Method for positioning tower crane by using sensor data

Abstract

The invention provides a method for positioning a tower crane by using sensor data, and relates to the technical field of building construction.The method comprises the steps that all sensors are installed at preset key parts of a target tower crane, and all sensing data signals are obtained based on the sensors; the preset key part comprises a lifting arm, a lifting hook and a slewing mechanism; performing de-noising processing on each sensing data signal to obtain a de-noised signal; all the de-noised signals are fused, and a fused signal is obtained; determining initial position attitude information of the target tower crane according to the fusion signal; and acquiring a panoramic image of an area where the target tower crane is located, and identifying key marker information of the panoramic image so as to optimize the initial position attitude information according to the key marker information to obtain optimized position attitude information. The construction efficiency and safety are improved, a foundation is laid for intelligent development of the building industry, and wide application prospects and economic benefits are achieved.

Description

Technical Field

[0001] The invention relates to the technical field of building construction, and in particular to a method for positioning a tower crane using sensor data. Background Art

[0002] With the rapid development of the construction industry, the demand for automation and intelligence of tower cranes in high-rise building construction is increasing. The positioning of traditional tower cranes mainly relies on manual operation, which has problems such as low precision, low efficiency and complex operation. Although some automated tower crane positioning technologies have been developed, most of them rely on a single sensor or simple control algorithm, which is difficult to meet the high-precision positioning requirements in complex construction environments.

[0003] Therefore, there is an urgent need for an intelligent positioning method based on multi-sensor fusion to achieve unmanned and precise control of tower cranes. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a method for positioning a tower crane using sensor data, which can achieve high-precision positioning and unmanned control of the tower crane in a complex construction environment through multi-sensor data fusion and intelligent algorithms.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A method for positioning a tower crane using sensor data, comprising:

[0007] Installing various sensors at preset key positions of the target tower crane, and acquiring various sensor data signals based on the sensors; the preset key positions include a boom, a hook, and a slewing mechanism;

[0008] Performing denoising processing on each of the sensor data signals to obtain a denoised signal;

[0009] Fusing the denoised signals to obtain a fused signal;

[0010] Determining the initial position and posture information of the target tower crane according to the fusion signal;

[0011] A panoramic image of the area where the target tower crane is located is obtained, and key marker information of the panoramic image is identified, so as to optimize the initial position and posture information according to the key marker information to obtain optimized position and posture information.

[0012] Preferably, it also includes:

[0013] Perform path planning according to the optimized position and posture information and preset target hanging object information to obtain an optimized path;

[0014] A control instruction is generated according to the optimized path, and the control instruction is sent to an actuator of the target tower crane to achieve unmanned control of the target tower crane.

[0015] Preferably, various sensors are installed at preset key positions of the target tower crane, including:

[0016] Position sensors are arranged on the boom, hook and slewing mechanism of the target tower crane to monitor the three-dimensional spatial position of the tower crane in real time;

[0017] An inertial measurement unit is arranged on the slewing mechanism and the boom of the target tower crane to detect the tilt angle, acceleration and angular velocity of the tower crane;

[0018] A laser distance measuring sensor is arranged on the hook of the target tower crane to measure the distance between the hook and the target object.

[0019] Setting a visual sensor at the end of the boom of the target tower crane to capture a panoramic image and key landmark information of the construction site;

[0020] A wind speed sensor is arranged on the top of the target tower crane to monitor the wind speed in the construction environment and provide environmental compensation data.

[0021] Preferably, performing denoising processing on each of the sensor data signals to obtain a denoised signal comprises:

[0022] For each type of the sensor data signal, the sensor data signal is shifted to obtain a shifted signal;

[0023] Performing wavelet decomposition on the shifted signal to obtain a plurality of wavelet coefficients;

[0024] Determining a filtering threshold according to the decomposition scale and length of the sensor data signal;

[0025] Constructing a denoising function according to the filtering threshold;

[0026] The denoising function is used to remove noise from the translated signal to obtain the denoised signal.

