CN120024817B - A crane anti-collision detection method, system, device and medium - Google Patents

A crane anti-collision detection method, system, device and medium

Info

Publication number
CN120024817B
CN120024817B CN202510497510.2A CN202510497510A CN120024817B CN 120024817 B CN120024817 B CN 120024817B CN 202510497510 A CN202510497510 A CN 202510497510A CN 120024817 B CN120024817 B CN 120024817B
Authority
CN
China
Prior art keywords
data
crane
trajectory
gru
lstm
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202510497510.2A
Other languages
Chinese (zh)
Other versions
CN120024817A (en
Inventor
宋诗振
殷志海
高璇璇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Bangze Crane Equipment Co ltd
Original Assignee
Bangze Crane Equipment Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Bangze Crane Equipment Co ltd filed Critical Bangze Crane Equipment Co ltd
Priority to CN202510497510.2A priority Critical patent/CN120024817B/en
Publication of CN120024817A publication Critical patent/CN120024817A/en
Application granted granted Critical
Publication of CN120024817B publication Critical patent/CN120024817B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B66—HOISTING; LIFTING; HAULING
    • B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
    • B66C13/00—Other constructional features or details
    • B66C13/16—Applications of indicating, registering, or weighing devices
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B66—HOISTING; LIFTING; HAULING
    • B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
    • B66C13/00—Other constructional features or details
    • B66C13/18—Control systems or devices
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B66—HOISTING; LIFTING; HAULING
    • B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
    • B66C15/00—Safety gear
    • B66C15/06—Arrangements or use of warning devices
    • B66C15/065—Arrangements or use of warning devices electrical
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/10—Pre-processing; Data cleansing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/10—Pre-processing; Data cleansing
    • G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
    • G06F18/256—Fusion techniques of classification results, e.g. of results related to same input data of results relating to different input data, e.g. multimodal recognition
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Mechanical Engineering (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Biomedical Technology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Automation & Control Theory (AREA)
  • Health & Medical Sciences (AREA)
  • Control And Safety Of Cranes (AREA)

Abstract

本发明公开了一种起重机械防碰撞检测方法、系统、设备和介质，涉及计算机平台负载平衡技术领域，包括采集起重机械的多模态数据，将多模态数据进行预处理；通过卡尔曼滤波对采集的多模态数据进行融合，构建LSTM‑GRU预测模型，将融合数据输入LSTM‑GRU预测模型对起重机械运行轨迹进行预测；根据轨迹预测结果，通过AI边缘计算设备动态生成防碰撞控制指令，PLC接受并执行防碰撞控制指令。本发明所述方法基于多模态数据融合，提高起重机运行状态感知精度，基于LSTM‑GRU轨迹预测，实现精准的未来轨迹计算，基于AI边缘计算设备，实现动态防碰撞控制，有效提升了防碰撞检测的智能化水平。

The present invention discloses a method, system, device, and medium for anti-collision detection of a crane, relating to the technical field of computer platform load balancing. The method comprises collecting multimodal data of the crane and preprocessing the multimodal data; fusing the collected multimodal data through Kalman filtering to construct an LSTM-GRU prediction model; inputting the fused data into the LSTM-GRU prediction model to predict the crane's operating trajectory; and dynamically generating anti-collision control instructions based on the trajectory prediction results through an AI edge computing device, which the PLC receives and executes. The method of the present invention improves the accuracy of crane operating status perception based on multimodal data fusion, achieves accurate future trajectory calculation based on LSTM-GRU trajectory prediction, and implements dynamic anti-collision control based on the AI edge computing device, effectively enhancing the intelligent level of anti-collision detection.

