A positioning blind area correction control system for an automatic wire-laying robot

Abstract

The present invention relates to the technical field of automatic construction line laying control, and particularly to a positioning blind area deviation correction control system for an automatic line laying robot. By determining the contour of an obstacle. When the line laying robot encounters an obstacle and causes signal loss, the system records the loss and recovery positions, and determines the obstacle area by connecting the lines. The robot is made to move around the obstacle at a constant distance, and coordinate data is collected in real time and mapped onto a digital drawing. The system processes the signal blind area completely blocked by the obstacle, and by correcting the line laying coordinates in the drawing, ensures that the robot continues to perform precise line laying operations in the blind area and corrects the errors in the blind area. The system uses a path optimization algorithm to optimize the blind area entry point in combination with the number of turns and the path length, reducing error superposition. After each obstacle bypass or blind area line laying, the robot returns to the calibration point, and the position deviation is adjusted in real time through the Kalman filter algorithm to correct the error, improving the line laying accuracy in the blind area operation and reducing deviation accumulation.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic construction layout control, and particularly to a positioning blind area correction control system for an automatic layout robot. Background Art

[0002] With the development of the construction industry and the complexity of large-scale project construction, the requirements for the accuracy and efficiency of ground markings are getting higher and higher. In traditional construction processes, manual layout is a common construction preparation task. However, manual layout is not only inefficient but also prone to positioning errors and an increase in repetitive labor.

[0003] In the prior art, an automatic layout robot uses a laser locator to receive a preset digital drawing and emit laser points to the ground, enabling the layout robot to track the emitted laser points to complete the layout work. However, existing automatic layout systems still face some technical challenges. Especially when encountering obstacles (walls or sundries) blocking during the layout process, the robot cannot accurately complete the layout task. The presence of obstacles not only causes the robot to lose the positioning signal but may even lead to layout deviation and inability to complete high-precision layout operations in complex environments. Summary of the Invention

[0004] To solve the above problems, the present invention provides a positioning blind area correction control system for an automatic layout robot. Through the collaborative work of a tracking device, a layout robot, and a local area network control module, it realizes the automatic detection and mapping of obstacle areas, and corrects the layout path through an algorithm to ensure that the robot can still complete precise layout operations in signal occlusion or blind area environments.

[0005] To achieve the above object, the technical solution adopted by the present invention is:

[0006] A positioning blind area correction control system for an automatic layout robot, comprising: a tracking device, a layout robot, and a local area network control module. The tracking device and the layout robot are respectively communicatively connected to the local area network control module. The local area network control module includes an obstacle mapping unit, a blind area processing unit, and a layout control unit;

[0007] The tracking device is used for emitting laser signals point by point on the actual ground according to the layout coordinates pre-marked on the digital drawing;

[0008] The layout robot is used for moving point by point and performing inkjet layout by receiving laser signals and control signals from the local area network control module;

[0009] The obstacle mapping unit is used to determine the positioning blind area according to the lost position and recovery position of the laser signal when the wire laying robot moves while receiving the laser signal, control the wire laying robot to bypass the obstacle at a preset constant distance and record the bypass coordinate data, draw the obstacle contour according to the bypass coordinate data, and map the obstacle on the digital drawing;

[0010] The blind area processing unit is used to correct the wire laying coordinates of the digital drawing after the obstacle mapping is completed to obtain the execution drawing;

[0011] The wire laying control module includes a receiving area wire laying unit and a blind area wire laying unit. The receiving area wire laying unit is used to control the wire laying robot to perform inkjet wire laying when the wire laying robot can receive the laser signal; the blind area wire laying unit is used to control the wire laying robot to perform wire laying operation in the positioning blind area according to the execution drawing when the wire laying robot cannot receive the laser signal.

[0012] Further, the tracking device includes a tracking device body and a laser transmitter, an image acquisition device, a calibration unit, and a moving unit arranged on the device body;

[0013] The laser transmitter is used to receive the digital drawing and emit laser signals point by point on the real scene ground according to the pre-marked wire laying coordinates on the digital drawing;

[0014] The image acquisition device is used to acquire the image data of the current orientation of the tracking device;

[0015] The calibration unit is used to compare and verify the image data of the current orientation with the digital drawing based on the convolutional neural network, calculate the coincidence degree of the current orientation of the tracking device and the trajectory of the marked tracking device in the digital drawing, and issue a calibration reminder according to the orientation coincidence degree;

[0016] The moving unit is used to control the device body to move on a preset track.

[0017] Further, the wire laying robot includes a carrier and a laser signal receiving device, a ranging sensor, an inkjet device, a positioning device, and a moving device arranged on the carrier.

[0018] Further, the recording of the lost position and recovery position of the laser signal when the wire laying robot moves while receiving the laser signal includes:

[0019] When the wire laying robot loses the laser signal, obtain the lost position of the laser signal, make the wire laying robot avoid obstacles by keeping a distance from the obstacle through the ranging sensor, and obtain the recovery position of the laser signal.

[0020] Further, the determination of the positioning blind area according to the lost position and recovery position of the laser signal includes:

[0021] Generate lost coordinates and recovery coordinates according to the lost position and recovery position of the laser signal respectively;

[0022] Connect the lost coordinates, recovery coordinates and the tracking device respectively to obtain a closed area, and record the closed area as the positioning blind area.