[0027] Preferably, the formula of the filtering threshold is:

[0028]

[0029] Among them, λ is the filtering threshold, j is the decomposition scale of the translated signal, and d j is the decomposition of the wavelet coefficients with a scale of j, N represents the length of the shifted signal, and median represents the median operation.

[0030] Preferably, removing the noise of the translated signal by using the denoising function to obtain the denoised signal comprises:

[0031] The corresponding wavelet coefficients are removed by using the denoising function to obtain a smoothed signal; wherein the denoising function is:

[0032]

[0033] Among them, X represents the wavelet coefficient, n represents the adjustment coefficient;

[0034] Performing inverse translation on the smoothed signal to obtain an inverse translated signal;

[0035] The denoised signal is obtained by averaging a plurality of inverse-translated signals.

[0036] Preferably, determining the initial position and posture information of the target tower crane according to the fusion signal includes:

[0037] Obtain a sample fusion signal sample set;

[0038] Construct an initial convolutional neural network;

[0039] Using the sample fusion signal sample set to train the initial convolutional neural network to obtain a trained position and posture prediction model;

[0040] The fused signal is input into the position and posture prediction model to obtain the initial position and posture information.

[0041] Preferably, the initial position and posture information is optimized according to the key marker information to obtain the optimized position and posture information, including:

[0042] Calculate an error vector based on the actual spatial coordinates of the key marker information and the initial position and posture information;

[0043] Constructing an error function, taking the sum of squares of errors between the actual coordinates and the predicted coordinates of the key marker information as an optimization target;

[0044] Based on the error function, the marker information is used as the observation value, and the initial position and posture information of the tower crane is used as the prediction value. The Kalman filter algorithm is used to fuse the observation value and the prediction value to obtain the optimized position and posture information.

[0045] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0046] The present invention provides a method for positioning a tower crane using sensor data, comprising: installing various sensors at preset key positions of a target tower crane, and obtaining various sensor data signals based on the sensors; the preset key positions include a boom, a hook, and a slewing mechanism; performing denoising processing on each of the sensor data signals to obtain a denoised signal; fusing each of the denoised signals to obtain a fused signal; determining the initial position and posture information of the target tower crane according to the fused signal; obtaining a panoramic image of the area where the target tower crane is located, and identifying key marker information of the panoramic image, so as to optimize the initial position and posture information according to the key marker information, and obtain the optimized position and posture information. The present invention not only improves construction efficiency and safety, but also lays a foundation for the intelligent development of the construction industry, and has broad application prospects and economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0048] Figure 1 A flow chart of a method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] The purpose of the present invention is to provide a method for tower crane positioning using sensor data, which not only improves construction efficiency and safety, but also lays a foundation for the intelligent development of the construction industry and has broad application prospects and economic benefits.

[0051] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] Figure 1 A flow chart of a method provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides a method for positioning a tower crane using sensor data, comprising:

[0053] Step 100: installing various sensors at preset key positions of the target tower crane, and acquiring various sensor data signals based on the sensors; the preset key positions include a boom, a hook, and a slewing mechanism;

[0054] Step 200: performing denoising processing on each sensor data signal to obtain a denoised signal;

[0055] Step 300: Fusing the denoised signals to obtain a fused signal;

[0056] Step 400: determining the initial position and posture information of the target tower crane according to the fused signal;

[0057] Step 500: Acquire a panoramic image of the area where the target tower crane is located, and identify key marker information of the panoramic image, so as to optimize the initial position and posture information according to the key marker information to obtain optimized position and posture information.

[0058] Preferably, it also includes:

[0059] Step 600: performing path planning according to the optimized position and posture information and preset target hanging object information to obtain an optimized path;

[0060] Step 700: Generate a control instruction according to the optimized path, and send the control instruction to the actuator of the target tower crane to achieve unmanned control of the target tower crane.

[0061] Preferably, various sensors are installed at preset key positions of the target tower crane, including:

[0062] Position sensors are arranged on the boom, hook and slewing mechanism of the target tower crane to monitor the three-dimensional spatial position of the tower crane in real time.

