinate measurement equipment. It adopts self-organizing network protocols and data compression algorithms to support distributed control, real-time collaboration, splicing coverage of multi-region laying-out ranges, and dynamic switching of absolute three-dimensional coordinate measurement equipment.

[0011] Furthermore, the parameter information includes real-time position information, posture information and environmental information of the wire-releasing robot.

[0012] Furthermore, the operation process of the positioning tracking module includes the following steps:

[0013] The position and posture of each wire-laying robot are measured with high precision through absolute three-dimensional coordinate measuring equipment, and its parameter information is obtained in real time;

[0014] The acquired parameter information is input into the causal convolutional neural network algorithm, and combined with the working path in the wire-laying task, the motion state model of the wire-laying robot is established to capture the dynamic behavior characteristics and motion laws of the robot;

[0015] Based on the motion state model, the current position of the wire-laying robot is located in real time, and its motion trend is predicted;

[0016] Compare the actual position and motion state with the output of the prediction model, analyze the deviation between the two, and use the adaptive error correction algorithm to correct the robot positioning results in real time;

[0017] Based on the positioning information after error correction, accurate real-time parameter information is provided to the collision avoidance planning module and the synchronous correction module.

[0018] Furthermore, the formula of the motion state model of the wire-laying robot is as follows:

[0019]

[0020] Among them, s t represents the predicted state vector of the wire-laying robot at time t, including position, velocity, acceleration, and attitude angle, which is used for real-time positioning and motion trend prediction; σ represents the activation function; L represents the total number of convolutional layers, which is used to capture motion characteristics at different time scales; Represents the weight matrix of the i-th convolution kernel in the l-th layer; Indicates that the lth layer is at time td l The input feature vector of i, including the parameter information obtained by the positioning and tracking module; d l represents the expansion coefficient of the lth layer; k l represents the size of the convolution kernel in layer l; b represents the bias term.

[0021] Furthermore, the operation process of the collision avoidance planning module includes the following steps:

[0022] Based on the parameter information provided by the positioning and tracking module, the current state data of each wire-laying robot is obtained, and the current state information includes the position, speed, direction and distribution of surrounding obstacles of the wire-laying robot;

[0023] Use dynamic obstacle avoidance algorithms to analyze the path of the wire-laying robot, combine the robot's work tasks and path restrictions, and adjust the path in real time according to the dynamic changes of obstacles;

[0024] Through the path planning algorithm, the optimal collision avoidance path between the robot and the obstacle is calculated according to the current obstacle position and movement trend, and the optimal collision avoidance path is generated;

[0025] The optimal collision avoidance path instruction is sent to the relevant wire-laying robot.

[0026] Furthermore, the calculation formula of the optimal collision avoidance path is as follows:

[0027]

[0028] Among them, P opt (t) represents the optimal collision avoidance path position; P(t) represents the current position of the wire-releasing robot at time t; v(t) represents the velocity vector of the wire-releasing robot at time t, that is, the current travel speed of the wire-releasing robot; Δt represents the time step, which is used to predict the position change of the robot in a short period of time in the future; R s Indicates safe distance; d i (t) represents the distance between the line-releasing robot and the i-th obstacle at time t; It represents the unit direction vector of the current position of the wire-laying robot pointing to the i-th obstacle; n represents the number of surrounding obstacles.

[0029] Furthermore, the operation process of the synchronous correction module includes the following steps:

[0030] When two or more wire-laying robots draw the same straight line from both ends at the same time in different areas, obtain the current real-time position information, posture information and trajectory data of the two or more robots;

[0031] Based on real-time data, the fuzzy control algorithm is used to calculate the trajectory deviation between the two robots and generate correction instructions;

[0032] Based on the correction instruction, the deviation is calculated using the least square method, and the trajectory is smoothed using the Bezier curve fitting algorithm;

[0033] Use the corrected trajectory data to adjust the motion paths and drawing behaviors of two or more robots in real time;

[0034] During the correction process, the actual position and the position after correction are continuously compared, and adjustments are repeated until the trajectory error of two or more robots drops to within the predetermined threshold, thereby completing synchronous correction.

[0035] Furthermore, the construction process of the fuzzy control algorithm includes the following steps:

[0036] The trajectory deviation, position error and attitude angle error between two or more wire-laying robots are used as input variables of fuzzy control, and the fuzzy processing is performed according to the error size and change rate to generate the corresponding fuzzy set.

[0037] Based on the fuzzy set of input variables, a fuzzy inference rule base is established to set appropriate control instructions according to the error size and change trend;

[0038] According to the position, posture and trajectory information of two or more wire-laying robots in actual operation, the fuzzy reasoning rule library is used to generate the deviation correction control instructions through the fuzzy reasoning method;

[0039] Defuzzify the inference results and generate precise control instructions.

