Positioning blind area deviation rectification control system of automatic paying-off robot

Abstract

The invention relates to the technical field of construction automatic pay-off control, in particular to a positioning blind area deviation rectification control system of an automatic pay-off robot. When the paying-off robot encounters an obstacle and causes signal loss, the system records loss and recovery positions, and an obstacle area is determined through a connecting line. The robot keeps moving around an obstacle by a constant distance, and coordinate data are collected in real time and mapped to a digital drawing. The system processes a signal blind area which is completely shielded by an obstacle, ensures that the robot continues to execute accurate paying-off operation in the blind area by correcting paying-off coordinates in a drawing, and corrects errors in the blind area. The system optimizes blind area entry points by using a path optimization algorithm in combination with steering times and path length, and error superposition is reduced. After each time of obstacle avoiding or blind area paying-off, the robot returns to a calibration point, position deviation is adjusted in real time through a Kalman filtering algorithm, errors are corrected, the paying-off precision in blind area operation is improved, and deviation accumulation is reduced.

Description

Technical Field

[0001] The invention relates to the technical field of automatic line-laying control in construction, and in particular to a positioning blind area deviation correction control system of an automatic line-laying robot. Background Art

[0002] With the development of the construction industry and the complexity of large-scale project construction, the accuracy and efficiency of ground marking are becoming increasingly demanding. In the traditional construction process, manual marking is a common construction preparation work. However, manual marking is not only inefficient, but also easily leads to positioning errors and increased duplication of labor.

[0003] The existing technology uses a laser locator to receive a preset digital drawing and emit a laser point to the ground, so that the robot can track the emitted laser point to complete the laying of wires. However, the existing automatic laying system still faces some technical challenges, especially when encountering obstacles (walls or debris) during the laying process, the robot cannot accurately complete the laying task. The existence of obstacles will not only cause the robot to lose the positioning signal, but also cause the laying deviation, and it is impossible to complete the high-precision laying operation in a complex environment. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a positioning blind spot correction control system for an automatic wire-laying robot. Through the coordinated work of a tracking device, a wire-laying robot and a local area network control module, automatic detection and mapping of obstacle areas are realized, and the wire-laying path is corrected through an algorithm to ensure that the robot can still complete precise wire-laying operations in signal obstruction or blind spot 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 wire-laying robot, comprising: 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 comprises an obstacle mapping unit, a blind area processing unit and a wire-laying control unit;

[0007] 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;

[0008] 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;

[0009] 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;

[0010] The blind spot processing unit is used to perform coordinate correction on the digital drawing that completes obstacle mapping to obtain an 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 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.

[0012] Furthermore, the tracking device includes a tracking device body and a laser emitter, an image acquisition device, a verification unit and a moving unit arranged on the device body;

[0013] 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;

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

[0015] 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 position overlap;

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

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

[0018] 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:

[0019] 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.

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

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

[0022] 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.

[0023] Furthermore, the controlling of the wire-laying robot to detour the obstacle at a preset constant distance 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:

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

[0025] 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;

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

[0027] According to the detour coordinate data sequence, the outline of the obstacle is drawn through the curve fitting algorithm;

[0028] 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;

[0029] Obstacle mapping and marking are performed on the digital drawing based on the extracted coordinates of several points.

[0030] Furthermore, the setting-out coordinate correction of the digital drawing after obstacle mapping includes logically deleting the pre-marked setting-out coordinates in the digital drawing that overlap with the obstacle after obstacle mapping is marked on the digital drawing.

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

[0032] 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;

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

[0034] 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;

[0035] 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;

[0036] 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.

[0037] Furthermore, the S1 comprises the following steps:

[0038] 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;

[0039] According to the weight evaluation results, the positioning blind area entry point with the least number of turns and the shortest path is selected.