Description

Anti-collision detection method, system, equipment and medium for hoisting machinery
Technical Field
The invention relates to the technical field of intelligent control and safety detection of hoisting machinery, in particular to a method, a system, equipment and a medium for detecting collision resistance of hoisting machinery.
Background
The hoisting machinery is used as key equipment in industrial production, logistics transportation and building construction, and is widely applied to scenes such as ports and docks, warehouse logistics, building sites, large-scale manufacturing enterprises and the like. However, in the running process of the crane, collision accidents are very easy to occur due to complex working environment, large load change and nonlinear running track, so that equipment damage, casualties and economic loss are caused. Therefore, how to realize intelligent anti-collision detection of hoisting machinery and ensure operation safety becomes a key technical problem to be solved in industry.
The hoisting machinery anti-collision method in the industry mainly comprises a fixed sensor detection method, a rule-based collision early warning method and a manual monitoring method, but the methods have the defects that the detection range of a sensor is limited, and the sensor is easily influenced by shielding, so that a detection blind area is caused. The sensor has poor adaptability to ambient light, rain and fog weather, dust interference and the like, and influences the detection precision. The current state can only be passively known, the track cannot be predicted in advance, and collision in high-speed movement cannot be avoided. Depending on fixed rules, the dynamic change under different working conditions cannot be adapted. Future tracks cannot be predicted in advance, and burst collision risks under high-speed running cannot be effectively avoided. When the crane needs to work in a narrow area, the alarm is frequently triggered by mistake, and the working efficiency is affected. Depending on manual experience, there are false positives and reaction lags. The operator has high working strength, and the long-time operation is easy to fatigue, so that potential safety hazards are caused. The crane can not be accurately controlled in a complex environment, and particularly in a cooperative operation scene of a plurality of cranes, the problem of collision caused by misoperation is easy to occur.
Disclosure of Invention
The present invention has been made in view of the above-described problems.
The invention solves the technical problems that the existing hoisting machinery anti-collision detection method has insufficient sensing precision, weak track prediction capability, unintelligy anti-collision strategy and low instantaneity, and how to realize high-precision track prediction and dynamic anti-collision control based on multi-mode data fusion and artificial intelligence technology.
In order to solve the technical problems, the invention provides a hoisting machinery anti-collision detection method which comprises the steps of collecting multi-mode data of a hoisting machinery and preprocessing the multi-mode data.
And fusing the acquired multi-mode data through Kalman filtering, constructing an LSTM-GRU prediction model, and inputting the fused data into the LSTM-GRU prediction model to predict the crane mechanical running track.
And dynamically generating an anti-collision control instruction through the AI edge computing equipment according to the track prediction result, and receiving and executing the anti-collision control instruction by the PLC.
The multi-mode data fusion comprises the steps of carrying out data fusion through Kalman filtering, eliminating measurement errors and improving state estimation accuracy.
Constructing an LSTM-GRU prediction model comprises capturing long-term track trend by adopting LSTM and combining GRU to treat short-term track fluctuation so as to realize short-term and long-term track prediction fusion.
Based on the time sequence prediction, the track of the crane for N time steps in the future is calculated in advance, and the potential collision risk is determined.
And optimizing the track calculation by adopting a track smoothing optimization method.
The crane machinery anti-collision detection method is characterized in that multi-mode data of the crane machinery are synchronously collected through an AI camera, a laser radar, an absolute value encoder and an inertial sensor, the collected multi-mode data of the crane machinery are used as observation values, and a time sequence data set of crane machinery operation is constructed.
The multimodal data includes object category, position coordinates, reflected intensity, target distance, hook height, running speed, angular velocity, acceleration.
As a preferable scheme of the hoisting machinery anti-collision detection method, the method comprises the steps of aligning the data of the sensor according to the highest frequency, and filling the low-frequency sensor data by adopting a linear interpolation method.
Denoising the aligned data based on particle filtering, detecting abnormal values of the denoised sensor data through a mean value-standard difference method, removing abnormal parts, filling again through a linear interpolation method, and repeating the multi-mode data preprocessing process until abnormal value data are not detected in the abnormal value detection process.
Particle filtering denoising comprises dynamically estimating a real signal value through a plurality of particle simulation system states and eliminating measurement noise.
Outlier detection includes calculating the average of all data points, measuring the central trend of the data, by standard deviationAnd measuring the discrete degree of the data, judging the deviation between the data point and the mean value according to the discrete degree, and determining the data with overlarge deviation as an abnormal value.
The hoisting machinery anti-collision detection method is characterized by comprising the following steps of fusing preprocessed multi-mode sensing data based on Kalman filtering, setting position, speed and acceleration information of the hoisting machinery through the preprocessed multi-mode data, and predicting the running state of the hoisting machinery according to a Newton kinematics equation by taking the set position, speed and acceleration information of the hoisting machinery as input data.
And (3) inputting state information at the previous moment by adopting a state transition model, calculating the position, the speed and the acceleration of the crane at the next moment by combining the preprocessed acceleration, and defining a motion rule.
And calculating a Kalman gain, and correcting the predicted state by using the measurement data according to the Kalman gain to obtain fused position information, speed information and acceleration information.
Calculating the Kalman gain comprises inputting a Kalman gain calculation formula through a measurement matrix, a measurement noise covariance matrix and a covariance matrix of state estimation to obtain the Kalman gain.
The covariance matrix of the state estimation measures the uncertainty of the predicted value and dynamically adjusts the confidence level of the prediction.
The measurement matrix maps the state variables to the measurement space and calculates correction values.
The measurement noise covariance matrix is set according to the self error characteristic of the sensor, the uncertainty of measurement data is reflected, and the accuracy of state correction is optimized.
The hoisting machinery anti-collision detection method is characterized in that fused position information, speed information and acceleration information data are taken as input, normalization processing is carried out on the input position information, speed information and acceleration information data, and a time sequence is formed by the normalized position information, speed information and acceleration information data.
And introducing a traditional LSTM prediction model into GRU calculation, and constructing a GRU layer and a full connection layer.
And taking the time sequence as input, adapting to the instant motion change of the hoisting machine through the GRU layer, outputting the overall track trend of the hoisting machine through the LSTM layer, carrying out track prediction calculation by combining the output results of the GRU layer and the LSTM layer, and outputting predicted track points.