[0040] Furthermore, the communication coordination module introduces an intelligent collaborative scheduling mechanism to adaptively adjust the data transmission path and frequency according to the robot's task priority and network load.

[0041] The beneficial effects of the present invention are:

[0042] The present invention can dynamically allocate tasks between multiple wire-laying robots through the robot collaboration module, and coordinate based on the real-time position information and work area of ​​each robot, so as to optimize the task allocation to the maximum extent and improve the overall efficiency of the wire-laying task. The positioning and tracking module can obtain the robot position and posture parameters in real time and predict its movement trend through the absolute three-coordinate measurement device and the causal convolutional neural network algorithm, which provides a basis for subsequent collision avoidance and path planning, and ensures that the robot's movement path is accurate. The collision avoidance planning module uses dynamic obstacle avoidance and path planning algorithms to plan the optimal collision avoidance path according to the real-time position and environmental information of the robot, effectively avoiding the collision problem when multiple robots are working at the same time, and improving safety and operation continuity. The synchronous correction module can perform synchronous control when multiple robots draw the same straight line at the same time through real-time communication and position feedback, and adopts fuzzy control algorithm and Bezier curve fitting to correct and smooth the trajectory, ensuring that the trajectory of the wire-laying is accurate and without deviation, and improving the quality of the wire-laying. The communication coordination module establishes a high-speed wireless communication network. Through self-organizing network protocols and data compression algorithms, it supports real-time collaboration among multiple robots and dynamic switching of absolute three-coordinate measurement equipment, ensuring the smooth completion of distributed control and large-area layout tasks, and further improving the stability and scalability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a module schematic diagram of a networked multi-machine linkage wire-laying robot control system of the present invention.

[0044] Figure 2 It is a flowchart of the operation process of the positioning and tracking module provided by an embodiment of the present invention.

[0045] Figure 3 It is a flowchart of the operation process of the collision avoidance planning module provided by one embodiment of the present invention.

[0046] Figure 4 It is a schematic diagram of the structure of a robot collaboration module provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0047] See also Figure 1-4 As shown, the present invention relates to a networked multi-machine linkage wire-laying robot control system.

[0048] Example

[0049] A networked multi-machine linkage wire-laying robot control system, comprising a robot collaboration module, a positioning and tracking module, a collision avoidance planning module, a synchronous deviation correction module and a communication coordination module connected in sequence;

[0050] The robot collaboration module includes multiple wire-laying robots, which are used to establish a collaborative working mechanism between the multiple wire-laying robots and dynamically allocate wire-laying tasks based on the real-time position information, task load and working area of ​​each robot;

[0051] It should be noted that the robot collaboration module includes multiple wire-laying robots. Specifically, the central controller controls robots 1, 2, 3 to n for collaborative control. Each wire-laying robot is equipped with the following main hardware devices:

[0052] Embedded Control Unit (ECU): used to perform various task scheduling and calculations. The control unit has high computing performance and can process sensor data, path planning, and task allocation in real time.

[0053] Wireless communication unit: supports high-speed, low-latency wireless communication (such as 5G, Wi-Fi Mesh, LoRa, etc.). Each robot maintains continuous two-way communication with the central control system or other robots.

[0054] Motion control unit: A system of motors and sensors to achieve precise movement and posture control.

[0055] Sensor unit: including cameras, inertial measurement units (IMUs), etc., used for environmental perception, obstacle detection and self-positioning.

[0056] Specifically, the main workflow of the task allocation mechanism of the robot collaboration module is as follows:

[0057] Initial task allocation: At the beginning of a task, the system makes a preliminary task division based on the global map, task scope, and the initial position and working status of each robot. The core of task allocation is to enable each robot to cover the assigned area in the shortest time while maximizing its working capacity.

[0058] Dynamic task adjustment: When some robots complete their tasks or encounter obstacles that hinder their work, the system will reallocate tasks by monitoring their task progress in real time. For example, when a robot needs to return to recharge due to low battery, the collaborative module assigns its unfinished tasks to nearby robots.

[0059] Task priority mechanism: Tasks in certain areas have higher priorities (such as line laying tasks in key areas or tasks that need to be completed within a specific time). The collaborative module ensures that these high-priority tasks are assigned to the robot in the best state. By sorting the importance of tasks, it ensures that key tasks can be completed first.