[0040] The beneficial effect of the present invention is that the present invention realizes the determination of the outline of the obstacle through the obstacle mapping unit. When the wire-laying robot receives the laser signal during the working process, if it encounters an obstacle and causes the signal to be lost, 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 lines. In order to more accurately determine the outline of the obstacle, the system controls the wire-laying robot to move around the obstacle at a preset constant distance. During the detour, the robot collects coordinate data around the obstacle in real time through a ranging sensor. These data are used to accurately draw the outline of the obstacle and map it to a 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 outline 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 obstruction of the obstacle, the system corrects the wire-laying coordinates in the digital drawing according to the obstacle outline previously obtained. This ensures that the wire-laying robot can continue to perform accurate 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 the laying-out operation according to the corrected coordinates, correcting the laying-out error in the signal blind area. By setting the Dijkstra algorithm and combining the weighted calculation of the number of turns and path length parameters, the system can generate the next positioning blind area entry point with the least error for the laying-out robot after the obstacle mapping is completed, ensuring that it maintains the shortest path and the least number of turns while performing laying-out inkjet in the positioning blind area, greatly reducing the error superposition. After each laying-out operation in the blind area or the obstacle bypassing area, the laying-out robot will return to the predetermined calibration point, obtain the current position information and compare it with the original coordinate system, and confirm the deviation between the current position of the robot and the ideal position. Through this calibration step, the system can correct the possible error accumulation after long-term operation and ensure the accuracy of subsequent laying-out. For example, after the robot completes the obstacle bypassing, it will realign with the global coordinate system based on the previously collected coordinate points when returning to the calibration point, and adjust the deviation through the Kalman filter algorithm to ensure that the laying-out operation continues to maintain high precision. This calibration function is particularly important because in operations over long distances or in complex terrain, any small deviations will accumulate and affect the overall accuracy of the layout. Therefore, calibration after each key node can effectively reduce errors and ensure the consistency and accuracy of the entire layout process. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0042] Figure 2It is a flow chart of the steps of controlling the wire-laying robot to perform wire-laying operations in a positioning blind area according to an execution drawing when the wire-laying robot cannot receive a laser signal in the present invention. DETAILED DESCRIPTION

[0043] See also Figure 1-2 As shown, the present invention relates to a positioning blind area correction control system for an automatic wire-laying robot, comprising: 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 comprises an obstacle mapping unit, a blind area processing unit and a wire-laying control unit;

[0044] 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;

[0045] 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;

[0046] 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;

[0047] The blind spot processing unit is used to perform coordinate correction on the digital drawing that completes obstacle mapping to obtain an 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 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.

[0049] In some embodiments, the tracking device is the core device in the system responsible for laser guidance of the wire-laying robot. The device includes but is not limited to projecting the pre-set coordinate points in the digital drawing onto the real ground one by one through a total station or a three-coordinate measuring device with an integrated high-power laser transmitter. The laser transmitter can adopt an industrial-grade laser transmitter, which can ensure the stability and accuracy of the laser point in a large range. The high-resolution image sensor built into 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 into the system performs image processing in the verification unit to verify whether the coordinate points emitted by the laser transmitter coincide with the preset coordinates, ensuring that the wire-laying robot always moves along the accurate preset trajectory. The wire-laying robot is the executive body in the entire 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 laser guidance signals under different light conditions. In order to ensure the accuracy of the laying-out path, the robot has a built-in inertial measurement unit (IMU), including a multi-sensor fusion system such as accelerometers and gyroscopes, which ensures that the robot does not drift during movement through real-time data correction. The laying-out robot is also equipped with an inkjet device. The inkjet module uses industrial-grade droplet injection technology to ensure the clarity and consistency of the laying-out lines. During the movement, the robot controls the movement path through a servo drive system. The drive system combines multi-wheel drive and suspension shock absorption devices to ensure that the robot can run smoothly under different ground conditions. The ranging sensor of the laying-out robot is used to accurately detect the distance between the robot and obstacles. The ranging sensor is based on laser radar (LiDAR) technology, has millimeter-level accuracy, and can provide real-time feedback on the relative position and distance information of surrounding obstacles.