[0063] An inertial measurement unit is arranged on the slewing mechanism and the boom of the target tower crane to detect the tilt angle, acceleration and angular velocity of the tower crane;

[0064] The laser distance measuring sensor is arranged on the hook of the target tower crane to measure the distance between the hook and the target object;

[0065] Setting a visual sensor at the end of the boom of the target tower crane to capture a panoramic image and key landmark information of the construction site;

[0066] A wind speed sensor is arranged on the top of the target tower crane to monitor the wind speed in the construction environment and provide environmental compensation data.

[0067] Specifically, in this embodiment, position sensors are installed on the boom, hook and slewing mechanism of the target tower crane to monitor the three-dimensional spatial position of the tower crane in real time. Specifically, the position sensor can use a high-precision GPS module or an inertial navigation system (INS), which can provide the position information of the tower crane in the three directions of X, Y, and Z. During installation, the position sensor should be fixed in a stable structural position to ensure that it is not affected by vibration and impact during the operation of the tower crane. At the same time, the output signal of the sensor should be transmitted to the central control system in real time through a wired or wireless communication module (such as CAN bus or Wi-Fi) for subsequent data processing and analysis.

[0068] Optionally, in this embodiment, an inertial measurement unit (IMU) is set on the slewing mechanism and boom of the target tower crane to detect the tilt angle, acceleration and angular velocity of the tower crane. IMU is usually composed of an accelerometer and a gyroscope, and can monitor the dynamic state of the tower crane in real time. During installation, the IMU should be placed near the center of gravity of the tower crane to improve the accuracy of the measurement. The output data of the IMU will be used to calculate the attitude information of the tower crane, and combined with the data of the position sensor, the motion state of the tower crane can be more fully understood. The data of the IMU also needs to be transmitted to the central control system through the communication module for real-time monitoring and data fusion.

[0069] Furthermore, in this embodiment, a laser rangefinder sensor is installed on the hook of the target tower crane to measure the distance between the hook and the target object. The laser rangefinder sensor can provide high-precision distance measurement and is suitable for various hoisting scenarios. During installation, the sensor should be facing the hoisting target to ensure that its field of view is unobstructed. The output signal of the laser rangefinder sensor will be used to monitor the height of the hook and its relative position to the target object in real time, helping to optimize the path planning and control strategy during the hoisting process. The sensor data also needs to be transmitted to the central control system in real time so that it can be integrated with other sensor data.

[0070] Furthermore, in this embodiment, a visual sensor (such as a camera) is arranged at the end of the boom of the target tower crane to capture panoramic images and key marker information of the construction site. The visual sensor should have high resolution and wide-angle field of view so that the details of the surrounding environment can be clearly captured. During installation, the camera should be fixed on a stable bracket to avoid vibration that affects the image quality. Through image processing algorithms, the system can identify key markers (such as QR codes, reflective signs, etc.) at the construction site and extract their location information. This information will be used to optimize the initial position and posture of the tower crane to ensure the safety and accuracy of the hoisting process.

[0071] In addition, this embodiment also installs a wind speed sensor on the top of the target tower crane to monitor the wind speed in the construction environment and provide environmental compensation data. The wind speed sensor should be selected with high sensitivity and anti-interference ability to ensure accurate measurement of wind speed under various climatic conditions. During installation, the sensor should be placed at the highest point of the tower crane to avoid being blocked by the tower crane structure. The wind speed data will be used to monitor changes in the construction environment in real time, especially during high-altitude operations, where changes in wind speed may affect the stability and safety of the tower crane. By combining wind speed data with other sensor data, this embodiment can dynamically adjust the operation strategy of the tower crane to ensure safe operation.

[0072] Through the above five steps, this embodiment can effectively install various sensors at key positions of the target tower crane, thereby realizing real-time monitoring and precise control of the tower crane.