[0060] It should be noted that each robot will transmit its own location information, task completion progress, current load and other data to other robots and the central control system through the wireless network. Each robot will also receive the working status information of other robots in real time in order to adjust its own tasks. In order to prevent the overlapping of working areas between different robots, the collaborative module divides the entire task area into several sub-areas. Each robot only works in its assigned area, and the tasks at the boundary of the area need to be coordinated through communication between robots. When two or more robots need to lay out lines at the boundary at the same time, they will avoid path overlap through real-time negotiation. When a robot is about to complete the task in its current area, the collaborative module will notify the neighboring robot to hand over the task before its work is about to end. Through the handover mechanism, two or more robots can be prevented from entering the same working area at the same time, reducing the risk of duplication of work and collision.

[0061] The collaborative module monitors the battery power of each robot in real time. When the battery power of a robot approaches the critical point, it will automatically reduce its task load and arrange it to return to the charging station for charging. At the same time, other robots will take over its unfinished tasks to ensure that the work is not interrupted. During the execution of the task, if the workload of a robot is too high or the work efficiency is reduced due to certain external factors (such as obstacles, complex terrain), the collaborative module will dynamically adjust its task area and redivide the work tasks to ensure that other robots can quickly replenish their task areas. In the line-laying task, some tasks have strict time requirements, while some tasks can tolerate certain delays. The collaborative module will adjust the robot's working order according to the task type, give priority to time-critical tasks, and assign tasks with high delay tolerance to robots with larger loads.

[0062] If a robot fails (such as motor failure or sensor failure), the collaborative module will automatically transfer the current task of the robot to other robots and notify maintenance personnel to repair the faulty robot. The dynamic transfer of tasks can prevent the entire system from stopping due to problems with a single robot. If a robot temporarily loses communication with other robots due to signal problems, the collaborative module will record the last known position and status of the robot for a period of time, and enable it to complete the current task independently through a self-correction mechanism. When communication is restored, the collaborative module will re-incorporate the robot into the coordinated management of the entire system. When new obstacles or environmental changes appear in the working area (such as wind causing unstable line laying), the collaborative module will monitor environmental data in real time and dynamically adjust the robot's movement route or task allocation based on this data.

[0063] The positioning and tracking module is used to perform high-precision positioning and real-time tracking of the position and posture of each wire-laying robot through an absolute three-coordinate measuring device, obtain its parameter information, and use a causal convolutional neural network algorithm to establish a motion state model of the wire-laying robot, perform real-time positioning of the robot's current position and predict its motion trend; the parameter information includes the real-time position information, posture information and environmental information of the wire-laying robot.

[0064] The operation process of the positioning and tracking module includes the following steps:

[0065] The position and posture of each wire-laying robot are measured with high precision through absolute three-dimensional coordinate measuring equipment, and its parameter information is obtained in real time;

[0066] Specifically, the absolute three-dimensional coordinate measurement equipment includes but is not limited to laser trackers and total stations, and can use global positioning systems (GPS), inertial navigation systems (INS) or other optical sensors to obtain accurate position information. Each robot is equipped with multiple sensors for collecting real-time motion data, as follows:

[0067] Position acquisition: The GPS system can obtain the robot's position information (x, y, z coordinates) in real time in three-dimensional space, with a measurement accuracy of centimeters. For local fine movements, an inertial navigation system (INS) is used to supplement the position information. When the GPS signal is poor, the inertial sensor calculates the robot's displacement through acceleration and angular velocity.

[0068] Attitude measurement: The inertial measurement unit (IMU) is used to obtain the attitude information of the robot when performing the line-laying task, including pitch angle, roll angle and yaw angle, to ensure that the robot can maintain a stable attitude in different terrains and environments.

[0069] Parameter information integration: Position and attitude information are integrated through the sensor fusion algorithm to generate a complete parameter set including position, velocity, acceleration, attitude, etc., and updated in real time.

[0070] The acquired parameter information is input into the causal convolutional neural network algorithm, and combined with the working path in the wire-laying task, the motion state model of the wire-laying robot is established to capture the dynamic behavior characteristics and motion laws of the robot;

[0071] Specifically, the specific construction process of the causal convolutional neural network algorithm (CCNN) is as follows:

[0072] The robot position, velocity, acceleration, and posture information are passed to the CCNN model as input features. The model takes the data of the current and historical time steps as input to capture the dynamic behavior characteristics of the wire-laying robot during movement.

[0073] In the early stage of system operation, the CCNN model is trained with a large amount of robot motion data, including the robot's trajectory, speed change, posture change, etc. During the task execution, the CCNN model can adjust the model parameters and optimize the prediction accuracy of the robot's motion state in combination with the working path in the specific wire laying task.

[0074] The model output includes the robot's current motion state (position, speed, posture) and the motion trend prediction at future moments. The model can infer the robot's future motion trajectory based on historical data and estimate the potential change direction and speed.