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

[0051] 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;

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

[0053] 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 position overlap;

[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 equipment is located. These image data are input into the convolutional neural network for processing. The working process of the convolutional neural network is divided into multiple steps: Convolution layer: First, the input ground image passes through several convolution layers. Each convolution layer performs a convolution operation on the image through a set of learned filters (kernel). The formula for the convolution operation is as follows:

[0056]

[0057] Among them, y(i, j, k) represents the value of the kth 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 kth 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 needed to locate the coordinates of the line, helping the system to identify key areas in complex environments. Activation function: After the convolution layer, the system applies a nonlinear activation function (usually the ReLU function) to enhance the nonlinear expression of features in the image. This enables the network to recognize more complex geometric shapes and structures and improve the perception of the location of the line coordinates in complex construction scenes. Pooling layer: In order to reduce data dimensions and computational complexity, CNN introduces a pooling layer (such as maximum pooling or average pooling) after the convolution layer. The pooling layer retains important features in the image while reducing unnecessary details by downsampling the convolution results. This step helps filter out irrelevant noise information in the scene and enhances the focus on the line path. After the convolution and pooling operations, the extracted high-dimensional feature data is integrated into a fixed-length feature vector through a fully connected layer. This vector is used to describe the spatial distribution and characteristics of key points in the image, which is crucial for comparing the line coordinates on the real ground with the digital drawing. Finally, the output layer of the network matches the calculated features with the preset coordinates in the digital drawing. Specifically, CNN determines the degree of overlap between the coordinate points identified in the current image and the calibrated coordinate points on the digital drawing by calculating the cosine similarity or Euclidean distance measurement with the annotated data in the digital drawing. The tracking device is set on one side of the area to be laid out, and it emits laser signals point by point to the actual construction area according to the pre-marked laying out coordinates in the digital drawing. The control of the laser transmitter is realized through close integration with the digital drawing. The system emits laser signals to the ground one by one in a preset order according to the calibrated coordinate points in the drawing. These laser signals form multiple reference points in space, and the laying out robot completes the precise guidance of the path by receiving feedback from these points. The laser transmitter adopts a servo control system, which combines closed-loop control technology and real-time position feedback. Through angle sensors and gyroscopes, the horizontal and vertical angles of the transmitter are adjusted 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] Furthermore, the wire-laying robot includes a carrier and a laser signal receiving device, a distance measuring sensor, an inkjet device, a positioning device and a moving device arranged 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 advance scan of obstacles is first performed, that is, the wire-laying robot simulates wire-laying movement according to a preset digital drawing. When the wire-laying robot loses the received laser signal, the algorithm module of the system immediately starts the positioning and processing of the lost position. The recording of the lost position depends on the robot's built-in positioning device and sensor data fusion algorithm. Specifically, the system first predicts the position through the Kalman filter algorithm (Kalman Filter) combined 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 (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 perceives the environment through the ranging sensor and starts the obstacle avoidance mechanism. The ranging sensor usually uses a laser radar (LiDAR) or an ultrasonic sensor, which can detect the distance and position of surrounding obstacles in real time when the laser signal is lost. The obstacle avoidance algorithm uses a path planning based on the A* algorithm (A-Star Algorithm). The A* algorithm calculates an optimal detour path through known surrounding obstacle data. The algorithm can effectively evaluate the spatial distribution of obstacles, and plan the movement path that the wire-laying robot should take by combining the distance data fed back by the ranging sensor. The advantage of the A* algorithm is that it can quickly calculate the shortest path through heuristic search and achieve efficient obstacle avoidance in complex environments. 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. PID control continuously adjusts the robot's movement speed and direction to ensure that the robot can move along the planned detour path and always maintain a safe distance from 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 obstacles and avoid collisions. During the obstacle avoidance process, the robot will continuously monitor the surrounding environmental signals, especially the recovery of the laser signal. Once the laser signal is restored, the system will immediately record the signal recovery position. When the laser signal is restored and the recovery position is recorded, the system will compare the robot's current position information with the preset wire-laying trajectory. This process evaluates the overlap between the robot's current position and the trajectory through cosine similarity calculation or Euclidean distance measurement. Based on these measurements, the system will replan 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] Generate lost coordinates and restored coordinates according to the lost position and restored position of the laser signal respectively;

[0066] 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.