[0023] Furthermore, the control of the wire-laying robot to bypass an obstacle at a preset constant distance and record the bypass coordinate data, and draw the contour of the obstacle according to the bypass coordinate data and perform obstacle mapping on the digital drawing includes the following steps:

[0024] Control the wire-laying robot to move towards the positioning blind area, and obtain distance parameters in real time through a distance sensor;

[0025] Preset a constant distance standard value. When the distance parameter reaches the preset constant distance standard value, control the wire-laying robot to bypass the obstacle at the preset constant distance standard value;

[0026] Record the position coordinates of the wire-laying robot in real time to generate a bypass coordinate data sequence;

[0027] According to the bypass coordinate data sequence, draw the contour line of the obstacle through a curve fitting algorithm;

[0028] Uniformly extract several point coordinates from the contour line of the obstacle, and the extraction quantity of the point coordinates is in geometric proportion to the total length of the contour line of the obstacle;

[0029] Perform obstacle mapping marking on the digital drawing based on the extracted several point coordinates.

[0030] Furthermore, the wire-laying coordinate correction for the digital drawing with obstacle mapping completed includes logically deleting the pre-labeled wire-laying coordinates that coincide with the obstacle in the digital drawing after performing obstacle mapping marking on the digital drawing.

[0031] Furthermore, the operation of controlling the wire-laying robot to perform wire-laying in the positioning blind area according to the execution drawing when the wire-laying robot cannot receive the laser signal includes:

[0032] S1. Select the entry point into the positioning blind area with the fewest turning times according to the wire-laying coordinates of the next inkjet for wire-laying;

[0033] S2. Control the wire-laying robot to reach the entry point and perform positioning calibration;

[0034] S3. Control the wire-laying robot to enter the positioning blind area from the entry point and perform wire-laying inkjet according to the wire-laying coordinates in the execution drawing;

[0035] S4. Obtain the current position of the wire laying robot in real time, and use the Kalman filtering algorithm to predict and correct the current position information;

[0036] S4. After each straight wire laying and inkjetting in the positioning blind area, control the wire laying robot to pause wire laying and inkjetting and leave the positioning blind area, and re - execute S1.

[0037] Further, the S1 includes the following steps:

[0038] Calculate the feasible entry points of each positioning blind area through the Dijkstra algorithm, and perform weight evaluation based on the number of turns and path length parameters;

[0039] According to the weight evaluation results, select the positioning blind area entry point with the fewest turns and the shortest path.

[0040] The beneficial effects of the present invention are as follows: The present invention realizes the determination of the contour of obstacles through the obstacle mapping unit. When the wire-laying robot receives a laser signal during operation, if the signal is lost due to an obstacle, the signal loss position and the recovery position are recorded. The obstacle mapping unit uses these two positions to determine the preliminary area of the obstacle by connecting the lines. To more accurately determine the contour of the obstacle, the system controls the wire-laying robot to move around the obstacle at a preset constant distance. During the circumvention process, the robot collects the coordinate data around the obstacle in real time through the distance measurement sensor. These data are used to accurately draw the contour line of the obstacle and map it onto the digital drawing. By obtaining the peripheral data points of the obstacle, the system can draw the true shape and size of the obstacle. After the obstacle contour is determined, the blind area processing unit further ensures the operation accuracy of the wire-laying robot in the signal blind area. For those signal blind areas formed by complete occlusion of the obstacle, the system corrects the wire-laying coordinates in the digital drawing according to the previously obtained obstacle contour. This ensures that the wire-laying robot can continue to perform precise wire-laying tasks in the blind area where the laser signal cannot be received based on the corrected drawing. Through this step, the robot no longer relies on external laser signals but performs wire-laying operations according to the corrected coordinates, correcting the wire-laying error in the signal blind area. By setting the Dijkstra algorithm and combining the weighted calculation of the number of turns and the path length parameters, the system can generate the next positioning blind area entry point with the least error for the wire-laying robot after the obstacle mapping is completed, ensuring that it maintains the shortest path and the least number of turns while performing wire-laying inkjet in the positioning blind area, greatly reducing the error superposition. After each wire-laying operation in the blind area or the obstacle circumvention area is completed, the wire-laying robot returns to the predetermined calibration point, and by obtaining the current position information and comparing it with the original coordinate system, the deviation between the current position of the robot and the ideal position is confirmed. Through this calibration step, the system can correct the possible error accumulation after long-term operation, ensuring the accuracy of subsequent wire-laying. For example, when the robot returns to the calibration point after completing the obstacle circumvention, it will realign based on the previously collected coordinate points with the global coordinate system, and adjust the deviation through the Kalman filtering algorithm to ensure that the wire-laying operation continuously maintains high precision. This calibration function is particularly important because in operations over long distances or in complex terrains, any small deviation will accumulate and affect the overall wire-laying accuracy. Therefore, calibration after each key node can effectively reduce errors and ensure the consistency and accuracy of the entire wire-laying process. Description of the Drawings

[0041] Figure 1 It is a schematic structural diagram of a positioning blind area deviation correction control system for an automatic wire-laying robot in the present invention.