The historical position, speed and acceleration data are input into a prediction model, a track smoothing penalty term is introduced on the basis of mean square error calculation, a mean square error loss function is constructed, a historical track prediction result is input into the mean square error loss function, and the error between the historical prediction track and the real track is calculated.
And inputting a fusion result obtained by fusing the multi-mode data into an LSTM-GRU prediction model, and outputting a hoisting machinery running track.
The hoisting machinery anti-collision detection method is a preferable scheme, wherein track data output by a prediction model and position information of an obstacle are used as input, the minimum safety distance between the hoisting machinery and the obstacle is calculated, and a safety distance threshold is set according to the minimum safety distance.
And generating a control strategy according to the set safe distance threshold value and converting the control strategy into control instruction data.
Setting the safety distance threshold comprises sending out an early warning signal when the distance from the crane to the obstacle is smaller than the warning distance.
And when the distance from the crane to the obstacle is smaller than the braking distance, judging to execute braking operation.
Generating the control strategy comprises the steps of setting a control state according to a set safety distance threshold, and enabling the hoisting machinery to normally operate when the distance from the crane to the obstacle is larger than the warning distance.
And when the distance from the crane to the obstacle is between the braking distance and the warning distance, performing audible and visual alarm.
And when the distance from the crane to the obstacle is equal to the braking distance, sending a deceleration instruction to the PLC, and reducing the running speed of the crane.
And when the distance from the crane to the obstacle is smaller than the braking distance, sending a braking instruction to the PLC, and stopping the work of the hoisting machinery.
As a preferable scheme of the anti-collision detection method of the hoisting machinery, control instruction data is input into a PLC control execution unit.
The PLC controls the execution unit to keep the original running speed under the normal running instruction of the hoisting machinery, does not trigger an alarm, does not adjust the frequency of the frequency converter, and does not trigger a brake system.
And under the sound-light alarm instruction, the PLC controls the execution unit to trigger the LED indicator lamp to flash, the buzzer sounds, and warning information is displayed on the control panel or the remote monitoring system, so that the speed of the crane is not adjusted, and the braking system is not triggered.
The PLC controls the execution unit to carry out speed adjustment under the crane running speed instruction, adjusts the output frequency of the frequency converter, keeps audible and visual alarm and does not trigger the braking system.
And the PLC controls the execution unit to cut off the power supply of the crane driving motor under the working instruction of stopping the hoisting machinery, triggers the braking system, executes emergency braking, keeps audible and visual alarm and sends a shutdown report.
Another object of the present invention is to provide a lifting machinery anti-collision detection system, which can solve the problems of insufficient sensing precision, weak track prediction capability, unintelligible anti-collision control strategy and low instantaneity of the existing lifting machinery anti-collision detection technology by using an AI edge calculation scheme based on multi-mode data fusion and LSTM-GRU track prediction.
The hoisting machinery anti-collision detection system comprises a data acquisition and preprocessing module, a data fusion and prediction module and an instruction generation and execution module. The acquisition and preprocessing module is used for acquiring multi-mode data of the hoisting machinery and preprocessing the multi-mode data. The data fusion and prediction module is used for fusing the acquired multi-mode data through Kalman filtering, constructing an LSTM-GRU prediction model, and inputting the fused data into the LSTM-GRU prediction model to predict the crane mechanical running track. The instruction generation and execution module is used for dynamically generating an anti-collision control instruction through the AI edge computing equipment according to the track prediction result, and the PLC receives and executes the anti-collision control instruction.
A computer device comprising a memory storing a computer program and a processor executing the computer program is a step of implementing a hoist anti-collision detection method.
A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of a hoist anti-collision detection method.
The crane anti-collision detection method has the beneficial effects that the crane operation state sensing precision is improved based on multi-mode data fusion, accurate future track calculation is realized based on LSTM-GRU track prediction, dynamic anti-collision control is realized based on AI edge computing equipment, the intelligent level of anti-collision detection is effectively improved, and the crane anti-collision detection method has better effects in the aspects of crane operation safety, track prediction precision and anti-collision response speed.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is an overall flowchart of a method for detecting collision of a hoisting machine according to a first embodiment of the present invention.
Detailed Description
So that the manner in which the above recited objects, features and advantages of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
Embodiment 1, referring to fig. 1, for an embodiment of the present invention, there is provided a method for detecting collision of a hoisting machine, including:
s1, collecting multi-mode data of the hoisting machinery, and preprocessing the multi-mode data.
The multi-modal data of the hoisting machinery are synchronously collected through an AI camera, a laser radar, an absolute value encoder and an inertial sensor, and the collected multi-modal data of the hoisting machinery are used as observation values to construct a time sequence data set of the hoisting machinery operation.
The multimodal data includes object category, position coordinates, reflected intensity, target distance, hook height, running speed, angular velocity, acceleration.
Further, the data of the sensor are aligned according to the highest frequency, and for the low-frequency sensor data, a linear interpolation method is adopted for filling, and a preferable scheme of the linear interpolation method is as follows:
;
Wherein, the Representing the data points after the interpolation,Time of presentationThe raw data of the moment in time,Time of presentationThe raw data of the moment in time,Indicating a time stamp.
Denoising the aligned data based on particle filtering, detecting abnormal values of the denoised sensor data through a mean value-standard difference method, removing abnormal parts, filling again through a linear interpolation method, and repeating the multi-mode data preprocessing process until abnormal value data are not detected in the abnormal value detection process.
Particle filtering denoising comprises dynamically estimating a real signal value through a plurality of particle simulation system states, eliminating measurement noise and improving data stability.
A preferable scheme of particle filtering denoising is as follows:
;
Wherein, the Represents the optimal state estimation value after denoising,The number of particles is indicated and the number of particles,Represent the firstThe previous state of the particles at the moment,The control input is represented as such,The process noise is represented by a characteristic of the process,The state transition equation is represented as such,Represent the firstWeight of individual particles.
Outlier detection includes calculating the average of all data points, measuring the central trend of the data, by standard deviationAnd measuring the discrete degree of the data, and judging the deviation of the data point and the mean value according to the discrete degree.
One preferred scheme for outlier detection is:
;
;
Wherein, the Representing the mean value of the sample data,Representing the number of samples to be taken,Represent the firstA data point is provided for each of the data points,The standard deviation is indicated as such,Representing the deviation of each data point from the mean.