[0067] In some embodiments, first, when the laser signal is lost, the system records the position information of the robot at that moment through the built-in positioning module (such as inertial measurement unit IMU) and marks it as lost coordinates. The generation of lost coordinates depends on the Kalman filter algorithm. The Kalman filter fuses various data such as acceleration and angular velocity collected by the sensor, filters out noise information, and ensures the high accuracy 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. In order to ensure the accuracy of lost coordinates and restored coordinates, the system will combine GPS data and the motion trajectory provided by IMU, and further correct the errors of the two sets of coordinates through data fusion algorithm, especially when the robot is disturbed by ground undulations or slight obstacles during the period when the laser signal is lost. The data fusion algorithm can greatly reduce the error accumulation. Once the lost coordinates and the restored coordinates are recorded, the system then connects the two coordinates with the fixed position of the tracking device through geometric analysis and vector calculation to generate a triangular closed area. Specifically, the system first calculates the straight-line distances from the lost coordinates and the restored coordinates to the tracking device, which can be achieved through the Euclidean distance formula. In two-dimensional or three-dimensional space, this formula obtains the straight-line distance between two points by calculating the square and square root of the coordinate difference. After the closed area is generated, the system will further refine the shape and area of ​​the area through a plane geometry analysis algorithm. In this process, the system converts the lines connecting the lost coordinates, the restored coordinates and the tracking device into closed triangles or polygons, and calculates the actual area of ​​the area through the triangle area formula or the vertex coordinate formula of the polygon. After calculation, the closed area is identified as a positioning blind spot. After determining the positioning blind spot, the system will integrate the blind spot data with the digital drawing and generate the corresponding mapping coordinates so that the subsequent wire laying robot can optimize and correct the path in the area blocked by the signal.

[0068] Furthermore, the controlling of the wire-laying robot to detour the obstacle at a preset constant distance 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:

[0069] Controlling the wire-laying robot to move toward the positioning blind area, and obtaining distance parameters in real time through a distance sensor;

[0070] 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;

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

[0072] According to the detour coordinate data sequence, the outline of the obstacle is drawn through the curve fitting algorithm;

[0073] 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;

[0074] Obstacle mapping and marking are performed on the digital drawing based on the extracted coordinates of several points.