[0042] Figure 2It is a flowchart of the steps for controlling the wire laying robot to perform wire laying operations in the positioning blind area according to the execution drawing when the wire laying robot cannot receive the laser signal in the present invention. Detailed implementation manner

[0043] Please refer to Figure 1-2 As shown in the figure, the positioning blind area correction control system of the automatic wire laying robot of the present invention includes: a tracking device, a wire laying robot, and a local area network control module. The tracking device and the wire laying robot are respectively communicatively connected to the local area network control module. The local area network control module includes an obstacle mapping unit, a blind area processing unit, and a wire laying control unit;

[0044] The tracking device is used to emit laser signals point by point on the actual ground according to the wire laying coordinates pre-marked on the digital drawing;

[0045] The wire laying robot is used to move point by point and perform inkjet wire laying by receiving laser signals and control signals of the local area network control module;

[0046] The obstacle mapping unit is used to determine the positioning blind area according to the lost position and restored position of the laser signal when the wire laying robot moves by receiving the laser signal, control the wire laying robot to bypass the obstacle at a preset constant distance and record the bypass coordinate data, draw the obstacle contour according to the bypass coordinate data, and perform obstacle mapping on the digital drawing;

[0047] The blind area processing unit is used to correct the wire laying coordinates of the digital drawing after completing the obstacle mapping to obtain the execution drawing;

[0048] The wire laying control module includes a receiving area wire laying unit and a blind area wire laying unit. The receiving area wire laying unit is used to control the wire laying robot to perform inkjet wire laying when the wire laying robot can receive the laser signal; the blind area wire laying unit is used to control the wire laying robot to perform wire laying operations in the positioning blind area according to the execution drawing when the wire laying robot cannot receive the laser signal.

[0049] In some embodiments, the tracking device is the core device in the system responsible for laser guiding the wire-laying robot. The device includes, but is not limited to, a total station or a three-coordinate measuring device integrated with a high-power laser emitter, which projects the coordinate points preset in the digital drawing onto the actual ground one by one. The laser emitter can adopt an industrial-grade laser emission device, which can ensure the stability and accuracy of the laser points in a large range. The high-resolution image sensor built in the device can capture environmental image data in real time and compare it with the digital drawing in the local area network control module. The convolutional neural network algorithm built in the system performs image processing in the verification unit to verify whether the coordinate points emitted by the laser emitter coincide with the preset coordinates, ensuring that the wire-laying robot always moves along the accurate preset trajectory. The wire-laying robot is the execution body in the whole system, mainly responsible for receiving the laser signal emitted by the tracking device and performing wire-laying operations according to the instructions issued by the local area network control module. The robot is equipped with a highly sensitive laser receiving device, which can stably receive the laser guiding signal under different light conditions. To ensure the accuracy of the wire-laying path, the robot is built in with an inertial measurement unit (IMU), including a multi-sensor fusion system such as an accelerometer and a gyroscope, which ensures that the robot does not drift during movement through real-time data correction. The wire-laying robot is also equipped with an inkjet device, and the inkjet module adopts industrial-grade micro-droplet spraying technology, which can ensure the clarity and consistency of the wire-laying lines. During movement, the robot controls the movement path through a servo drive system, and this drive system combines multi-wheel drive and suspension damping devices to ensure that the robot can operate smoothly under different ground conditions. The ranging sensor of the wire-laying robot is used to accurately detect the distance between the robot and the obstacle. This ranging sensor is based on lidar (LiDAR) technology and has millimeter-level accuracy, which can real-time feedback the relative position and distance information of the surrounding obstacles.

[0050] Further, the tracking device includes a tracking device body and a laser emitter, an image acquisition device, a verification unit, and a moving unit provided on the device body;

[0051] The laser emitter is used to receive the digital drawing and emit laser signals point by point on the actual ground according to the wire-laying coordinates pre-marked in the digital drawing;

[0052] The image acquisition device is used to acquire image data of the current orientation of the tracking device;

[0053] The verification unit is used to compare and verify the image data of the current orientation with the digital drawing based on the convolutional neural network, calculate the trajectory coincidence degree between the current orientation of the tracking device and the marked tracking device in the digital drawing, and issue a verification reminder according to the orientation coincidence degree;

[0054] The moving unit is used to control the device body to move on a preset track.

[0055] It should be noted that the image acquisition device of the tracking device captures the ground images in the current environment in real time. These images contain the actual terrain and construction details of the area where the device is located. These image data are input into a convolutional neural network for processing. The working process of the convolutional neural network is divided into multiple steps: Convolutional layer: First, the input ground images pass through several convolutional layers. Each convolutional layer performs a convolution operation on the image through a set of learned filters (kernels). The formula for the convolution operation is as follows:

[0056]

[0057] where y(i, j, k) represents the value of the k-th feature map at position (i, j); m and n are index variables in the convolution operation; x(i + m, j + n) represents the pixel value of the input image at position (i + m, j + n); w(m, n, k) represents the weight matrix of the convolution kernel in the k-th feature map; b(k) is the bias of the convolution kernel; M and N are the sizes of the convolution kernel.