[0073] Preferably, performing denoising processing on each of the sensor data signals to obtain a denoised signal comprises:

[0074] For each type of the sensor data signal, the sensor data signal is shifted to obtain a shifted signal;

[0075] Performing wavelet decomposition on the shifted signal to obtain a plurality of wavelet coefficients;

[0076] Determining a filtering threshold according to the decomposition scale and length of the sensor data signal;

[0077] Constructing a denoising function according to the filtering threshold;

[0078] The denoising function is used to remove noise from the translated signal to obtain the denoised signal.

[0079] Preferably, the formula of the filtering threshold is:

[0080]

[0081] Among them, λ is the filtering threshold, j is the decomposition scale of the translated signal, and d j is the decomposition of the wavelet coefficients with a scale of j, N represents the length of the shifted signal, and median represents the median operation.

[0082] Preferably, removing the noise of the translated signal by using the denoising function to obtain the denoised signal comprises:

[0083] The corresponding wavelet coefficients are removed by using the denoising function to obtain a smoothed signal; wherein the denoising function is:

[0084]

[0085] Among them, X represents the wavelet coefficient, n represents the adjustment coefficient;

[0086] Performing inverse translation on the smoothed signal to obtain an inverse translated signal;

[0087] The denoised signal is obtained by averaging a plurality of inverse-translated signals.

[0088] Specifically, this embodiment uses wavelet coefficients to measure the attenuation deviation, so that the attenuation of wavelet coefficients with larger absolute values ​​decreases as their absolute values ​​increase, thereby effectively avoiding the loss of high-frequency information and improving the signal-to-noise ratio of the signal.

[0089] Optionally, the process of fusing the denoised signals to obtain a fused signal in this embodiment is achieved through multi-sensor data fusion technology. First, denoised signals from different sensors (such as position sensors, inertial measurement units, laser ranging sensors, and visual sensors) are collected, and these signals may have different timestamps and measurement accuracy. Next, these signals are aligned using a time synchronization algorithm to ensure that they are compared at the same time point. Then, the denoised signals are fused using a weighted average method or a Kalman filter algorithm, where the weighted average method assigns different weights according to the reliability and accuracy of each sensor, while the Kalman filter dynamically fuses the signals through prediction and update steps to eliminate noise and uncertainty. Ultimately, the fused signal will provide a comprehensive and accurate tower crane position and attitude information as the basis for subsequent processing and decision-making.

[0090] Preferably, determining the initial position and posture information of the target tower crane according to the fusion signal includes:

[0091] Obtain a sample fusion signal sample set;

[0092] Construct an initial convolutional neural network;

[0093] Using the sample fusion signal sample set to train the initial convolutional neural network to obtain a trained position and posture prediction model;

[0094] The fused signal is input into the position and posture prediction model to obtain the initial position and posture information.

[0095] Specifically, this embodiment first collects a large amount of fusion signal data from each sensor to construct a sample fusion signal sample set. This process includes real-time acquisition of data from position sensors, inertial measurement units, laser ranging sensors, and visual sensors, and denoising and fusing these data to form high-quality fusion signals. The sample set should cover different working states and environmental conditions to ensure the generalization ability of the model. Specifically, the sample set should include the initial position and posture information of multiple tower cranes, and record sensor data at different time points and in different construction environments to ensure the diversity and representativeness of the data.

[0096] Furthermore, after obtaining the sample fusion signal sample set, the present embodiment needs to construct an initial convolutional neural network (CNN) model. The design of the model should take into account the characteristics of the input data and the requirements of the target output. Typically, a convolutional neural network consists of multiple convolutional layers, pooling layers, and fully connected layers, which can effectively extract spatial features from the input signal. The input layer will receive the fusion signal sample set, the convolutional layer will extract features through convolution operations, the pooling layer will reduce the feature dimension to reduce the computational complexity, and finally output the predicted initial position and posture information through the fully connected layer. During the construction process, it is also necessary to select appropriate activation functions (such as ReLU) and loss functions (such as mean square error) to optimize the training effect of the model.