[0075] In some embodiments, when the wire-laying robot enters the positioning blind spot, the system controls it to approach the obstacle through instructions. During this process, the robot's built-in distance sensor, such as a laser radar (LiDAR) or an ultrasonic sensor, obtains the distance parameters between the robot and the obstacle in real time. The distance sensor has a millimeter-level measurement accuracy and can accurately feedback the distance while the robot approaches the obstacle. In order to ensure that the robot maintains a fixed safe distance from the obstacle during the entire detour, the system continuously adjusts the position of the robot through a PID control algorithm (proportional-integral-differential control algorithm). PID control can dynamically adjust the robot's moving path and speed based on the real-time distance data provided by the sensor, ensuring that the robot slows down when approaching the obstacle and accurately reaches the preset constant distance standard value. Once the distance sensor detects that the distance between the robot and the obstacle reaches the preset value, the system immediately performs a detour operation. Through the real-time feedback mechanism of PID control, the wire-laying robot can detour along the periphery of the obstacle at a constant distance. In order to ensure the accuracy of the robot's path in a complex environment, during the detour, the robot will also combine the data of the inertial measurement unit (IMU) and further correct the position through the Kalman filter algorithm to ensure that the motion trajectory during the detour is not subject to external interference. While the robot is bypassing obstacles, the system will record the position coordinates of the robot in real time. These coordinate data are obtained by the robot's built-in multi-sensor fusion system, which mainly includes a combination of GPS, IMU and laser radar. The multi-source data fusion algorithm ensures the high accuracy and consistency of the bypass coordinate data. 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 a curve fitting algorithm to draw the contour of the obstacle. Curve fitting algorithms usually use polynomial fitting or spline curve fitting methods. These algorithms can generate smooth curves based on a series of discrete coordinate points to accurately reflect the shape of the obstacle. Polynomial fitting uses the least squares method to fit the coordinate points and obtain an equation that describes the obstacle contour. Spline curve fitting is suitable for more complex and irregular obstacle shapes. It generates a continuous curve by splicing multiple quadratic or cubic polynomial fragments to ensure the accuracy of the obstacle contour. When the contour is drawn, the system will evenly extract several point coordinates from it. The point coordinates are extracted based on the total length of the obstacle contour and are proportionally distributed at different locations along the contour. The number of point coordinates is proportional to the total length of the contour to ensure that more points are extracted on longer contour segments and fewer points are extracted on shorter segments. These evenly distributed points ensure that the shape of the obstacle is presented more accurately in the digital drawing, avoiding blurry or incomplete contours in some local areas due to missing coordinate points.Finally, the system maps these extracted point coordinates onto digital drawings to complete the mapping of obstacles. Obstacle mapping marks are quickly stored and retrieved through a spatial indexing algorithm. The system uses these marks as an important reference for construction layout path planning, ensuring that the robot can avoid obstacles and accurately perform subsequent layout tasks.

[0076] Furthermore, the setting-out coordinate correction of the digital drawing after obstacle mapping includes logically deleting the pre-marked setting-out coordinates in the digital drawing that overlap with the obstacle after obstacle mapping is marked on the digital drawing.

[0077] In some embodiments, first, when the system completes the mapping of the obstacle and marks the obstacle on the digital drawing, the system will detect the overlap between the pre-marked set-up coordinates and the obstacle marking area through a spatial intersection algorithm. The algorithm is based on geometric calculations, and performs spatial analysis on the set-up coordinates and the obstacle contour to determine whether the set-up path conflicts with the obstacle. The spatial intersection algorithm calculates the geometric relationship between the coordinate points of the obstacle contour line and the pre-marked set-up coordinates, for example, using a polygon intersection detection algorithm to detect whether the set-up coordinates are located in the inner area of ​​the obstacle or intersect with the obstacle boundary. Once it is detected that the set-up coordinates coincide with the obstacle, the system will start a logical deletion mechanism to process the pre-marked set-up coordinates that coincide with the obstacle. This process first uses a point-in-polygon detection algorithm to identify which set-up coordinates fall within the boundary or interior of the obstacle. The algorithm determines whether a point is located within a polygon based on a ray method or an angle sum method. If it is determined that some set-up coordinate points are located within the obstacle area, the system will mark 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 coordinate point, and the flag of the coordinate point that coincides with an obstacle will be set to an "invalid" state. The purpose of this is to retain the original data in its entirety and ensure that the original coordinate information is still available in any case where recovery or recalculation is required. This logical deletion method quickly marks and retrieves coordinate points through an indexing mechanism, thereby ensuring the structural integrity of the entire digital drawing.

[0078] Furthermore, when the wire-laying robot cannot receive the laser signal, controlling the wire-laying robot to perform the wire-laying operation in the positioning blind area according to the execution drawing includes:

[0079] 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;

[0080] S2, controlling the wire-laying robot to reach the entry point and perform positioning calibration;

[0081] 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;

[0082] 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;

[0083] 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.