[0058] Extract low-level features such as edges, lines, and textures. These features are the basic information required to locate the setting-out coordinates, helping the system identify key areas in complex environments. Activation function: After the convolutional layer, the system applies a non-linear activation function (usually the ReLU function) to enhance the non-linear expression ability of the features in the image. This enables the network to recognize more complex geometric shapes and structures and improves the perception of the setting-out coordinate positions in complex construction scenarios. Pooling layer: To reduce the data dimension and computational complexity, the CNN introduces a pooling layer (such as max pooling or average pooling) after the convolutional layer. The pooling layer downsamples the convolutional results, retaining the important features in the image while reducing unnecessary details. This step helps filter out irrelevant noise information in the scene and enhances the attention to the setting-out path. After the convolutional and pooling operations, the extracted high-dimensional feature data is integrated into a fixed-length feature vector through the fully connected layer. This vector is used to describe the spatial distribution and features of the key points in the image, which is crucial for comparing the setting-out coordinates on the actual ground and the digital drawing. Finally, the output layer of the network matches the calculated features with the preset coordinates in the digital drawing. Specifically, the CNN determines the coincidence degree between the identified coordinate points in the current image and the calibrated coordinate points on the digital drawing by calculating the cosine similarity or Euclidean distance metric with the annotation data in the digital drawing. The tracking device is set on one side of the area to be set out. According to the pre-annotated setting-out coordinates in the digital drawing, it emits laser signals point by point to the actual construction area. The control of the laser emitter is achieved through close integration with the digital drawing. The system emits laser signals to the ground one by one in the preset order according to the coordinate points calibrated in the drawing. These laser signals form multiple reference points in space, and the setting-out robot completes the precise guidance of the path by receiving the feedback of these points. The laser emitter adopts a servo control system, which combines closed-loop control technology and real-time position feedback. Through angle sensors and gyroscopes, it adjusts the horizontal and vertical angles of the emitter in real time to ensure that the emitted laser beam is always accurately aligned with the preset coordinate points. The servo control system is adjusted by the PID control algorithm. The system dynamically adjusts the deviation of the laser emission angle according to the position feedback signal, and finally realizes the high-precision projection of the laser signal.

[0059] Further, the setting-out robot includes a carrier and a laser signal receiving device, a ranging sensor, an inkjet device, a positioning device, and a moving device provided on the carrier.

[0060] It should be noted that, first of all, the carrier is the physical basis of the wire-laying robot, which is usually made of high-strength lightweight materials (such as aluminum alloy or carbon fiber composite materials) to ensure that the wire-laying robot has a light weight while maintaining structural stability, facilitating its flexible movement. The design of the carrier takes into account the complexity of the construction site, and has high impact resistance and environmental corrosion resistance to ensure stable operation in harsh outdoor environments. In addition, a multi-axis suspension system is installed at the bottom of the carrier to reduce the impact of terrain undulations on the balance of the wire-laying robot and ensure that the robot can move stably on uneven ground. The laser signal receiving device is a key component of the wire-laying robot for receiving laser signals emitted by the tracking device. The device includes a highly sensitive photodetector and a multi-band filter, which can accurately receive and analyze the laser signal transmitted by the laser transmitter. The photodetector captures the direction of the laser signal through a high-precision sensing unit, and transmits the received signal to the central control module for processing through a real-time feedback loop. The multi-band filter can effectively filter out interfering light sources in the environment to ensure stable reception of laser signals in strong or low light environments. In order to enhance the receiving accuracy, the receiving device is also equipped with an automatic aperture adjustment, which adjusts the amount of light entering the detector by controlling the aperture size to ensure that the received signal strength is moderate and not saturated or distorted. The distance sensor is the core module used in the wire-laying robot to measure the distance to obstacles or the ground in real time, usually using laser radar (LiDAR) technology. The sensor accurately measures the distance to the object by emitting laser pulses to the target object and calculating the time required for the laser to return. The high-precision distance sensor can provide real-time feedback on the relative distance information between the robot and the surrounding obstacles at the millimeter level, ensuring the safe operation of the robot in complex construction environments. The sensor adopts a 360-degree omnidirectional scanning design, which can continuously monitor the surrounding environment and calculate the best detour path in real time through the built-in obstacle detection algorithm, ensuring that the robot can avoid obstacles while maintaining the preset wire-laying path. In addition, the distance sensor can also work in conjunction with the ground distance keeper to ensure that the robot always maintains a constant height with the ground during the wire-laying process, preventing wire-laying errors caused by undulating terrain.

[0061] Furthermore, the recording and placing robot receives the laser signal and moves to obtain the lost position and the restored position of the laser signal, including:

[0062] When the wire-laying robot loses the laser signal, the position where the laser signal is lost is obtained, and the wire-laying robot maintains a distance from the obstacle to avoid the obstacle through the distance measuring sensor, and obtains the recovery position of the laser signal.

[0063] In some embodiments, an early scan of obstacles is first performed, that is, the wire-laying robot is made to move in a simulated wire-laying manner according to a preset digital drawing. When the laser signal received by the wire-laying robot is lost, the algorithm module of the system will immediately start the positioning and processing of the lost position. The recording of the lost position depends on the built-in positioning device of the robot and the sensor data fusion algorithm. Specifically, the system first performs position prediction through the Kalman Filter algorithm in combination with the historical data of the inertial measurement unit (IMU) and the laser signal receiving device. The Kalman Filter can model the current motion state (such as speed, acceleration, angular velocity, etc.) of the wire-laying robot. Even if the laser signal is lost, the approximate current position of the robot can be obtained through prediction. After recording the loss of the laser signal, the system performs environmental perception through a ranging sensor and activates the obstacle avoidance mechanism. The ranging sensor usually uses a lidar (LiDAR) or an ultrasonic sensor. These sensors can detect the distance and position of surrounding obstacles in real time when the laser signal is lost. The obstacle avoidance algorithm uses path planning based on the A* algorithm. The A* algorithm calculates an optimal detour path through the known data of surrounding obstacles. This algorithm can effectively evaluate the spatial distribution of obstacles and, combined with the distance data fed back by the ranging sensor, plan the movement path that the wire-laying robot should take. The advantage of the A* algorithm is that through heuristic search, it can quickly calculate the shortest path and achieve efficient obstacle avoidance in a complex environment. During the obstacle avoidance process, the robot also performs motion control based on PID control according to the preset safety distance and real-time ranging data. The PID control continuously adjusts the moving speed and direction of the robot to ensure that the robot can move along the planned detour path and always maintain a safe distance from the obstacles. The PID controller calculates the error in real time and performs feedback control. By adjusting the steering angle and speed, the robot can smoothly bypass the obstacles and avoid collisions. During the obstacle avoidance process, the robot continuously monitors the surrounding environmental signals, especially the recovery situation of the laser signal. Once the laser signal is restored, the system immediately records the signal restoration position. When the laser signal is restored and the restoration position is recorded, the system compares the current position information of the robot with the preset wire-laying trajectory. This process evaluates the coincidence degree between the current position of the robot and the trajectory through cosine similarity calculation or Euclidean distance measurement. Based on these measurement results, the system will re-plan the path through a path correction algorithm to ensure that the robot returns to the preset trajectory.