And S2, fusing the acquired multi-mode data through Kalman filtering, constructing an LSTM-GRU prediction model, and inputting the fused data into the LSTM-GRU prediction model to predict the running track of the hoisting machinery.
The preprocessed multi-mode sensing data are fused based on Kalman filtering, the position, speed and acceleration information of the hoisting machinery are set according to the preprocessed multi-mode data, the set position, speed and acceleration information of the hoisting machinery are used as input data, and the running state of the hoisting machinery is predicted according to Newton kinematics equations.
One preferred solution for predicting the operating state of a hoisting machine is:
;
;
;
;
Wherein, the Representing the current time stepIs used to determine the state estimation vector of (c),A state transition matrix is represented and is used to represent,Representing a matrix of control inputs,The process noise is represented by a characteristic of the process,Representing the current time stepIs provided with a control input for the control of the (c),The time step is represented by a time step,、、Representing acceleration values in three axes.
And (3) inputting state information at the previous moment by adopting a state transition model, calculating the position, the speed and the acceleration of the crane at the next moment by combining the preprocessed acceleration, and defining a motion rule.
And calculating a Kalman gain, and correcting the predicted state by using the measurement data according to the Kalman gain to obtain fused position information, speed information and acceleration information.
One preferred scheme for state correction is:
;
;
Wherein, the Representing the kalman gain matrix,The state covariance matrix is represented as such,Representing the measurement matrix of the device,Representing the measurement noise covariance matrix,Representing the state vector of the final estimate,A measurement value representing the current time step is displayed,Representing the projection of the predicted state into the measurement space,Representation based on Kalman gainIs updated by the measurement of (a).
Calculating the Kalman gain comprises inputting a Kalman gain calculation formula through a measurement matrix, a measurement noise covariance matrix and a covariance matrix of state estimation to obtain the Kalman gain.
Further, a state transition model is adopted, state information of the previous moment is input, the position, the speed and the acceleration of the crane at the next moment are calculated by combining the preprocessed acceleration, a motion rule is defined, and the predicted state is corrected according to the measured data.
And inputting a Kalman gain calculation formula through a measurement matrix, a measurement noise covariance matrix and a covariance matrix of state estimation, and calculating the Kalman gain.
The covariance matrix of the state estimation is used for measuring uncertainty of the predicted value and dynamically adjusting confidence level of the prediction.
The measurement matrix is used to map the state variables to the measurement space to ensure that the calculated correction values conform to the data provided by the sensor.
The measurement noise covariance matrix is set according to the own error characteristic of the sensor so as to reflect the uncertainty of measurement data and optimize the accuracy of state correction.
Furthermore, the fused position information, speed information and acceleration information data are taken as input, normalization processing is carried out on the input position information, speed information and acceleration information data, and the normalized position information, speed information and acceleration information data form a time sequence.
One preferred scheme for composing the time series is:
;
Wherein, the Representing a time series data matrix input to the LSTM-GRU,Representing the current time stepIs provided with a position information of (a),Representing the current time stepIs set in the database, the speed information of (a),Current time stepIs provided with an acceleration information of (a),Representing the time window length.
And introducing a traditional LSTM prediction model into GRU calculation, and constructing a GRU layer and a full connection layer.
One preferred scheme for LSTM predictive model calculation is:
;
Wherein, the Indicating that LSTM is in time stepIs used to determine the hidden state of the (c),Representing the output gate of the LSTM,Representing the state of the memory cell of the LSTM,Indicating activation of the long-term trajectory information by a hyperbolic tangent function,Representing element-wise multiplication.
One preferred approach to introducing the GRU calculation is:
;
Wherein, the Indicating GRU at time stepIs used to determine the hidden state of the (c),An update gate representing the gre is presented,The bias term(s) representing the GRU,Input data representing the current time step,A weight matrix representing the GRU is presented,Representing the reset gate of the GRU.
And taking the time sequence as input, adapting to the instant motion change of the hoisting machine through the GRU layer, outputting the overall track trend of the hoisting machine through the LSTM layer, carrying out track prediction calculation by combining the output results of the GRU layer and the LSTM layer, and outputting predicted track points.
One preferred scheme for combining the GRU layer with the LSTM layer is:
;
Wherein, the Representing predicted trajectory data at the next time instant,Representing a weight matrix of the output layer,Representing the LSTM-GRU combination scale factor,Representing the long-term trajectory characteristics computed by LSTM,Representing short-term trajectory characteristics calculated by the GRU,Representing the bias term of the output layer.
A preferred scheme for trajectory prediction is:
;
Wherein, the Representing predicted futureTrace data for a single time step,The representation is based on inputFuture trajectories are predicted.
The historical position, speed and acceleration data are input into a prediction model, a track smoothing penalty term is introduced on the basis of mean square error calculation, a mean square error loss function is constructed, a historical track prediction result is input into the mean square error loss function, and the error between the historical prediction track and the real track is calculated.
One preferred scheme for constructing the mean square error loss function is:
;
Wherein, the Representing the total loss function of the device,Representing the actual future trajectory data,Representing LSTM-GRU predicted future trajectory data,The number of samples of the training data is represented,Representing the smoothing penalty coefficient(s),Representing the second derivative of the predicted trajectory,Representing the length of the time window of the smoothing calculation.
And S3, dynamically generating an anti-collision control instruction through the AI edge computing equipment according to the track prediction result, and receiving and executing the anti-collision control instruction by the PLC.
And taking the track data output by the prediction model and the position information of the obstacle as inputs, calculating the minimum safety distance between the hoisting machinery and the obstacle, and setting a safety distance threshold according to the minimum safety distance.
One preferred solution for calculating the minimum safe distance between the hoisting machine and the obstacle is:
;
Wherein, the Indicating the time steps of the crane and the obstacleIs used for the distance between euclidean distance(s),、、Indicating the crane is at time stepsIs provided with a three-dimensional coordinate position of (c),,,Indicating that the obstacle is at a step in timeIs used for the three-dimensional coordinate position of the lens.
And generating a control strategy according to the set safe distance threshold value and converting the control strategy into control instruction data.
Further, setting the safety distance threshold includes sending an early warning signal when the distance from the crane to the obstacle is smaller than the warning distance.
And when the distance from the crane to the obstacle is smaller than the braking distance, judging to execute braking operation.
Generating the control strategy comprises the steps of setting a control state according to a set safety distance threshold, and enabling the hoisting machinery to normally operate when the distance from the crane to the obstacle is larger than the warning distance.
And when the distance from the crane to the obstacle is between the braking distance and the warning distance, performing audible and visual alarm.