[0075] Based on the motion state model, the current position of the wire-laying robot is located in real time, and its motion trend is predicted;

[0076] Compare the actual position and motion state with the output of the prediction model, analyze the deviation between the two, and use the adaptive error correction algorithm to correct the robot positioning results in real time;

[0077] It should be noted that the adaptive error correction algorithm uses a recursive update method to adjust the position and attitude of the robot based on the real-time error analysis results. The adaptive algorithm can dynamically adjust the correction parameters according to different types of error sources to ensure that the system can adapt to error fluctuations in different environments. For example, when the sensor data fluctuates over a large range, the system will increase its reliance on historical data to smooth short-term errors. In order to improve the correction accuracy, the adaptive algorithm usually combines the data of multiple sensors for correction. For example, when the GPS signal is interfered with, the system will rely more on the acceleration and angular velocity information provided by the IMU, and correct the position and attitude through a multi-sensor fusion algorithm (such as the extended Kalman filter EKF). This method can effectively reduce the impact of a single sensor error. The adaptive algorithm will dynamically adjust the parameters used for correction. For example, when the robot is in an area with large errors and complex environment, the system will increase the frequency and weight of correction to ensure that the error can be corrected in time; when the error is small and the environment is stable, the system will reduce the correction frequency and reduce the computational load.

[0078] Based on the positioning information after error correction, accurate real-time parameter information is provided to the collision avoidance planning module and the synchronous correction module.

[0079] Furthermore, the formula of the motion state model of the wire-laying robot is as follows:

[0080]

[0081] Among them, s trepresents the predicted state vector of the wire-laying robot at time t, including position, velocity, acceleration, and attitude angle, which is used for real-time positioning and motion trend prediction; σ represents the activation function; L represents the total number of convolutional layers, which is used to capture motion characteristics at different time scales; Represents the weight matrix of the i-th convolution kernel in the l-th layer; Indicates that the lth layer is at time td l The input feature vector of i, including the parameter information obtained by the positioning and tracking module; d l represents the expansion coefficient of the lth layer; k l represents the size of the convolution kernel in layer l; b represents the bias term.

[0082] The collision avoidance planning module is used to plan the motion path of each wire-laying robot based on the parameter information provided by the positioning and tracking module, using a dynamic obstacle avoidance algorithm and a path planning algorithm, and generate an optimal collision avoidance path;

[0083] The operation process of the collision avoidance planning module includes the following steps:

[0084] Based on the parameter information provided by the positioning and tracking module, the current state data of each wire-laying robot is obtained, and the current state information includes the position, speed, direction and distribution of surrounding obstacles of the wire-laying robot;

[0085] Use dynamic obstacle avoidance algorithms to analyze the path of the wire-laying robot, combine the robot's work tasks and path restrictions, and adjust the path in real time according to the dynamic changes of obstacles;

[0086] Specifically, the dynamic obstacle avoidance algorithm continuously monitors changes in the robot's surrounding environment, especially the movement of dynamic obstacles (such as pedestrians, vehicles or other machines and equipment), and adjusts the robot's path in real time. This algorithm is usually combined with classic path planning algorithms such as the A* algorithm and the RRT (rapid random tree) algorithm, and is based on advanced algorithms such as machine learning or model predictive control (MPC), enabling the robot to avoid obstacles in complex environments. The collision avoidance planning module also needs to consider the robot's mission objectives and path restrictions. When planning an obstacle avoidance path, the algorithm must not only ensure that the robot can avoid obstacles, but also ensure that the robot is still moving towards the mission objective. Path restrictions may include prohibited areas, slopes, narrow passages, etc., and the system will constrain the robot's route based on these restrictions.

[0087] Through the path planning algorithm, the optimal collision avoidance path between the robot and the obstacle is calculated according to the current obstacle position and movement trend, and the optimal collision avoidance path is generated;

[0088] The optimal collision avoidance path instruction is sent to the relevant wire-laying robot.

[0089] It should be noted that the path planning instructions include a series of navigation points or continuous path curves. The robot adjusts its direction and speed according to these instructions to avoid obstacles. The collision avoidance planning module transmits these instructions to the robot controller through wireless communication or control bus. After the controller parses the instructions, it adjusts the robot's speed, steering angle, etc. in real time. When the robot executes the collision avoidance path, it will continuously feedback the current status information to the collision avoidance planning module. The collision avoidance planning module dynamically adjusts the path instructions based on these feedback data. If there are new changes in the environment, such as the sudden movement of obstacles or the appearance of new obstacles, the system will recalculate the path in real time and send updated path instructions.