[0064] Further, determining the positioning blind area according to the lost position and the restored position of the laser signal includes:

[0065] Generating a lost coordinate and a restored coordinate according to the lost position and the restored position of the laser signal respectively;

[0066] Connect the lost coordinates, the restored coordinates, and the tracking device respectively to obtain a closed area, and record the closed area as a positioning blind area.

[0067] In some embodiments, first, when the laser signal is lost, the system records the position information of the robot at that moment through a built-in positioning module (such as an inertial measurement unit IMU) and marks it as lost coordinates. The generation of lost coordinates depends on the Kalman filtering algorithm. The Kalman filter fuses various data such as acceleration and angular velocity collected by the sensor, filtering out noise information to ensure the high precision of the lost coordinates. In the state of lost signal, the robot continues to perform short-distance movement through inertial navigation until the laser signal is restored. When the laser signal is restored, the system also records the position at this time through sensor data and marks it as restored coordinates. To ensure the accuracy of the lost coordinates and the restored coordinates, the system combines GPS data and the motion trajectory provided by the IMU, and further corrects the errors of the two sets of coordinates through a data fusion algorithm. Especially in the case where the robot is affected by ground undulations or slight obstacles during the laser signal loss, the data fusion algorithm can greatly reduce the error accumulation. Once the lost coordinates and the restored coordinates are recorded, the system then generates a triangular closed area by connecting these two coordinates with the fixed position of the tracking device through geometric analysis and vector calculation. Specifically, the system first calculates the straight-line distances from the lost coordinates and the restored coordinates to the tracking device respectively, which can be achieved through the Euclidean distance formula. In a two-dimensional or three-dimensional space, this formula calculates the straight-line distance between two points by taking the square root of the sum of the squares of the coordinate differences. After the generation of the closed area, the system further refines the shape and area of the area through a plane geometry analysis algorithm. During this process, the system converts the connection lines of the lost coordinates, the restored coordinates, and the tracking device into a closed triangle or polygon, and calculates the actual area of the area through the triangle area formula or the vertex coordinate formula of the polygon. After the calculation of the closed area, it is recognized as a positioning blind area. After determining the positioning blind area, the system integrates the blind area data with the digital drawing and generates corresponding mapping coordinates, so that the subsequent cable-laying robot can optimize and correct the path in the signal-blocked area.

[0068] Further, the steps of controlling the cable-laying robot to bypass an obstacle at a preset constant distance and record the bypass coordinate data, and drawing the obstacle contour according to the bypass coordinate data and performing obstacle mapping on the digital drawing include the following:

[0069] Control the cable-laying robot to move towards the positioning blind area, and obtain distance parameters in real time through a distance sensor;

[0070] Preset a constant distance standard value. When the distance parameter reaches the preset constant distance standard value, control the cable-laying robot to bypass the obstacle while maintaining the preset constant distance standard value;

[0071] Record the position coordinates of the wire-laying robot in real time to generate a sequence of bypass coordinate data;

[0072] According to the sequence of bypass coordinate data, draw the contour line of the obstacle through the curve fitting algorithm;

[0073] Uniformly extract several point coordinates from the contour line of the obstacle, and the extraction quantity of the point coordinates is in geometric proportion to the total length of the contour line of the obstacle;

[0074] Perform obstacle mapping and marking on the digital drawing based on the extracted several point coordinates.