And when the distance from the crane to the obstacle is equal to the braking distance, sending a deceleration instruction to the PLC, and reducing the running speed of the crane.
And when the distance from the crane to the obstacle is smaller than the braking distance, sending a braking instruction to the PLC, and stopping the work of the hoisting machinery.
Further, the control instruction data is input to the PLC control execution unit.
The PLC controls the execution unit to keep the original running speed under the normal running instruction of the hoisting machinery, does not trigger an alarm, does not adjust the frequency of the frequency converter, and does not trigger a brake system.
And under the sound-light alarm instruction, the PLC controls the execution unit to trigger the LED indicator lamp to flash, the buzzer sounds, and warning information is displayed on the control panel or the remote monitoring system, so that the speed of the crane is not adjusted, and the braking system is not triggered.
The PLC controls the execution unit to carry out speed adjustment under the crane running speed instruction, adjusts the output frequency of the frequency converter, keeps audible and visual alarm and does not trigger the braking system.
One preferred scheme for adjusting the output frequency of the frequency converter is:
;
Wherein, the Representing the new frequency converter output frequency of the crane,Representing the current frequency of the output of the frequency converter of the crane,Representing the speed adjustment factor.
And the PLC controls the execution unit to cut off the power supply of the crane driving motor under the working instruction of stopping the hoisting machinery, triggers the braking system, executes emergency braking, keeps audible and visual alarm and sends a shutdown report.
Embodiment 2 provides a method for detecting collision of hoisting machinery for one embodiment of the invention, and in order to verify the beneficial effects of the invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.
In order to verify the effectiveness of the hoisting machinery anti-collision detection method based on LSTM-GRU track prediction, the experiment is carried out in the indoor crane operation environment of a large-scale manufacturing factory. The experimental site comprises a plurality of fixed obstacles, a moving vehicle, a manual operator and a plurality of cranes working in parallel. The aim of the experiment is to evaluate the anti-collision detection precision, track prediction accuracy and control response speed of the system under different working conditions.
Firstly, equipment and sensors are configured, a bridge crane is selected, the maximum lifting load is 10 tons, and the running speed is 1-3 m/s.
The sensor employs an AI camera for target recognition, frame rate 30FPS. The laser radar adopts the detection of the distance of an obstacle and scans the frequency of 10Hz. The absolute value encoder records the hook position, resolution 0.01m. The inertial sensor IMU measures angular velocity and acceleration precision of 0.01m/s2.
The experiment firstly collects the operation data of the crane and preprocesses the data. And synchronizing multi-mode data, including crane position, speed, acceleration, obstacle distance and other information. And particle filtering is adopted to remove noise, so that sensor errors are reduced, and data quality is improved. And filling the low-frequency sensor data by using linear interpolation to ensure the time synchronization of the data. And detecting abnormal values based on a mean value-standard difference method, removing abnormal data and re-interpolating and filling.
And (5) fusing the data by adopting Kalman filtering, and estimating the real-time state of the crane. And inputting the fused data into an LSTM-GRU track prediction model to predict the future track of the crane for 5 seconds. And the track is used for smooth optimization, so that the stability of a prediction curve is ensured, and the abrupt change error is reduced.
And calculating the safety distance between the crane and the obstacle, setting an early warning threshold value 3m, a deceleration threshold value 2m and a braking threshold value 1m, and generating an anti-collision control instruction by the AI computing equipment when the predicted track is about to approach the obstacle. Key indexes such as track error, false alarm rate, control response time, collision occurrence rate and the like in the experimental process are collected, and experimental data are shown in table 1.
Table 1 table of experimental data
Embodiment 3 provides a hoist anti-collision detection system according to an embodiment of the present invention, including a data acquisition and preprocessing module 100, a data fusion and prediction module 200, and an instruction generation and execution module 300.
And S4, the acquisition and preprocessing module 100 is used for acquiring multi-mode data of the hoisting machinery and preprocessing the multi-mode data.
It should be further noted that, the data acquisition and preprocessing module 100 is responsible for sensing the state of the hoisting machine, and acquires data of sensors such as an AI camera, a laser radar, an absolute value encoder, an inertial sensor, and the like, including position information, speed information, and acceleration information, in real time. In order to improve the data quality, the module can perform time alignment, noise removal, outlier detection and data complementation on the acquired multi-mode data, so that the reliability and consistency of the input data are ensured. The output of the module is the preprocessed high quality sensor data and is passed to the data fusion and prediction module 200.
And S5, the data fusion and prediction module 200 is used for fusing the acquired multi-mode data through Kalman filtering, constructing an LSTM-GRU prediction model, and inputting the fused data into the LSTM-GRU prediction model to predict the running track of the hoisting machinery.
It should be further noted that, the data fusion and prediction module 200 is responsible for fusing the preprocessed multi-mode data, and pre-judging the operation trend of the crane in advance based on the track prediction algorithm. And (3) fusing the sensor data by using Kalman filtering, eliminating measurement errors and constructing a unified state estimation model. And inputting the fused data into a prediction model by adopting an LSTM-GRU prediction model, analyzing and calculating a future track point through a time sequence, and determining the motion state of the crane at a future time step. The output of the module is predicted trajectory data and is passed to instruction generation and execution module 300.
And S6, the instruction generation and execution module 300 is used for dynamically generating an anti-collision control instruction through the AI edge computing equipment according to the track prediction result, and the PLC receives and executes the anti-collision control instruction.
It should also be noted that the generating and executing module 300 is responsible for making intelligent anti-collision decisions based on the trajectory prediction data and controlling the operation state of the crane. The AI edge computing device analyzes the predicted trajectory and computes a minimum safe distance of the crane from the obstacle. And according to the safety threshold, the system dynamically generates an anti-collision control instruction, and the PLC executes corresponding operation.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method of the embodiments of the present invention. The storage medium includes a U disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, an optical disk, or other various media capable of storing program codes.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium include an electrical connection (an electronic device) having one or more wires, a portable computer diskette (a magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium may even be paper or other suitable medium upon which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of techniques known in the art, discrete logic circuits with logic gates for implementing logic functions on data signals, application specific integrated circuits with appropriate combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like. It should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention may be modified or substituted without departing from the spirit and scope of the technical solution of the present invention, which is intended to be covered in the scope of the claims of the present invention.