[0097] Furthermore, this embodiment uses the acquired sample fusion signal sample set to train the initial convolutional neural network to obtain a trained position and posture prediction model. During the training process, the sample set is divided into a training set and a validation set, and the model is iteratively trained using the training set. The network weights are adjusted through the back propagation algorithm to minimize the error between the predicted output and the true initial position and posture information. After each training cycle (epoch), the performance of the model is evaluated using the validation set to ensure the generalization ability of the model on unseen data. After the training is completed, the fusion signal is input into the trained position and posture prediction model, and the model will output the initial position and posture information of the target tower crane, providing high-precision positioning results to support subsequent path planning and control decisions.

[0098] Preferably, the initial position and posture information is optimized according to the key marker information to obtain the optimized position and posture information, including:

[0099] Calculate an error vector based on the actual spatial coordinates of the key marker information and the initial position and posture information;

[0100] Constructing an error function, taking the sum of squares of errors between the actual coordinates and the predicted coordinates of the key marker information as an optimization target;

[0101] Based on the error function, the marker information is used as the observation value, and the initial position and posture information of the tower crane is used as the prediction value. The Kalman filter algorithm is used to fuse the observation value and the prediction value to obtain the optimized position and posture information.

[0102] Specifically, this embodiment obtains a panoramic image of the area where the target tower crane is located by installing a visual sensor (such as a high-resolution camera) at the end of the tower crane boom. The image should cover the key areas of the construction site, including the working range and surrounding environment of the tower crane. Next, the key marker information in the panoramic image is identified using image processing algorithms (such as edge detection, feature point extraction, etc.). These markers can be artificially set (such as QR codes, reflective signs) or naturally exist (such as corners and edges of buildings, etc.), and the actual spatial coordinates of each marker should be known in advance or obtained through calibration.

[0103] Furthermore, after identifying the key markers, the present embodiment compares the actual spatial coordinates of these markers with the coordinates predicted by the initial position and posture information. Specifically, the initial position and posture information is first converted into corresponding predicted coordinates. Then, the difference between the actual coordinates and the predicted coordinates of each marker is calculated to form an error vector. The calculation formula of the error vector is:

[0104] Error vector = actual coordinates - predicted coordinates

[0105] This error vector will be used in the subsequent optimization process to quantify the accuracy of the initial position and posture information.

[0106] Furthermore, this embodiment constructs an error function to quantify the optimization target. The error function is in the form of the sum of square errors of all markers. By minimizing this error function, the optimal tower crane position and posture information can be found to minimize the difference between the predicted coordinates and the actual coordinates. This process provides a clear goal for the subsequent optimization algorithm.

[0107] After constructing the error function, the Kalman filter algorithm is used to fuse the marker information as the observed value and the initial position and posture information of the crane as the predicted value. Kalman filtering is a recursive algorithm that can combine predicted and observed data in a dynamic system to update the state estimate. Specifically, the current state is first predicted based on the motion model of the crane, and then the observed marker information is compared with the predicted state, the Kalman gain is calculated, and the state estimate is updated. The update formula of the Kalman filter includes a prediction step and an update step to ensure that the fused result can effectively reduce the error.

[0108] Through the iterative process of the Kalman filter algorithm, the optimized tower crane position and attitude information is finally obtained. This information not only takes into account the predicted value of the initial position and attitude, but also combines the actual observation values ​​from key landmarks, which significantly improves the positioning accuracy. The optimized position and attitude information will be used for subsequent path planning and control decisions to ensure the safe and efficient operation of the tower crane in a complex construction environment. In this way, the system can dynamically adapt to environmental changes, adjust the tower crane's operating strategy in real time, and further improve construction efficiency and safety.