[0075] In some embodiments, when the wire-laying robot enters the positioning blind area, the system controls it to approach the obstacle through instructions. During this process, the distance sensors built into the robot, such as LiDAR (Light Detection and Ranging) or ultrasonic sensors, continuously obtain the distance parameters between the robot and the obstacle. The distance sensors have a measurement accuracy of millimeter level, and can accurately feedback the distance while the robot approaches the obstacle. To ensure that the robot maintains a fixed safe distance from the obstacle during the entire bypass process, the system continuously adjusts the position of the robot through the PID control algorithm (Proportional-Integral-Derivative control algorithm). PID control can dynamically adjust the movement path and speed of the robot according to the real-time distance data provided by the sensor, ensuring that the robot decelerates when approaching the obstacle and precisely reaches the preset constant distance standard value. Once the distance sensor detects that the distance between the robot and the obstacle reaches this preset value, the system immediately executes the bypass operation. Through the real-time feedback mechanism of PID control, the wire-laying robot can bypass along the outer periphery of the obstacle at a constant distance. To ensure the accuracy of the robot's path in a complex environment, during the bypass process, the robot also combines the data of the Inertial Measurement Unit (IMU) and further corrects its position through the Kalman filtering algorithm to ensure that the movement trajectory during the bypass process is not affected by external interference. While the robot is bypassing the obstacle, the system will record the position coordinates of the robot in real time during the bypass process. These coordinate data are obtained by the multi-sensor fusion system built into the robot, mainly including the combination of GPS, IMU, and LiDAR. Through the multi-source data fusion algorithm, the high accuracy and consistency of the bypass coordinate data are ensured. The system stores these coordinate data as a bypass coordinate data sequence, which contains the precise position of the robot at each moment during the bypass process. Next, the system will process these bypass coordinate data through the curve fitting algorithm to draw the contour line of the obstacle. The curve fitting algorithm usually uses polynomial fitting or spline curve fitting methods. These algorithms can generate smooth curves based on a series of discrete coordinate points, accurately reflecting the shape of the obstacle. Polynomial fitting calculates the fitting of the coordinate points through the least squares method to obtain the equation describing the contour of the obstacle. Spline curve fitting is applicable to more complex and irregular obstacle shapes. By splicing multiple quadratic or cubic polynomial segments, a continuous curve is generated to ensure the accuracy of the obstacle contour. When the contour line is drawn, the system will uniformly extract several point coordinates from it. The extraction method of these point coordinates is based on the total length of the obstacle contour line and is distributed proportionally at different positions on the contour line. The number of point coordinates is proportional to the total length of the contour line to ensure that more points are extracted on the longer contour segments and fewer points are extracted on the shorter segments. These evenly distributed points ensure that the shape of the obstacle is presented more precisely in the digital drawing, avoiding the contour blur or incompleteness caused by the lack of coordinate points in some local areas.Finally, the system maps these extracted point coordinates onto the digital drawing to complete the mapping and marking of obstacles. The obstacle mapping and marking are quickly stored and retrieved through a spatial indexing algorithm. The system uses these marks as important references for the construction line laying path planning to ensure that the robot can avoid obstacles and accurately execute the subsequent line laying tasks.

[0076] Further, the alignment coordinate correction for the digital drawing with completed obstacle mapping includes, after performing obstacle mapping and marking on the digital drawing, logically deleting the pre-marked alignment coordinates that coincide with the obstacles in the digital drawing.

[0077] In some embodiments, first, when the system completes the mapping of obstacles and marks the obstacles on the digital drawing, the system detects the coincidence degree between the pre-marked alignment coordinates and the obstacle marking area through a spatial intersection algorithm. This algorithm is based on geometric calculations and performs spatial analysis on the alignment coordinates and the obstacle contour to determine whether the line laying path conflicts with the obstacles. The spatial intersection algorithm calculates the geometric relationship between the coordinate points of the obstacle contour line and the pre-marked alignment coordinates. For example, using a polygon intersection detection algorithm, it detects whether the alignment coordinates are located in the inner area of the obstacle or intersect with the obstacle boundary. Once it detects that the alignment coordinates coincide with the obstacles, the system activates a logical deletion mechanism to process the pre-marked alignment coordinates that coincide with the obstacles. This process first uses a point-in-polygon detection algorithm (Point-in-Polygon Algorithm) to identify which alignment coordinates fall on or inside the boundary of the obstacle. This algorithm is based on the ray method or the angle sum method to determine whether a point is inside a polygon. If it is determined that some alignment coordinate points are located within the obstacle area, the system marks these points as invalid coordinate points. When performing logical deletion, the system does not directly remove these coordinate points from the drawing, but logically deletes them through a marking and indexing mechanism. Logical deletion sets a validity flag for each alignment coordinate point. For the coordinate points that coincide with the obstacles, their flag will be set to the "invalid" state. The purpose of this is to retain the complete original data and ensure that the original coordinate information is still available in any case of recovery or recalculation. This logical deletion method quickly marks and retrieves the coordinate points through an indexing mechanism, thus ensuring the structural integrity of the entire digital drawing.

[0078] Further, the operation of controlling the line laying robot to perform line laying operations in the positioning blind area according to the execution drawing when the line laying robot cannot receive laser signals includes:

[0079] S1. Select the entry point in the positioning blind area with the fewest number of turning times according to the alignment coordinates of the next inkjet for line laying.

[0080] S2. Control the wire laying robot to reach the entry point and perform positioning calibration;

[0081] S3. Control the wire laying robot to enter the positioning blind area from the entry point and perform wire laying and inkjet according to the wire laying coordinates in the execution drawing;

[0082] S4. Real-time obtain the current position of the wire laying robot, and use the Kalman filter algorithm to predict and correct the current position information;

[0083] S4. After each straight wire laying and inkjet is completed in the positioning blind area, control the wire laying robot to pause wire laying and inkjet and leave the positioning blind area, and re-execute S1.