[0090] Furthermore, the calculation formula of the optimal collision avoidance path is as follows:

[0091]

[0092] Among them, P opt (t) represents the optimal collision avoidance path position; P(t) represents the current position of the wire-releasing robot at time t; v(t) represents the velocity vector of the wire-releasing robot at time t, that is, the current travel speed of the wire-releasing robot; Δt represents the time step, which is used to predict the position change of the robot in a short period of time in the future; R s Indicates the safety distance, which is used to ensure that the robot maintains a certain minimum safety distance from obstacles to prevent collision; d i (t) represents the distance between the line-releasing robot and the i-th obstacle at time t; It represents the unit direction vector of the current position of the wire-laying robot pointing to the ith obstacle. It describes the direction of the obstacle relative to the robot and is used to determine the direction in which the robot should adjust to avoid the obstacle. It is calculated by the collision avoidance planning module based on the obstacle position information provided by the positioning and tracking module. n represents the number of surrounding obstacles, which is the total number of obstacles detected in real time by the positioning and tracking module. This number will change dynamically, reflecting the real-time obstacle information of the robot in its environment.

[0093] The synchronous deviation correction module is used to perform synchronous control and trajectory correction based on real-time communication and position feedback when two or more wire-laying robots simultaneously draw the same straight line from both ends in different areas, and calculate the line deviation through the least squares method, and use Bezier curve fitting to achieve trajectory smoothing, so as to correct the deviation in the drawing process in real time;

[0094] The operation process of the synchronous correction module includes the following steps:

[0095] When two or more wire-laying robots draw the same straight line from both ends at the same time in different areas, obtain the current real-time position information, posture information and trajectory data of the two or more robots;

[0096] Based on real-time data, the fuzzy control algorithm is used to calculate the trajectory deviation between the two robots and generate correction instructions;

[0097] Based on the correction instruction, the deviation is calculated using the least square method, and the trajectory is smoothed using the Bezier curve fitting algorithm;

[0098] Specifically, the least squares method calculates the distance difference between the trajectory points drawn in real time by two or more robots to obtain the trajectory deviation. The system minimizes the sum of the squares of these deviation points to ensure that the drawn straight lines remain consistent in space. The system fits the trajectory error between the two robots through the least squares method, so that their trajectory differences are gradually reduced and achieve an ideal synchronization state.

[0099] The Bezier curve generates a smooth curve based on a small number of control points. In this embodiment, the system selects several control points based on the position data of two or more robots to generate a smooth Bezier curve as the ideal path for the robot after correction. The system will continue to generate a smooth trajectory while drawing a straight line to ensure that the robot's route will not change dramatically when executing the correction instruction, thereby avoiding excessive vibration or deviation.

[0100] Use the corrected trajectory data to adjust the motion paths and drawing behaviors of two or more robots in real time;

[0101] During the correction process, the actual position and the position after correction are continuously compared, and adjustments are repeated until the trajectory error of the two robots drops to within the predetermined threshold, thus completing synchronous correction.

[0102] Specifically, after each deviation correction, the system will collect the robot's position and trajectory information again through sensors and compare it with the ideal trajectory. If the new trajectory still has a large deviation, the system will continue to generate new deviation correction instructions for fine-tuning.

[0103] The system will repeatedly adjust until the trajectory error of two or more robots falls within the predetermined tolerance range, ensuring that their trajectories are completely synchronized. Usually the error threshold can be set according to the specific application scenario, such as a permissible error of 1 mm or less.

[0104] The construction process of the fuzzy control algorithm includes the following steps:

[0105] The trajectory deviation, position error and attitude angle error between two or more wire-laying robots are used as input variables of fuzzy control, and the fuzzy processing is performed according to the error size and change rate to generate the corresponding fuzzy set.

[0106] Based on the fuzzy set of input variables, a fuzzy inference rule base is established to set appropriate control instructions according to the error size and change trend;

[0107] According to the position, posture and trajectory information of two or more wire-laying robots in actual operation, the fuzzy reasoning rule library is used to generate the deviation correction control instructions through the fuzzy reasoning method;

[0108] Defuzzify the inference results and generate precise control instructions.