Claims (6)

1.一种起重机械防碰撞检测方法，其特征在于，包括：1. A method for detecting collision avoidance of a lifting machine, comprising: 采集起重机械的多模态数据，将多模态数据进行预处理；Collect multimodal data of lifting machinery and pre-process the multimodal data; 通过卡尔曼滤波对采集的多模态数据进行融合，构建LSTM-GRU预测模型，将融合数据输入LSTM-GRU预测模型对起重机械运行轨迹进行预测；The collected multimodal data is fused through Kalman filtering to build an LSTM-GRU prediction model. The fused data is input into the LSTM-GRU prediction model to predict the operation trajectory of the crane. 根据轨迹预测结果，通过AI边缘计算设备动态生成防碰撞控制指令，PLC接受并执行防碰撞控制指令；Based on the trajectory prediction results, the AI edge computing device dynamically generates anti-collision control instructions, which the PLC receives and executes. 多模态数据融合包括，通过卡尔曼滤波进行数据融合，消除测量误差，提高状态估计精度；Multimodal data fusion includes data fusion through Kalman filtering to eliminate measurement errors and improve state estimation accuracy; 构建LSTM-GRU预测模型包括，采用LSTM捕捉长期轨迹趋势，结合GRU处理短时轨迹波动，实现短时+长时轨迹预测融合；Building an LSTM-GRU prediction model involves using LSTM to capture long-term trajectory trends and combining it with GRU to process short-term trajectory fluctuations, achieving a fusion of short-term and long-term trajectory predictions. 基于时间序列预测，提前计算起重机未来N个时间步的轨迹，确定潜在的碰撞风险；Based on time series prediction, the crane's trajectory is calculated N time steps in the future to determine potential collision risks; 采用轨迹平滑优化方法，优化轨迹计算；Adopt trajectory smoothing optimization method to optimize trajectory calculation; 所述采集起重机械的多模态数据包括，The multimodal data collection of lifting machinery includes: 通过AI摄像头、激光雷达、绝对值编码器、惯性传感器同步采集起重机械的多模态数据，将采集的起重机械多模态数据作为观测值，构建起重机械运行的时间序列数据集；The multimodal data of the crane is collected synchronously through AI cameras, lidar, absolute encoders, and inertial sensors. The collected multimodal data of the crane is used as observation values to construct a time series dataset of the crane operation. 多模态数据包括，物体类别、位置坐标、反射强度、目标距离、吊钩高度、运行速度、角速度、加速度；Multimodal data includes object category, location coordinates, reflection intensity, target distance, hook height, running speed, angular velocity, and acceleration; 所述将多模态数据进行预处理包括，The preprocessing of the multimodal data includes: 将传感器的数据按照最高频率对齐，对于低频传感器数据，采用线性插值法填充；Align the sensor data according to the highest frequency, and use linear interpolation to fill in the low-frequency sensor data; 将对齐后的数据基于粒子滤波进行去噪处理，对去噪处理后的传感器数据通过均值-标准差法进行异常值检测，去除异常部分并再次使用线性插值法进行填充，重复多模态数据预处理过程直至异常值检测过程未检测出异常值数据；The aligned data is denoised using a particle filter. The denoised sensor data is then used for outlier detection using the mean-standard deviation method. The outliers are removed and filled again using linear interpolation. The multimodal data preprocessing process is repeated until no outlier data is detected in the outlier detection process. 粒子滤波去噪包括，通过多个粒子模拟系统状态，动态估计真实信号值，消除测量噪声；Particle filter denoising includes simulating the system state through multiple particles, dynamically estimating the true signal value, and eliminating measurement noise; 异常值检测包括，计算所有数据点的平均值，衡量数据的中心趋势，通过标准差衡量数据的离散程度，根据离散程度判断数据点与均值的偏差，将偏差过大的数据确定为异常值；Outlier detection involves calculating the mean of all data points, measuring the central tendency of the data, and using the standard deviation Measure the degree of dispersion of the data, determine the deviation of the data points from the mean based on the degree of dispersion, and identify data with excessive deviation as outliers; 所述多模态数据进行融合包括，The multimodal data fusion includes: 基于卡尔曼滤波对预处理后的多模态传感数据进行融合，通过预处理后的多模态数据，设定起重机械的位置、速度和加速度信息，将设定的起重机械的位置、速度和加速度信息作为输入数据，根据牛顿运动学方程对起重机械的运行状态进行预测；The pre-processed multi-modal sensor data is fused based on Kalman filtering. The position, velocity and acceleration information of the crane are set through the pre-processed multi-modal data. The set position, velocity and acceleration information of the crane are used as input data to predict the operating status of the crane according to Newton's kinematic equations. 对起重机械的运行状态进行预测具体表示为：The specific prediction of the operating status of the lifting machinery is as follows: ； ; ； ; ； ; ； ; 其中，表示当前时间步的状态估计向量，表示状态转移矩阵，表示控制输入矩阵，表示过程噪声，表示当前时间步的控制输入，表示时间步长，、、表示在三个轴上的加速度值；in, Indicates the current time step The state estimation vector, represents the state transition matrix, represents the control input matrix, represents the process noise, Indicates the current time step The control input, represents the time step, 、 、 Indicates the acceleration value on three axes; 采用状态转移模型输入前一时刻的状态信息，结合预处理后的加速度，计算起重机在下一时刻的位置、速度和加速度，定义运动规律；The state transition model is used to input the state information of the previous moment, combined with the pre-processed acceleration, to calculate the position, velocity and acceleration of the crane at the next moment and define the motion law; 计算卡尔曼增益，根据卡尔曼增益利用测量数据对预测状态进行修正，得到融合后的位置信息、速度信息、加速度信息；Calculate the Kalman gain and use the measurement data to correct the predicted state based on the Kalman gain to obtain the fused position information, velocity information, and acceleration information; 计算卡尔曼增益包括，通过测量矩阵、测量噪声协方差、状态估计的协方差矩阵输入卡尔曼增益计算公式，得到卡尔曼增益；Calculating the Kalman gain includes inputting the Kalman gain calculation formula through the measurement matrix, the measurement noise covariance, and the state estimation covariance matrix to obtain the Kalman gain; 状态估计的协方差矩阵衡量预测值的不确定性，动态调整预测的置信程度；The covariance matrix of the state estimate measures the uncertainty of the predicted value and dynamically adjusts the confidence level of the prediction; 测量矩阵将状态变量映射到测量空间，计算修正值；The measurement matrix maps the state variables to the measurement space and calculates the correction values; 测量噪声协方差矩阵根据传感器自身误差特性进行设置，反映测量数据的不确定性，优化状态修正的准确性；The measurement noise covariance matrix is set according to the sensor's own error characteristics to reflect the uncertainty of the measurement data and optimize the accuracy of state correction; 所述构建LSTM-GRU预测模型包括，The construction of the LSTM-GRU prediction model includes: 将融合后的位置信息、速度信息、加速度信息数据作为输入，将输入的位置信息、速度信息、加速度信息数据进行归一化处理，通过归一化处理后的位置信息、速度信息、加速度信息数据组成时间序列；The fused position information, velocity information, and acceleration information data are used as input, the input position information, velocity information, and acceleration information data are normalized, and the normalized position information, velocity information, and acceleration information data are used to form a time series; 将传统LSTM预测模型引入GRU计算，构建GRU层与全连接层；Introducing the traditional LSTM prediction model into GRU calculation, constructing the GRU layer and the fully connected layer; 将时间序列作为输入，通过GRU层适应起重机械即时运动变化，通过LSTM层输出起重机械的整体轨迹趋势，结合GRU层与LSTM层的输出结果进行轨迹预测计算，输出预测轨迹点；The time series is used as input, and the GRU layer adapts to the real-time motion changes of the crane. The LSTM layer outputs the overall trajectory trend of the crane. The output results of the GRU layer and LSTM layer are combined to perform trajectory prediction calculation and output the predicted trajectory points. 