[0084] In some embodiments, first, in step S1, the system selects the best point to enter the positioning blind area according to the coordinates of the next inkjet wire to be released. The selection of this point is based on the principle of the least number of turns, ensuring that the robot reduces unnecessary turns in path planning to improve the movement efficiency. In specific implementation, the system uses the Dijkstra algorithm or the A* algorithm for path planning. These algorithms calculate the lengths and the number of turns of all feasible paths, and take the path with the minimum comprehensive cost as the preferred path for the entry point. To optimize the calculation process, the algorithms will combine the coordinate information of the obstacles to ensure that the path planning avoids the influence area of the obstacles. In addition, the optimization of the number of turns will further reduce unnecessary complex paths through heuristic search. Next, step S2 involves the wire-releasing robot reaching the entry point and performing positioning calibration. To ensure accurate positioning when reaching the entry point, the system combines the Kalman filtering algorithm to predict and correct the motion data collected by the robot in real time. The Kalman filter estimates the position, speed, and acceleration of the robot at consecutive time steps, and removes the errors caused by sensor noise through continuous iterative updates, thereby improving the positioning accuracy. When the robot reaches the preset entry point, the system will calibrate the current position to ensure that the robot is in an accurate initial state before entering the blind area. In step S3, the robot enters the positioning blind area from the entry point and performs inkjet wire-releasing operations according to the wire-releasing coordinates in the execution drawing. This process depends on the execution drawing corrected after the previous obstacle mapping to ensure that the robot can perform the wire-releasing task along the preset path according to the calibrated coordinates. Since the laser signal is unavailable, the robot needs to rely on path tracking algorithms to move in the positioning blind area. Common path tracking algorithms include the Pure Pursuit algorithm, which ensures that the robot can travel along the optimal path in the blind area by continuously adjusting the direction and speed of the robot relative to the target coordinates. When the robot travels in the blind area, it will precisely control the inkjet device according to the position information of each coordinate point to ensure the continuity and accuracy of the wire-releasing. After each straight-line wire-releasing inkjet task is completed in the positioning blind area, step S4 will also control the robot to pause the wire-releasing inkjet and leave the positioning blind area. The path planning for leaving the blind area also depends on the previous path tracking algorithm and the inertial navigation system to ensure that the robot can smoothly exit the blind area. During this process, the system will continuously monitor the motion state of the robot to ensure that it can re-acquire the laser signal or calibration information after exiting the blind area. After completion of the exit, the system will execute step S1 again to select the next entry point to ensure the smooth progress of subsequent wire-releasing tasks.

[0085] Further, S1 includes the following steps:

[0086] Calculate the feasible entry points of each positioning blind area through the Dijkstra algorithm, and perform weight evaluation based on the number of turns and path length parameters;

[0087] Based on the weight evaluation results, select the positioning blind area entry point with the fewest turns and the shortest path.

[0088] Specifically, the formula of Dijkstra's algorithm is as follows:

[0089] W total = W L ·Path Length + W T ·Turn Count;

[0090] Among them, W total This is the total weight value, representing the comprehensive score for evaluating each path; W L Path length weight, which is a weight coefficient used to measure the impact of path length on the total weight; Path Length is the path length; W T is the turn count weight. The more turns there are, the greater the complexity of the robot's travel path and the possibility of error accumulation. Therefore, in some scenarios, a higher weight may be assigned; Turn Count is the number of turns, referring to the number of turns required for the wire-laying robot on the path from the current position to the positioning blind area entry point.

[0091] The system calculates the feasible entry points for each positioning blind area through the Dijkstra algorithm. The Dijkstra algorithm is a classic shortest path algorithm that can effectively handle the shortest path problem in a graph structure. In this embodiment, the positioning blind area and the current position of the robot form a graph structure, and each potential entry point in the blind area is regarded as a node in the graph. The Dijkstra algorithm starts from the current position of the wire-laying robot and calculates the path length to each entry point one by one, and assigns a shortest path weight to each path. The main advantage of the Dijkstra algorithm is that it can traverse all possible paths and ensure finding the route with the shortest path length. This step provides the basic data for selecting the entry points. Next, the system evaluates the weights of all paths. The criteria for weight evaluation include not only the path length but also the number of turns. The number of turns is an important factor affecting the movement efficiency of the robot. Frequent turning will increase energy consumption, extend the movement time, and even increase the error of the wire-laying path. Therefore, the weight evaluation will comprehensively consider these two factors of path length and number of turns. During the weight calculation process, the system assigns certain weight values to the path length and the number of turns respectively. Among them, the path length is directly obtained from the result of the Dijkstra algorithm, and the number of turns is calculated through the angle change of the path. Specifically, the system calculates the angle change of the movement direction of the robot on each section of the path. If the angle change between two sections of the path exceeds the set threshold, it is regarded as one turn. By accumulating the angle changes of the path, the total number of turns of the path is obtained. After the weight evaluation is completed, the system selects the optimal entry point according to the weight evaluation result. The selection criterion for the optimal entry point is the least number of turns and the shortest path. In practical applications, the system will preferentially select the path with the smallest weight to ensure that the robot enters the positioning blind area with the least number of turns and the shortest path. This can not only ensure the efficient entry of the robot into the blind area but also reduce the accumulation of path deviations caused by frequent turning.