[0109] The formula of the Bezier curve fitting algorithm is as follows:

[0110]

[0111] Among them, P(t) represents the smooth coordinate point on the motion trajectory of the wire-laying robot corresponding to a certain time t after the correction. This point is calculated in real time and is used to correct the trajectory deviation of two or more wire-laying robots in the synchronous drawing process; P i Indicates the coordinates of the control point used for correction during trajectory fitting. Control point P i The actual position and trajectory data of two or more wire-laying robots are obtained in real time through the positioning tracking module and the synchronous deviation correction module. This data comes from the real-time position information and posture information of the wire-laying robots. The number of sampling points is determined according to the complexity of the robot's current motion trajectory; B i，n (t) represents the Bessel basis function, which is used to determine the control point P i For the influence weight of the curve, the formula is as follows:

[0112]

[0113] in, represents the control point P i The weight of the control points in the curve generation. During the deviation correction process, the number of control points can be dynamically adjusted according to the complexity of the trajectory errors of two or more wire-laying robots; t represents the interpolation progress of the real-time trajectory data. When two or more robots draw straight lines from different areas at the same time, the system will determine t based on time and real-time data, so that the trajectories of the two robots gradually approach and finally align smoothly after the deviation correction; n represents the order of the Bezier curve, which is usually the number of real-time trajectory sampling points of the wire-laying robot minus 1. The more trajectory data points, the higher the order, and the smoother the generated curve. In the synchronous deviation correction module, the trajectory data is collected and input in real time by the positioning and tracking module to ensure that there are enough trajectory points for smoothing when fitting the Bezier curve.

[0114] The communication coordination module is used to build a high-speed data wireless communication network between robots and between robots and absolute three-dimensional coordinate measurement equipment. It adopts a self-organizing network protocol and a data compression algorithm to support distributed control, real-time collaboration, splicing coverage of multi-region laying-out ranges, and dynamic switching of absolute three-dimensional coordinate measurement equipment. The communication coordination module introduces an intelligent collaborative scheduling mechanism to adaptively adjust the data transmission path and frequency according to the robot's task priority and network load.

[0115] It should be noted that the communication coordination module uses a combination of multiple communication protocols, and the specific selection is dynamically switched according to the scenario requirements and real-time network status, as follows:

[0116] Self-organizing network protocol: Each robot and absolute three-dimensional coordinate measurement device can be used as a node in the network to form a self-organizing Mesh network. This network structure does not require a central control node and can dynamically adjust the communication path. When a node fails or cannot participate in communication, the Mesh network can automatically bypass the failed node and find the best path to transmit data.

[0117] QoS guarantee protocol: The communication coordination module supports the QoS mechanism to ensure that different data streams have different priorities. For example, high-priority data such as real-time location information and posture information will be guaranteed to be transmitted quickly, while task logs or non-critical status information can be allocated lower bandwidth.

[0118] Low-latency protocol: In order to ensure that the collaboration and collision avoidance planning between robots can respond quickly, the communication module introduces a low-latency communication protocol (such as UDP) to reduce the delay in communication and ensure that data can reach the receiving end within milliseconds.

[0119] Self-organizing network is one of the core technologies of the communication coordination module and has the following key features:

[0120] Dynamic topology management: When multiple robots are performing tasks at the same time, the network topology will change dynamically as the robots move. The self-organizing network can automatically sense the location of each robot and adjust the data transmission path according to the movement of the robot. When performing tasks, the robot can act as a relay node to transmit data to a distant robot, forming a stable communication link.

[0121] Redundant paths and fault-tolerance mechanisms: Each robot can connect to multiple adjacent nodes at the same time, thus forming redundant paths. Even if a communication link is interrupted due to environmental interference, the system can quickly switch to an alternative path to ensure that communication is not interrupted. This redundant design enhances the robustness of the system, especially in environments where signals are easily blocked (such as industrial scenarios with complex terrain and many metal obstacles).

[0122] Load balancing: The self-organizing network protocol also supports load balancing. When the load of some communication nodes is too high, the system will automatically transfer part of the communication tasks to the nodes with lighter load. This method effectively improves the efficiency of communication and avoids overloading of a single node.

[0123] It should be noted that the module uses lightweight data compression algorithms, such as differential coding and Huffman coding, to compress redundant data transmitted between robots. For example, for position information and posture data, the amount of data transmitted can be greatly reduced by only transmitting incremental information (i.e., the difference from the data at the previous time point). For data with high non-real-time requirements (such as task status, log information, etc.), the system will batch multiple data packets through a cache mechanism and send multiple data packets at one time, thereby reducing the communication frequency and saving bandwidth. Each robot has a certain edge computing capability and can perform preliminary processing on sensor data locally. For example, sensor data can be filtered locally on the robot first, and the noise information can be removed before being transmitted to other robots or the central control system, thereby reducing the burden on the communication network.