[0109] Further, step 600 of the present embodiment is specifically as follows: the process of path planning according to the optimized position posture information and the preset target hanging object information, firstly, the spatial position and size of the target hanging object, as well as the current optimized position posture information of the tower crane need to be defined. A path planning algorithm (such as A* algorithm, Dijkstra algorithm or RRT algorithm) is used to calculate the optimal path from the current tower crane position to the target hanging object position. During the path planning process, the algorithm will consider the obstacles at the construction site, the motion limitations of the tower crane (such as the maximum extension length of the boom, the rotation angle, etc.) and environmental factors (such as wind speed, terrain, etc.). The goal of path planning is to generate a safe and effective path to ensure that the hook can smoothly reach the position of the target hanging object during the hoisting process while avoiding collision with surrounding obstacles. The generated optimized path will include a series of waypoints, which will serve as the basis for subsequent control instructions.

[0110] Furthermore, step 700 of this embodiment is specifically as follows: after obtaining the optimized path, the system will generate specific control instructions according to the intermediate points in the path. These instructions include motion instructions of the tower crane, such as the extension and retraction of the boom, the rotation angle of the slewing mechanism, the lifting height of the hook, etc. The generation of control instructions needs to consider the motion model and dynamic characteristics of the tower crane to ensure the rationality and executability of the instructions. The generated control instructions will be sent to the actuator of the target tower crane through a wireless communication module (such as Wi-Fi or 5G). After receiving the instructions, the actuator will adjust the motion state of the tower crane according to the instructions to achieve unmanned control of the target hanging object. During the execution process, the system will also monitor the status of the tower crane in real time, and dynamically adjust the control instructions according to the actual situation to cope with possible environmental changes or emergencies, and ensure the safety and accuracy of the lifting process.

[0111] The beneficial effects of the present invention are as follows:

[0112] (1) The present invention can obtain the precise position and attitude information of the tower crane in real time by installing multiple sensors (such as position sensors, IMUs, laser ranging sensors, etc.) at key positions of the tower crane. After denoising and data fusion processing, the positioning accuracy can be significantly improved, and the positioning deviation caused by environmental interference or sensor error can be reduced.

[0113] (2) The present invention can effectively eliminate noise and interference by denoising the sensor data signal, ensuring the stability of the system in a complex construction environment. This stability is crucial to the safety of high-rise building construction and can reduce the risk of accidents caused by misoperation or sensor failure.

[0114] (3) The present invention can realize unmanned control of the tower crane by optimizing the position and posture information, reducing the reliance on manual operation. This not only improves the construction efficiency, but also reduces the labor cost and safety hazards, especially when working at high altitudes, and can effectively protect the safety of workers.

[0115] (4) By acquiring panoramic images and identifying key marker information, the system can dynamically adapt to changes in the construction site (such as obstacles, wind speed, etc.) and adjust the tower crane's operating strategy in real time. This adaptability enables the tower crane to maintain efficient and safe operation in complex environments.

[0116] (5) After obtaining the optimized position and posture information, the present invention can perform more accurate path planning to ensure that the hook moves along the optimal path during the lifting process and avoids collision with surrounding obstacles. This optimization not only improves work efficiency, but also reduces errors and losses during the lifting process.

[0117] (6) The present invention can provide data support for the operation of the tower crane through real-time collection and processing of sensor data, forming a data-driven intelligent decision-making system. This system can continuously optimize the control strategy through historical data analysis and machine learning algorithms to improve the overall construction efficiency.

[0118] (7) The present invention can significantly reduce the waste of time and resources during the construction process by improving positioning accuracy, realizing unmanned control and optimizing path planning, thereby reducing the overall construction cost.

[0119] (8) The present invention can timely discover potential safety hazards, reduce the probability of accidents, and improve the safety of the construction site by real-time monitoring and optimizing the operating status of the tower crane.

[0120] (9) The present invention can flexibly adjust the configuration of sensors and data processing algorithms according to the needs of different construction sites, has good scalability, and is applicable to various types of tower cranes and construction environments.

[0121] (10) The implementation of the present invention provides technical support for the intelligent and automated development of the construction industry and promotes the construction of intelligent buildings and smart cities.