[0092] The above embodiments are only descriptions of the preferred embodiments of the present invention and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A positioning blind area correction control system for an automatic wire-laying robot, characterized in that: include: A tracking device, a wire-laying robot and a local area network control module, wherein the tracking device and the wire-laying robot are respectively connected to the local area network control module for communication, and the local area network control module includes an obstacle mapping unit, a blind spot processing unit and a wire-laying control unit; The tracking device is used to transmit laser signals point by point to the real ground according to the layout coordinates pre-marked on the digital drawing; The wire-laying robot is used to move point by point and ink-lay out wires by receiving laser signals and control signals from a local area network control module; The obstacle mapping unit is used to record the loss position and recovery position of the laser signal when the wire-laying robot receives the laser signal and moves, determine the positioning blind area according to the loss position and recovery position of the laser signal, control the wire-laying robot to bypass the obstacle at a preset constant distance and record the bypass coordinate data, draw the obstacle outline according to the bypass coordinate data and perform obstacle mapping on the digital drawing; The blind spot processing unit is used to perform coordinate correction on the digital drawing that completes obstacle mapping to obtain an execution drawing; The wire-laying control unit includes a receiving area wire-laying unit and a blind area wire-laying unit. The receiving area wire-laying unit is used to control the wire-laying robot to perform inkjet wire-laying when the wire-laying robot can receive a laser signal; the blind area wire-laying unit is used to control the wire-laying robot to perform wire-laying operations within the positioning blind area according to the execution drawing when the wire-laying robot cannot receive a laser signal.

2. The positioning blind area correction control system of an automatic wire-laying robot according to claim 1 is characterized in that: The tracking device comprises a tracking device body and a laser emitter, an image acquisition device, a verification unit and a moving unit arranged on the device body; The laser transmitter is used to receive the digital drawing and transmit the laser signal point by point to the real ground according to the layout coordinates pre-marked on the digital drawing; The image acquisition device is used to acquire image data of the current position of the tracking device; The verification unit is used to compare and verify the image data of the current position with the digital drawing based on the convolutional neural network, calculate the overlap between the current position of the tracking device and the track of the tracking device marked in the digital drawing, and issue a verification reminder according to the calculation result; The moving unit is used to control the device body to move on a preset track.

3. The positioning blind area correction control system of an automatic wire-laying robot according to claim 1 is characterized in that: The wire-laying robot comprises a carrier and a laser signal receiving device, a distance measuring sensor, an inkjet device, a positioning device and a moving device which are arranged on the carrier.

4. The positioning blind area correction control system of an automatic wire-laying robot according to claim 3 is characterized in that: The recording and placing robot receives the laser signal and moves to determine the position where the laser signal is lost and restored, including: When the wire-laying robot loses the laser signal, the position where the laser signal is lost is obtained, and the wire-laying robot maintains a distance from the obstacle to avoid the obstacle through the distance measuring sensor, and obtains the recovery position of the laser signal.

5. The positioning blind area correction control system of an automatic wire-laying robot according to claim 4 is characterized in that: Determining the positioning blind area according to the loss position and the recovery position of the laser signal includes: Generate lost coordinates and restored coordinates according to the lost position and restored position of the laser signal respectively; The lost coordinates, restored coordinates and tracking devices are connected respectively to obtain a closed area, and the closed area is recorded as a positioning blind area.

6. The positioning blind area correction control system of an automatic wire-laying robot according to claim 5 is characterized in that: The controlling of the wire-laying robot to keep a preset constant distance from the obstacle and to detour and record detour coordinate data, drawing the obstacle outline according to the detour coordinate data and performing obstacle mapping on the digital drawing comprises the following steps: Controlling the wire-laying robot to move toward the positioning blind area, and obtaining distance parameters in real time through a distance sensor; A constant distance standard value is preset. If the distance parameter reaches the preset constant distance standard value, the wire-laying robot is controlled to maintain the preset constant distance standard value and bypass obstacles; Record the position coordinates of the wire-laying robot in real time and generate a detour coordinate data sequence; According to the detour coordinate data sequence, the outline of the obstacle is drawn through the curve fitting algorithm; Uniformly extracting a number of point coordinates from the contour line of the obstacle, wherein the number of the extracted point coordinates is proportional to the total length of the contour line of the obstacle; Obstacle mapping and marking are performed on the digital drawing based on the extracted coordinates of several points.

7. The positioning blind area correction control system of the automatic wire-laying robot according to claim 6 is characterized in that: The correction of the layout coordinates on the digital drawing after obstacle mapping includes logically deleting the pre-marked layout coordinates on the digital drawing that overlap with the obstacle after obstacle mapping is marked on the digital drawing.

8. The positioning blind area correction control system of an automatic wire-laying robot according to claim 1 is characterized in that: The controlling the wire-laying robot to perform a wire-laying operation in a positioning blind area according to the execution drawing when the wire-laying robot cannot receive a laser signal comprises: S1, according to the next inkjet laying coordinates to be laid out, select the positioning blind area entry point with the least number of turns; S2, controlling the wire-laying robot to reach the entry point and perform positioning calibration; S3, controlling the wire laying robot to enter the positioning blind area from the entry point and to lay out the wire and ink according to the wire laying coordinates in the execution drawing; S4, obtaining the current position of the wire-laying robot in real time, and using the Kalman filter algorithm to predict and correct the current position information; S4, after each straight line laying and inkjetting is completed in the positioning blind area, the laying robot is controlled to pause the laying and inkjetting and leave the positioning blind area, and S1 is executed again.

9. The positioning blind area correction control system of the automatic wire-laying robot according to claim 8 is characterized in that: The S1 comprises the following steps: The Dijkstra algorithm is used to calculate the feasible entry points of each positioning blind area, and a weight evaluation is performed based on the number of turns and path length parameters; According to the weight evaluation results, the positioning blind area entry point with the least number of turns and the shortest path is selected.

Images