[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 line to be laid out and inkjet. The selection of this point is based on the principle of the minimum number of turns, ensuring that the robot reduces unnecessary turns in path planning to improve the efficiency of movement. In the specific implementation, the system uses the Dijkstra algorithm or the A* algorithm for path planning. These algorithms calculate the length and number of turns of all feasible paths, and use the path with the smallest comprehensive cost as the preferred path of the entry point. In order to optimize the calculation process, the algorithm will combine the coordinate information of the obstacle to ensure that the path planning avoids the influence area of ​​the obstacle. In addition, the optimization of the number of turns will further reduce unnecessary complex paths through heuristic search. Next, step S2 involves the line laying robot arriving at the entry point and performing positioning calibration. In order to ensure accurate positioning when arriving at the entry point, the system combines the Kalman filter algorithm to predict and correct the motion data collected by the robot in real time. The Kalman filter estimates the position, velocity and acceleration of the robot in continuous time steps, and removes the error 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 laying out operations according to the laying out coordinates in the execution drawing. This process relies on the execution drawing that has been corrected after the obstacle mapping to ensure that the robot can perform the laying out task along the preset path according to the calibrated coordinates. Since the laser signal is unavailable, the robot needs to rely on the path tracking algorithm to move in the positioning blind area. Common path tracking algorithms include the Pure Pursuit algorithm, which ensures that the robot can move 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 moves in the blind area, it will accurately control the inkjet device according to the position information of each coordinate point to ensure the continuity and accuracy of the laying out. Each time a straight line laying out inkjet task is completed in the positioning blind area, step S4 will also control the robot to pause the laying out inkjet and leave the positioning blind area. The path planning to leave the blind area also relies on the previous path tracking algorithm and inertial navigation system to ensure that the robot can smoothly exit the blind area. During this process, the system will continue to monitor the robot's motion status to ensure that it reacquires the laser signal or calibration information after exiting the blind area. After exiting, the system will execute step S1 again and select the next entry point to ensure the smooth progress of the subsequent wire laying task.

[0085] Furthermore, the S1 comprises the following steps:

[0086] 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;

[0087] According to the weight evaluation results, the positioning blind area entry point with the least number of turns and the shortest path is selected.

[0088] Specifically, the formula of Dijkstra 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, which represents the comprehensive score of evaluating each path; W L Path length weight 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 weight of the number of turns. The more turns there are, the greater the complexity of the robot's path and the possibility of error accumulation. Therefore, a higher weight may be given in some scenarios. Turn Count is the number of turns, which refers to the number of turns required by the wire-laying robot on the path from the current position to the entry point of the positioning blind area.

[0091] The system calculates the feasible entry points of 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 the graph structure. In this embodiment, the positioning blind area and the current position of the robot constitute a graph structure, and each potential entry point of the blind area is regarded as a node in the graph. The Dijkstra algorithm starts from the current position of the wire-laying robot, 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 that the route with the shortest path length is found. This step provides basic data for selecting entry points. Next, the system performs a weight evaluation on all paths. The standard for weight evaluation is not only the path length, but also the number of turns. The number of turns is an important factor affecting the robot's mobility efficiency. Frequent turns will increase energy consumption, extend the moving time, and may even increase the error of the wire-laying path. Therefore, the weight evaluation will comprehensively consider the 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. The path length is directly obtained through the results of the Dijkstra algorithm, while the number of turns is calculated through the path angle change. Specifically, the system calculates the angle change of the robot's movement direction on each path segment. If the angle change of the two paths exceeds the set threshold, it is regarded as a turn. The total number of turns of the path is obtained by accumulating the angle change of the path. After the weight evaluation is completed, the system selects the optimal entry point according to the weight evaluation results. The selection criteria for the optimal entry point are the least number of turns and the shortest path. In practical applications, the system will give priority to the path with the smallest weight to ensure that the robot enters the positioning blind spot with the least number of turns and the shortest path. This not only ensures that the robot enters the blind spot efficiently, but also reduces the accumulation of path deviations caused by frequent turns.

[0092] 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 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 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 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 position overlap; 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 inkjet 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