[0124] The system assigns different priorities to different data streams based on the type, urgency, and real-time requirements of the task. High-priority tasks (such as collision avoidance information and emergency fault alarms) will be transmitted immediately, while low-priority tasks (such as regular status updates or log data) can be transmitted later. The communication coordination module can monitor the load of the entire network in real time and adaptively adjust the data transmission frequency based on the current network bandwidth and node load. For example, when the network load is low, the robot can transmit data at a high frequency; when the load is high, the frequency of low-priority data transmission will automatically decrease to ensure the transmission of high-priority data. The communication coordination module can dynamically select the best communication path and node scheduling strategy based on the current task progress, location, and network load of each robot. By calculating the workload and transmission bandwidth of each node, the system can avoid certain nodes from becoming communication bottlenecks and make the communication network more balanced by adjusting the routing.

[0125] In summary, through the robot collaboration module, the present application enables multiple wire-laying robots to dynamically allocate and adjust work tasks during task execution, realize real-time collaborative work, and make full use of robot resources. The task allocation mechanism can ensure that each robot is assigned appropriate tasks according to its working status, position and task priority, avoiding waste of resources and improving the overall task execution efficiency. Through the positioning and tracking module, the causal convolutional neural network algorithm is used to perform high-precision real-time positioning and motion trend prediction of the robot. The system can adjust the positioning error in real time through multi-sensor data fusion, ensure that the robot can maintain high-precision positioning in complex environments, and improve the stability and reliability of the wire-laying task.

[0126] The collision avoidance planning module in this application can dynamically adjust the robot's path according to changes in the robot's surrounding environment, ensuring that the robot can safely avoid obstacles and maintain its work objectives in complex environments. The optimal collision avoidance path generated by the path planning algorithm can ensure the continuity and safety of task execution and reduce unexpected problems that may be encountered during work. The synchronous correction module can ensure that multiple robots remain synchronized when drawing the same route at the same time, and uses fuzzy control algorithms and Bezier curve fitting to achieve real-time trajectory correction, reducing trajectory errors and ensuring drawing accuracy, effectively improving the quality of work tasks.

[0127] The communication coordination module in this application realizes high-speed wireless communication between multiple robots and the central control system through a self-organizing network protocol. The system supports distributed control and real-time collaboration, and can dynamically adjust the data transmission path to ensure efficient and reliable communication between robots and reduce communication delays, thereby improving the flexibility and robustness of overall task execution.

[0128] This application has a powerful task dynamic adjustment function, which can make real-time adjustments and allocations according to the task progress, battery status and workload of each robot to ensure that the task is not interrupted. At the same time, if a robot fails, the system can transfer its task in time to avoid the interruption of the entire system due to a single robot failure, thereby improving the fault tolerance of the system. The communication module adopts data compression algorithm and load balancing mechanism, reduces the data transmission burden through edge computing capabilities, realizes efficient data transmission and processing, reduces communication delays and network bandwidth pressure, and further improves the overall performance and stability of the system.

[0129] The above implementation modes are merely descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary engineering and technical personnel in the field shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A networked multi-machine linkage wire-laying robot control system, characterized in that: It includes a robot collaboration module, a positioning and tracking module, a collision avoidance planning module, a synchronous deviation correction module and a communication coordination module which are connected in sequence; The robot collaboration module includes multiple wire-laying robots, which are used to establish a collaborative working mechanism between the multiple wire-laying robots and dynamically allocate wire-laying tasks based on the real-time position information, task load and working area of ​​each robot; The positioning and tracking module is used to perform high-precision positioning and real-time tracking of the position and posture of each wire-laying robot through an absolute three-coordinate measuring device, obtain its parameter information, and use a causal convolutional neural network algorithm to establish a motion state model of the wire-laying robot, perform real-time positioning of the robot's current position and predict its motion trend; The collision avoidance planning module is used to plan the motion path of each wire-laying robot based on the parameter information provided by the positioning and tracking module, using a dynamic obstacle avoidance algorithm and a path planning algorithm, and generate an optimal collision avoidance path; The synchronous deviation correction module is used to perform synchronous control and trajectory correction based on real-time communication and position feedback when two or more wire-laying robots simultaneously draw the same straight line from both ends in different areas, and calculate the line deviation through the least squares method, and use Bezier curve fitting to achieve trajectory smoothing, so as to correct the deviation in the drawing process in real time; The communication coordination module is used to build a high-speed data wireless communication network between robots and between robots and absolute three-dimensional coordinate measurement equipment. It adopts self-organizing network protocols and data compression algorithms to support distributed control, real-time collaboration, splicing coverage of multi-region laying-out ranges, and dynamic switching of absolute three-dimensional coordinate measurement equipment.

2. A networked multi-machine linkage wire-laying robot control system according to claim 1, characterized in that: The parameter information includes the real-time position information, posture information and environment information of the wire-releasing robot.