[0122] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0123] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for tower crane positioning using sensor data, characterized in that: include: Installing various sensors at preset key positions of the target tower crane, and acquiring various sensor data signals based on the sensors; The preset key parts include the boom, hook and slewing mechanism; Performing denoising processing on each of the sensor data signals to obtain a denoised signal; Fusing the denoised signals to obtain a fused signal; Determining the initial position and posture information of the target tower crane according to the fusion signal; A panoramic image of the area where the target tower crane is located is obtained, and key marker information of the panoramic image is identified, so as to optimize the initial position and posture information according to the key marker information to obtain optimized position and posture information.

2. The method for tower crane positioning using sensor data according to claim 1, characterized in that: Also includes: Perform path planning according to the optimized position and posture information and preset target hanging object information to obtain an optimized path; A control instruction is generated according to the optimized path, and the control instruction is sent to an actuator of the target tower crane to achieve unmanned control of the target tower crane.

3. The method for tower crane positioning using sensor data according to claim 1, characterized in that: Install various sensors at preset key locations of the target crane, including: Position sensors are arranged on the boom, hook and slewing mechanism of the target tower crane to monitor the three-dimensional spatial position of the tower crane in real time; An inertial measurement unit is arranged on the slewing mechanism and the boom of the target tower crane to detect the tilt angle, acceleration and angular velocity of the tower crane; The laser distance measuring sensor is arranged on the hook of the target tower crane to measure the distance between the hook and the target object; Setting a visual sensor at the end of the boom of the target tower crane to capture a panoramic image and key landmark information of the construction site; A wind speed sensor is arranged on the top of the target tower crane to monitor the wind speed in the construction environment and provide environmental compensation data.

4. The method for tower crane positioning using sensor data according to claim 1, characterized in that: Performing denoising processing on each of the sensor data signals to obtain a denoised signal includes: For each type of the sensor data signal, the sensor data signal is shifted to obtain a shifted signal; Performing wavelet decomposition on the shifted signal to obtain a plurality of wavelet coefficients; Determining a filtering threshold according to the decomposition scale and length of the sensor data signal; Constructing a denoising function according to the filtering threshold; The denoising function is used to remove noise from the translated signal to obtain the denoised signal.

5. The method for tower crane positioning using sensor data according to claim 4, characterized in that: The formula for the filtering threshold is: Among them, λ is the filtering threshold, j is the decomposition scale of the translated signal, and d j is the decomposition of the wavelet coefficients with a scale of j, N represents the length of the shifted signal, and median represents the median operation.

6. The method for tower crane positioning using sensor data according to claim 5, characterized in that: Using the denoising function to remove the noise of the translated signal to obtain the denoised signal includes: The corresponding wavelet coefficients are removed by using the denoising function to obtain a smoothed signal; wherein the denoising function is: Among them, X represents the wavelet coefficient, n represents the adjustment coefficient; Performing inverse translation on the smoothed signal to obtain an inverse translated signal; The denoised signal is obtained by averaging a plurality of inverse-translated signals.

7. The method for tower crane positioning using sensor data according to claim 1, characterized in that: Determining the initial position and posture information of the target tower crane according to the fusion signal includes: Obtain a sample fusion signal sample set; Construct an initial convolutional neural network; Using the sample fusion signal sample set to train the initial convolutional neural network to obtain a trained position and posture prediction model; The fused signal is input into the position and posture prediction model to obtain the initial position and posture information.

8. The method for tower crane positioning using sensor data according to claim 1, characterized in that: Optimizing the initial position and posture information according to the key marker information to obtain optimized position and posture information includes: Calculate an error vector based on the actual spatial coordinates of the key marker information and the initial position and posture information; Constructing an error function, taking the sum of squares of errors between the actual coordinates and the predicted coordinates of the key marker information as an optimization target; Based on the error function, the marker information is used as the observation value, and the initial position and posture information of the tower crane is used as the prediction value. The Kalman filter algorithm is used to fuse the observation value and the prediction value to obtain the optimized position and posture information.

Images