将历史位置、速度、加速度数据输入预测模型，在均方误差计算的基础上引入轨迹平滑惩罚项，构建均方误差损失函数，将历史轨迹预测结果输入均方误差损失函数，计算历史预测轨迹与真实轨迹之间的误差；The historical position, velocity, and acceleration data are input into the prediction model. Based on the mean square error calculation, a trajectory smoothing penalty term is introduced to construct a mean square error loss function. The historical trajectory prediction results are input into the mean square error loss function to calculate the error between the historical predicted trajectory and the actual trajectory. 将多模态数据进行融合的融合结果输入LSTM-GRU预测模型，输出起重机械运行轨迹。The fusion result of multimodal data is input into the LSTM-GRU prediction model to output the operation trajectory of the lifting machinery. 2.如权利要求1所述的起重机械防碰撞检测方法，其特征在于：所述通过AI边缘计算设备动态生成防碰撞控制指令包括，2. The crane anti-collision detection method according to claim 1, wherein the dynamic generation of anti-collision control instructions by the AI edge computing device includes: 将预测模型输出的轨迹数据与障碍物的位置信息作为输入，计算起重机械与障碍物之间的最小安全距离，根据最小安全距离设定安全距离阈值；The trajectory data output by the prediction model and the location information of the obstacle are used as input to calculate the minimum safe distance between the crane and the obstacle, and the safety distance threshold is set based on the minimum safe distance. 根据设定的安全距离阈值，生成控制策略并将控制策略转化为控制指令数据；Generate a control strategy based on the set safety distance threshold and convert the control strategy into control instruction data; 设定安全距离阈值包括，当起重机到障碍物的距离小于警告距离，发出预警信号；Setting the safety distance threshold includes issuing an early warning signal when the distance between the crane and the obstacle is less than the warning distance; 当起重机到障碍物的距离小于制动距离，判断执行刹车操作；When the distance between the crane and the obstacle is less than the braking distance, the braking operation is performed; 生成控制策略包括，根据设定的安全距离阈值，进行控制状态设置，当起重机到障碍物的距离大于警告距离，起重机械正常运行；Generating a control strategy includes setting the control state according to the set safety distance threshold. When the distance between the crane and the obstacle is greater than the warning distance, the crane operates normally. 当起重机到障碍物的距离处于制动距离与警告距离之间，进行声光报警；When the distance between the crane and the obstacle is between the braking distance and the warning distance, an audible and visual alarm will be issued; 当起重机到障碍物的距离等于制动距离，向PLC发送减速指令，降低起重机运行速度；When the distance between the crane and the obstacle is equal to the braking distance, a deceleration command is sent to the PLC to reduce the crane's operating speed; 当起重机到障碍物的距离小于制动距离，向PLC发送刹车指令，停止起重机械工作。When the distance between the crane and the obstacle is less than the braking distance, a braking command is sent to the PLC to stop the crane from working. 3.如权利要求2所述的起重机械防碰撞检测方法，其特征在于：所述PLC接受并执行防碰撞控制指令包括，3. The anti-collision detection method for hoisting machinery according to claim 2, wherein the PLC receives and executes the anti-collision control instruction including: 将控制指令数据作为输入，输入PLC控制执行单元；Take the control instruction data as input and input it into the PLC control execution unit; 起重机械正常运行指令下PLC控制执行单元保持原有运行速度，不触发报警，不调整变频器频率，不触发刹车系统；Under normal operation instructions of the lifting machinery, the PLC controls the execution unit to maintain the original operating speed, does not trigger the alarm, does not adjust the inverter frequency, and does not trigger the brake system; 进行声光报警指令下PLC控制执行单元触发LED指示灯闪烁，蜂鸣器鸣响，在控制面板或远程监控系统上显示警告信息，不调整起重机速度，不触发刹车系统；Under the sound and light alarm command, the PLC controls the execution unit to trigger the LED indicator to flash, the buzzer to sound, and the warning message to be displayed on the control panel or remote monitoring system, without adjusting the crane speed or triggering the brake system; 降低起重机运行速度指令下PLC控制执行单元进行速度调整，调整变频器输出频率，保持声光报警，不触发刹车系统；Reduce the crane's operating speed. The PLC controls the execution unit to adjust the speed and the inverter output frequency, maintaining the sound and light alarms without triggering the brake system. 停止起重机械工作指令下PLC控制执行单元切断起重机驱动电机电源，触发刹车系统，执行紧急制动，保持声光报警，发送停机报告。When the crane stops working, the PLC controls the execution unit to cut off the power supply of the crane drive motor, trigger the brake system, perform emergency braking, maintain the sound and light alarm, and send a shutdown report. 4.一种采用如权利要求1~3任一所述的起重机械防碰撞检测方法的系统，其特征在于：包括数据采集与预处理模块（100），数据融合与预测模块（200），指令生成与执行模块（300）；4. A system using the anti-collision detection method for lifting machinery according to any one of claims 1 to 3, characterized in that it comprises a data acquisition and preprocessing module (100), a data fusion and prediction module (200), and an instruction generation and execution module (300); 所述采集与预处理模块（100）用于采集起重机械的多模态数据，将多模态数据进行预处理；The acquisition and preprocessing module (100) is used to acquire multimodal data of the lifting machinery and preprocess the multimodal data; 所述数据融合与预测模块（200）用于通过卡尔曼滤波对采集的多模态数据进行融合，构建LSTM-GRU预测模型，将融合数据输入LSTM-GRU预测模型对起重机械运行轨迹进行预测；The data fusion and prediction module (200) is used to fuse the collected multimodal data through Kalman filtering, build an LSTM-GRU prediction model, and input the fused data into the LSTM-GRU prediction model to predict the operation trajectory of the lifting machinery; 所述指令生成与执行模块（300）用于根据轨迹预测结果，通过AI边缘计算设备动态生成防碰撞控制指令，PLC接受并执行防碰撞控制指令。The instruction generation and execution module (300) is used to dynamically generate anti-collision control instructions through the AI edge computing device according to the trajectory prediction result, and the PLC receives and executes the anti-collision control instructions. 5.一种计算机设备，包括存储器和处理器，所述存储器存储有计算机程序，其特征在于，所述处理器执行所述计算机程序时实现权利要求1至3中任一项所述的起重机械防碰撞检测方法的步骤。5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein the processor implements the steps of the crane anti-collision detection method according to any one of claims 1 to 3 when executing the computer program. 6.一种计算机可读存储介质，其上存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现权利要求1至3中任一项所述的起重机械防碰撞检测方法的步骤。6. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the crane anti-collision detection method according to any one of claims 1 to 3 are implemented.
CN202510497510.2A 2025-04-21 2025-04-21 A crane anti-collision detection method, system, device and medium Active CN120024817B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202510497510.2A CN120024817B (en) 2025-04-21 2025-04-21 A crane anti-collision detection method, system, device and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202510497510.2A CN120024817B (en) 2025-04-21 2025-04-21 A crane anti-collision detection method, system, device and medium