3. A networked multi-machine linkage wire-laying robot control system according to claim 2, characterized in that: The operation process of the positioning tracking module includes the following steps: The position and posture of each wire-laying robot are measured with high precision through absolute three-dimensional coordinate measuring equipment, and its parameter information is obtained in real time; The acquired parameter information is input into the causal convolutional neural network algorithm, and combined with the working path in the wire-laying task, the motion state model of the wire-laying robot is established to capture the dynamic behavior characteristics and motion laws of the robot; Based on the motion state model, the current position of the wire-laying robot is located in real time, and its motion trend is predicted; Compare the actual position and motion state with the output of the prediction model, analyze the deviation between the two, and use the adaptive error correction algorithm to correct the robot positioning results in real time; Based on the positioning information after error correction, accurate real-time parameter information is provided to the collision avoidance planning module and the synchronous correction module.

4. A networked multi-machine linkage wire-laying robot control system according to claim 3, characterized in that: The formula of the motion state model of the wire-laying robot is as follows: ； in, Represents the predicted state vector of the wire-laying robot at time t, including position, velocity, acceleration and attitude angle, which is used for real-time positioning and motion trend prediction; represents the activation function; Represents the total number of convolutional layers, which is used to capture motion features at different time scales; Represents the weight matrix of the i-th convolution kernel in the l-th layer; Indicates that the lth layer is at time The input feature vector includes parameter information obtained through the positioning and tracking module; represents the expansion coefficient of the lth layer; represents the size of the convolution kernel in layer l; b represents the bias term.

5. A networked multi-machine linkage wire-laying robot control system according to claim 1, characterized in that: The operation process of the collision avoidance planning module includes the following steps: Based on the parameter information provided by the positioning and tracking module, the current state data of each wire-laying robot is obtained, and the current state data includes the position, speed, direction and distribution of surrounding obstacles of the wire-laying robot; Use dynamic obstacle avoidance algorithms to analyze the path of the wire-laying robot, combine the robot's work tasks and path restrictions, and adjust the path in real time according to the dynamic changes of obstacles; Through the path planning algorithm, the optimal collision avoidance path between the robot and the obstacle is calculated according to the current obstacle position and movement trend, and the optimal collision avoidance path is generated; The optimal collision avoidance path instruction is sent to the relevant wire-laying robot.

6. A networked multi-machine linkage wire-laying robot control system according to claim 5, characterized in that: The calculation formula of the optimal collision avoidance path is as follows: ； in, represents the optimal collision avoidance path position; represents the current position of the wire-releasing robot at time t; represents the velocity vector of the wire-releasing robot at time t, that is, the current travel speed of the wire-releasing robot; Represents the time step, which is used to predict the position change of the robot in a short period of time in the future; Indicates safe distance; represents the distance between the line-releasing robot and the i-th obstacle at time t; It represents the unit direction vector of the current position of the wire-laying robot pointing to the i-th obstacle; n represents the number of surrounding obstacles.

7. A networked multi-machine linkage wire-laying robot control system according to claim 1, characterized in that: The operation process of the synchronous correction module includes the following steps: When two or more wire-laying robots simultaneously draw the same straight line from both ends in different areas, obtain the current real-time position information, posture information and trajectory data of the two or more robots; Based on real-time data, the fuzzy control algorithm is used to calculate the trajectory deviation between the two robots and generate correction instructions; Based on the correction instruction, the deviation is calculated using the least square method, and the trajectory is smoothed using the Bezier curve fitting algorithm; Use the corrected trajectory data to adjust the motion paths and drawing behaviors of two or more robots in real time; During the correction process, the actual position and the position after correction are continuously compared, and adjustments are repeated until the trajectory error of two or more robots drops to within the predetermined threshold, thereby completing synchronous correction.

8. A networked multi-machine linkage wire-laying robot control system according to claim 7, characterized in that: The construction process of the fuzzy control algorithm includes the following steps: The trajectory deviation, position error and attitude angle error between two or more wire-laying robots are used as input variables of fuzzy control, and the fuzzy processing is performed according to the error size and change rate to generate the corresponding fuzzy set. Based on the fuzzy set of input variables, a fuzzy inference rule base is established to set appropriate control instructions according to the error size and change trend; According to the position, posture and trajectory information of two or more wire-laying robots in actual operation, the fuzzy reasoning rule library is used to generate the deviation correction control instructions through the fuzzy reasoning method; Defuzzify the inference results and generate precise control instructions.

9. A networked multi-machine linkage wire-laying robot control system according to claim 1, characterized in that: The communication coordination module introduces an intelligent collaborative scheduling mechanism to adaptively adjust the data transmission path and frequency according to the robot's task priority and network load.

Images