Publications (2)

Publication Number Publication Date
CN120024817A CN120024817A (en) 2025-05-23
CN120024817B true CN120024817B (en) 2025-08-26

Family

ID=95738222

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202510497510.2A Active CN120024817B (en) 2025-04-21 2025-04-21 A crane anti-collision detection method, system, device and medium

Country Status (1)

Country Link
CN (1) CN120024817B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN120353205B (en) * 2025-06-23 2025-09-05 精兵特种装备(福建)有限公司 Equipment target control data acquisition system and acquisition method thereof

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118877740A (en) * 2024-09-10 2024-11-01 宁波越海码头经营有限公司 A blind spot monitoring system for quay crane operation
CN119787348A (en) * 2025-01-16 2025-04-08 安徽华赛能源科技股份有限公司 A wind power off-grid hydrogen production power supply topology and control method without step-down transformer

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11263896B2 (en) * 2020-04-06 2022-03-01 B&H Licensing Inc. Method and system for detecting jaywalking of vulnerable road users
CN114180460B (en) * 2021-11-01 2024-06-25 珠海三一港口机械有限公司 Anti-collision protection method, controller, system and quay crane for lifting appliance
CN114940166A (en) * 2022-05-05 2022-08-26 清华大学 Pedestrian anti-collision protection method, device, equipment and medium based on trajectory prediction
CN118419787A (en) * 2024-05-28 2024-08-02 中铁二十二局集团第三工程有限公司 Anti-collision method based on collision prevention early warning system of tower crane
CN119612367A (en) * 2025-01-16 2025-03-14 唐波 Intelligent tower crane anti-collision precision lifting system based on multi-sensor fusion

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118877740A (en) * 2024-09-10 2024-11-01 宁波越海码头经营有限公司 A blind spot monitoring system for quay crane operation
CN119787348A (en) * 2025-01-16 2025-04-08 安徽华赛能源科技股份有限公司 A wind power off-grid hydrogen production power supply topology and control method without step-down transformer

Also Published As

Publication number Publication date
CN120024817A (en) 2025-05-23

Similar Documents

Publication Publication Date Title
CN104723350B (en) Intelligent control method and system for industrial robot safety protection
CN119476939B (en) An intelligent seepage monitoring system for water conservancy projects
CN120024817B (en) A crane anti-collision detection method, system, device and medium
CN111392619A (en) Tower crane early warning method, device and system and storage medium
CN119683499B (en) Tower crane collision early warning system with multi-mode data fusion
CN118419788B (en) Intelligent control system for approaching protection and anti-collision protection of monorail crane personnel
US12321245B2 (en) Method and system for reconciling values of a feature
CN119785454A (en) Dynamic monitoring system for safe driving behavior and safe production of mobile machinery on-site based on AI big model
CN120538536A (en) Mobile robot path planning method based on deep reinforcement learning
CN119810570A (en) Crane identification system based on artificial intelligence
CN120122670B (en) A high-precision stacker crane intelligent obstacle avoidance system and method based on hybrid guidance
CN120097222B (en) A monitoring and early warning method and system for container gantry crane
CN120610558B (en) Cloud edge coordination management and control platform and method
CN119348624B (en) Obstacle sensing method of industrial vehicle ADAS (advanced automatic analysis and attachment system) fused with large model
CN120573601A (en) Overhead crane positioning control method and related equipment
CN120802953A (en) Robot moving path planning method, apparatus, electronic device and storage medium
CN120542924A (en) A digital twin-based active control method, system, and equipment for major engineering construction safety
CN119658721A (en) Robot rescue emergency control system based on voice interaction
CN120088188A (en) Steel billet tilt detection method, device, system and medium
CN119197511A (en) A mobile navigation system for intelligent service robots
CN117270385A (en) A human-robot fusion decision-making method with weight adaptive adjustment
CN117058211A (en) Grab anti-sway and collision strategy control method and system based on laser positioning
CN119339580B (en) A remote human-machine co-driving system, method and device
CN119526430B (en) Robot-based operation falling monitoring and early warning method and system
CN120039807A (en) Intelligent forklift management system and method based on AI anti-collision

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant