CN118419788B

CN118419788B - Intelligent control system for approaching protection and anti-collision protection of monorail crane personnel

Info

Publication number: CN118419788B
Application number: CN202410895026.0A
Authority: CN
Inventors: 薄福利; 连涛; 吴敬建; 刘朋朋; 殷宏杰; 高贵国; 孙祥勇; 李恩光; 李斌; 郭灿; 王文启
Original assignee: Yangcheng Coal Mine Of Shandong Jikuang Luneng Coal Power Co ltd
Current assignee: Yangcheng Coal Mine Of Shandong Jikuang Luneng Coal Power Co ltd
Priority date: 2024-07-05
Filing date: 2024-07-05
Publication date: 2024-09-06
Anticipated expiration: 2044-07-05
Also published as: CN118419788A

Abstract

The invention belongs to the technical field of intelligent control, and particularly relates to an intelligent control system for approaching protection and anti-collision protection of monorail crane personnel. The system comprises: the multi-source data acquisition and fusion unit is used for acquiring the operation data and the surrounding environment data of the monorail crane through a plurality of sensors to form a multi-source data set; nonlinear fusion is carried out on the multi-source data set to obtain a fusion multi-source data set; the personnel tracking and predicting unit is used for tracking personnel trajectories, predicting future trajectories of each personnel based on the personnel trajectories, and obtaining predicted trajectories of each personnel; and the collision analysis unit is used for calculating a dangerous area according to the fusion multisource data set, then calculating a collision risk value based on the predicted track, and controlling the monorail crane to brake when the collision risk value exceeds a set first risk threshold value. The invention obviously improves the safety, reliability and intelligent level of the approach protection and anti-collision protection system of monorail crane personnel.

Description

Intelligent control system for approaching protection and anti-collision protection of monorail crane personnel

Technical Field

The invention belongs to the technical field of intelligent control, and particularly relates to an intelligent control system for approaching protection and anti-collision protection of monorail crane personnel.

Background

The monorail crane is used as an important transport means and is widely applied to a plurality of fields such as industry, mining industry, construction and the like. With the development of industrial automation and intelligence, the use frequency and the operation complexity of the monorail crane are continuously increased. However, in practical operation, the safety problem of the monorail crane is always a key technical problem. Particularly in workplaces with dense personnel and complex environment, how to effectively prevent the monorail crane from colliding with personnel or other equipment is a problem to be solved.

Currently, there are a number of technical solutions for the safety control of monorail cranes in the market. These techniques can be broadly divided into three categories: sensor-based detection techniques, image recognition-based monitoring techniques, and wireless communication-based positioning techniques. The detection technology based on the sensor is mainly used for monitoring the environmental information around the monorail crane in real time through various sensors (such as an infrared sensor, an ultrasonic sensor, a laser radar and the like) arranged on the monorail crane. When the sensor detects an obstacle or a person approaching, the system will give an alarm and take corresponding braking action. The technology has the advantages of quick response, convenient installation and the like, but in a complex environment, the sensor is easily interfered by external factors such as dust, rainwater and the like, so that the detection precision is reduced. The monitoring technology based on image recognition captures images around the monorail crane in real time through a camera, and utilizes a computer vision algorithm to analyze the images so as to identify potential risk factors. Such techniques may provide more intuitive and comprehensive environmental information, but their accuracy depends on the performance of the image processing algorithm. In the case of insufficient light or severe occlusion, the accuracy and reliability of image recognition may be compromised. In addition, the image processing requires higher computing resources, and higher requirements are put on the real-time performance of the system. The positioning technology based on wireless communication utilizes signal intensity or time difference to calculate the positions of all equipment and personnel by arranging wireless beacons in the monorail crane and the surrounding environment. Such techniques can provide higher accuracy positioning information, but are costly to implement, requiring a large number of beacon devices to be deployed in a work environment. Moreover, the signal is susceptible to reflections, multipath effects, etc., resulting in unstable positioning accuracy.

Disclosure of Invention

In view of the above, the main purpose of the invention is to provide an intelligent control system for the approach protection and the anti-collision protection of monorail crane personnel, which significantly improves the safety, reliability and intelligent level of the approach protection and the anti-collision protection system of monorail crane personnel.

The technical scheme adopted by the invention is as follows:

Intelligent control system of monorail crane personnel proximity protection and anticollision protection, the system includes: the multi-source data acquisition and fusion unit is used for acquiring the operation data and the surrounding environment data of the monorail crane through a plurality of sensors to form a multi-source data set; nonlinear fusion is carried out on the multi-source data set to obtain a fusion multi-source data set; the personnel tracking and predicting unit is used for tracking personnel trajectories, predicting future trajectories of each personnel based on the personnel trajectories, and obtaining predicted trajectories of each personnel; and the collision analysis unit is used for calculating a dangerous area according to the fusion multisource data set, then calculating a collision risk value based on the predicted track, and controlling the monorail crane to brake when the collision risk value exceeds a set first risk threshold value.

Further, the operation data at least includes: the speed components of the monorail crane in the X axis, the Y axis and the Z axis respectively, the angular speed components of the monorail crane around the X axis, the Y axis and the Z axis respectively, and the rolling angle, the pitch angle and the yaw angle of the monorail crane; the environmental data includes at least: the components of wind speed in the X-axis, Y-axis and Z-axis directions and the initial three-dimensional coordinates of each person, respectively.

Further, the multi-source data set is set as; Wherein, Representing a three-dimensional coordinate set consisting of initial three-dimensional coordinates of each person, which takes the position of the monorail crane as an origin,Is the firstThe X-axis coordinate values of the individual person,Is the firstThe Y-axis coordinate values of the individual persons,Is the firstZ-axis coordinate values of the individuals; Representing a speed set of the monorail crane, 、AndThe speed components of the monorail crane in the X axis, the Y axis and the Z axis are respectively represented; representing a set of wind speeds, 、AndComponents of wind speed in X-axis, Y-axis and Z-axis directions are respectively represented; A set of angular velocities is represented and, 、AndRespectively representing angular velocity components of the monorail crane around an X axis, a Y axis and a Z axis; Represents a deflection angle set of the monorail crane, 、AndRespectively representing the roll angle, pitch angle and yaw angle of the monorail crane.

Further, the multi-source data acquisition and fusion unit uses an improved unscented kalman filter to perform nonlinear fusion on the multi-source data set, and the specific process comprises the following steps: a prediction step, an updating step and an adaptive factor updating step; wherein, the formula of the prediction step is as follows:

；

The formula of the updating step is as follows:

；

Wherein, Is the sigma point set at time k-1; is the state estimate at time k-1; is a scaling parameter of sigma point distribution; Is the covariance matrix at time k-1; is the control input at time k; is a parameter of the state transfer function; is a preset nonlinear state transfer function; Is a preset nonlinear observation function; is a parameter of the observation function; And Respectively the firstMean and covariance weights of the sigma points; Is an adaptive process noise covariance matrix used to represent uncertainty of state prediction; is a self-adaptive observation noise covariance matrix and represents the uncertainty of observation; is the actual observed value at time k; Is the kalman gain; is a process noise adaptive factor for dynamically adjusting a process noise covariance matrix; is an observation noise self-adaptive factor and is used for dynamically adjusting an observation noise covariance matrix; Is the cross covariance matrix between the states and observations; The posterior state estimation at the moment k, namely the updated optimal estimation, is used as a fusion multisource data set; the posterior covariance matrix at the moment k represents the updated state estimation uncertainty; Is the covariance matrix of the predicted observations; Is a sigma point set propagated through a state transfer function; is a priori state estimate at time k; Is a priori covariance matrix at time k; the result of sigma point set transmitted by observation function; is a predicted observation; Is propagated through state transfer function A sigma point; Is the first The result of the sigma points after being transmitted through the observation function; sigma point totalAnd each.

Further, the formula of the adaptive factor updating step is as follows:

；

Wherein, Is a process noise adaptive factorControlling the rate of change thereof; Is a process noise adaptive factor To influence the sensitivity of the variation thereof; Is an adaptive factor of observation noise Controlling the rate of change thereof; Representing the L2 norm.

Further, the person tracking and predicting unit tracks the person track, predicts the future track of each person based on the person track, and the process of obtaining the predicted track of each person is expressed by using the following formula:

；

Wherein, The operator is a Caputo fractional derivative operator; Is the order; is the time step; for a predicted time interval; The three-dimensional coordinate system is a personnel three-dimensional coordinate set function at the time t, and is a curve function, wherein the sum of the distances between a curve represented by the curve function and each three-dimensional coordinate in the personnel three-dimensional coordinate set is minimum; Rounding down the symbol; Is a Gaussian noise matrix; preset of The transfer matrix is a Caputo fractional derivative transfer matrix; by solving the formula, we obtainAs a predicted trajectory for the person; refer to Caputo, meaning that this is the fractional derivative of Caputo.

Further, the Caputo fractional derivative transfer matrix is expressed using the following formula:

;

Wherein, Is a gamma function.

Further, the collision analysis unit calculates the hazard zone from the fused multisource dataset by the following formula:

;

Wherein, The area of the dangerous area is round with the monorail crane as the center; The radius of the base of the monorail crane is the radius; The longest radial length of the monorail crane; The deviation angle of the monorail crane relative to the horizontal plane; for a second element of the fused multisource dataset, corresponding to a monorail crane velocity set in the fused multisource dataset; for a third element of the fused multisource dataset, corresponding to a wind speed set in the fused multisource dataset; for a fourth element of the fused multisource dataset, corresponding to an angular velocity set in the fused multisource dataset; the operation of vector modulus is obtained; Is the maximum allowable speed; is the maximum allowable wind speed; Is the maximum allowable load; ， And The roll angle, the pitch angle and the yaw angle respectively correspond to the fusion multisource data set; To adjust the coefficients.

Further, the adjustment coefficient is calculated by the following formula:

；

Wherein, As an initial value, the value range is 0.5 to 1.3; The maximum allowable value of the mode of the multi-source data set is a set value.

By adopting the technical scheme, the invention has the following beneficial effects: the invention realizes the comprehensive perception of the surrounding environment of the monorail crane by the multisource data fusion technology. The system acquires environmental data in real time by utilizing various sensors (such as an infrared sensor, an ultrasonic sensor, a laser radar and the like) arranged on the monorail crane, acquires visual information by combining an image recognition technology, and accurately positions the monorail crane and surrounding personnel by utilizing a wireless positioning technology. By fusion and comprehensive analysis of the multi-source data, the system can comprehensively grasp the running state and the environmental information of the monorail crane, and can still provide high-precision detection and analysis results in complex and changeable environments. The invention also realizes the prediction and early warning of potential danger. The collision analysis unit combines the multi-source data set, calculates the dangerous area through a mathematical formula, and timely gives an alarm and takes corresponding measures when the potential collision risk is detected. Specifically, the system dynamically calculates the size and shape of the hazard zone using parameters such as speed vector, wind speed vector, angular speed vector, and attitude angle that are fused with the multisource dataset. When the system detects that a person or object enters a dangerous area, an alarm is immediately sent out, and emergency braking or other safety measures are adopted according to requirements so as to prevent collision accidents. The prediction and early warning mechanism obviously reduces the occurrence probability of accidents and improves the safety of the system.

Drawings

Fig. 1 is a schematic system structure diagram of an intelligent control system for approaching protection and anti-collision protection of monorail crane personnel according to an embodiment of the present invention.

Detailed Description

All of the features disclosed in this specification, or all of the steps in a method or process disclosed, may be combined in any combination, except for mutually exclusive features and/or steps.

Any feature disclosed in this specification may be replaced by alternative features serving the same or equivalent purpose, unless expressly stated otherwise. That is, each feature is one example only of a generic series of equivalent or similar features, unless expressly stated otherwise.

Example 1: referring to fig. 1, an intelligent control system for monorail crane personnel access protection and collision avoidance, the system comprising: the multi-source data acquisition and fusion unit is used for acquiring the operation data and the surrounding environment data of the monorail crane through a plurality of sensors to form a multi-source data set; nonlinear fusion is carried out on the multi-source data set to obtain a fusion multi-source data set; the personnel tracking and predicting unit is used for tracking personnel trajectories, predicting future trajectories of each personnel based on the personnel trajectories, and obtaining predicted trajectories of each personnel; and the collision analysis unit is used for calculating a dangerous area according to the fusion multisource data set, then calculating a collision risk value based on the predicted track, and controlling the monorail crane to brake when the collision risk value exceeds a set first risk threshold value.

Specifically, in the intelligent control system for the approach protection and the anti-collision protection of monorail crane personnel, the multi-source data acquisition and fusion unit plays a critical role, and is the basis for the whole system to sense the environment and collect information. The working principle of the unit is based on the cooperative work of multiple sensors, and the limitation of a single sensor is overcome by integrating the advantages of different types of sensors, so that the comprehensive and accurate sensing of the monorail overhead travelling environment is realized. In practical applications, this unit may comprise various types of data acquisition devices such as vision sensors, lidar, ultrasonic sensors, infrared sensors, etc. The vision sensor can capture image information of the environment, and is helpful for identifying and tracking personnel; The laser radar can accurately measure the distance and the shape, and is beneficial to constructing a three-dimensional environment map; the ultrasonic sensor is suitable for detecting short-distance obstacles; the infrared sensor can then function in the event of insufficient light. The sensors are distributed at key positions of the monorail crane to form an omnibearing sensing network. Each sensor has unique data format and characteristics, and the maximum utility of the sensor is difficult to be achieved by simply superposing the heterogeneous data. Therefore, multi-source data fusion techniques have evolved. The core of the technology is to organically combine data with different sources and different characteristics to generate a more complete and accurate environment representation. In a monorail crane system, the data fusion process firstly needs to preprocess the original data acquired by each sensor, and comprises the steps of noise filtering, data calibration and the like so as to ensure the data quality. The system then aligns the data of the different sensors in time and space, which is critical because only correctly aligned data can be fused effectively. The system then integrates the preprocessed data using a nonlinear fusion algorithm. Nonlinear fusion is important because in a complex industrial environment, there is often a nonlinear relationship between various factors, and it is difficult for conventional linear fusion methods to accurately describe this complexity. Common nonlinear fusion methods include kalman filtering, particle filtering, deep learning, and other techniques. Taking deep learning as an example, a neural network can be designed, data of different sensors are taken as input, and a comprehensive environment representation is finally output through multi-layer nonlinear transformation. This approach has the advantage of being able to automatically learn complex relationships between data without the need for manually designing features. In the monorail crane system, the fused data set not only contains the running state information of the monorail crane, such as position, speed, load and the like, but also contains the dynamic information of the surrounding environment, such as personnel position, moving obstacles and the like. This comprehensive environmental awareness provides a solid data base for subsequent personnel tracking, trajectory prediction, and collision risk analysis. For example, the fused data can more accurately locate surrounding personnel, and even if a certain sensor temporarily fails due to occlusion or other reasons, the system can still rely on the data of other sensors to maintain the perception of the environment. In addition, the multisource data fusion can also improve the anti-interference capability and reliability of the system. In harsh industrial environments, a single type of sensor may be affected by factors such as dust, vibration, electromagnetic interference, and the like. By fusing the data of various sensors, the system can cross-verify the accuracy of information and filter abnormal data, thereby ensuring the reliability of the sensing result. This feature is particularly important for ensuring the safety of monorail crane operation, as any environmental perceived error may lead to serious safety accidents.

The personnel tracking and predicting unit acquires real-time images and three-dimensional space data of the surrounding environment of the monorail crane by using sensing equipment such as cameras, laser radars and the like. After preprocessing, the visual image captured by the camera is input into a deep learning model for target detection and recognition. Convolutional Neural Networks (CNNs) are powerful image processing tools that can effectively identify personnel objects in images and label the positions of the personnel and bounding boxes. For the three-dimensional space data of the laser radar, the system can generate accurate personnel position and posture information through a point cloud processing algorithm. The data are comprehensively processed through a multi-sensor data fusion technology to form a high-precision personnel position information set. After the position information of the person is acquired, the system enters a person track tracking stage. Trajectory tracking is a continuous process aimed at generating a moving trajectory for each person by analyzing the change in person's position between the previous and subsequent frames. Based on algorithms such as a Kalman filter or a particle filter, the system can carry out smooth processing on the personnel track, eliminate noise and uncertainty, and improve the accuracy and the continuity of the track. Meanwhile, the system also adopts a multi-target tracking (MOT) technology, and each target is ensured to be independently and accurately tracked under a multi-person scene through data association and track management. Personnel tracking data provides a basis for future trajectory predictions. Future trajectory prediction is based on historical trajectory data and current environmental information, and utilizes machine learning and statistical modeling techniques to predict a future movement path of a person. A long-short-term memory network (LSTM) is used as a recurrent neural network suitable for time series data, and can effectively capture time sequence characteristics in a personnel track. The system inputs the historical track data into an LSTM model, and generates future track prediction of personnel in a short period by learning a motion mode and a behavior rule in the track. Further, a statistical model such as a Hidden Markov Model (HMM) may be used to predict the future state of the person, and the possible position distribution of the person at the next time may be estimated by using the state transition probability and the observation model. The result of personnel trajectory prediction is critical to collision avoidance decisions. The system evaluates potential collision risks by comparing the predicted personnel trajectories with the running trajectories of the monorail crane. Specifically, the system calculates the relative position and velocity of the monorail crane and the person at a future time, and calculates a collision risk value based thereon. When the collision risk value exceeds a preset safety threshold, the system can give an alarm and take braking measures to avoid personnel injury and equipment damage.

The collision analysis unit needs to comprehensively analyze the fusion data provided by the multi-source data acquisition and fusion unit. Such data includes real-time operating conditions (e.g., position, speed, acceleration) of the monorail crane and information about the surrounding environment (e.g., obstacle location, personnel location, etc.). In order to accurately calculate the motion trail of the monorail crane, the collision analysis unit simulates the motion trail of the monorail crane by using a physical model, and takes kinetic parameters such as mass, friction, inertia and the like into consideration to obtain a predicted trail of the monorail crane in a future period of time. Meanwhile, future tracks of the personnel provided by the personnel tracking and predicting unit can be used as input data to participate in the collision risk assessment.

After obtaining the predicted trajectories of the monorail crane and the person, the collision analysis unit will enter a path planning and risk calculation stage. Path planning algorithms, such as the a-algorithm and Dijkstra algorithm, will be used to identify possible collision paths for monorail and personnel. These algorithms find areas where they are likely to collide by spatially and temporally matching future locations of monorail and personnel. To further increase the accuracy of the collision prediction, the collision analysis unit may dynamically adjust in connection with the environmental data, e.g. taking into account new obstacles detected in real time or changes in the position of the person. The risk of collision is assessed by calculating the relative position and velocity of the monorail crane and the person at a future time. The collision analysis unit will analyze these relative positions and velocities using a probabilistic model to estimate the probability of a collision. The probabilistic model may take into account various uncertainty factors such as sensor measurement errors, uncertainty in personnel movements, etc. to provide a more accurate risk assessment result. Based on these analysis results, the collision analysis unit calculates a collision risk value reflecting the probability and severity of a collision of the monorail crane with a person at a future time. When the collision risk value exceeds a preset safety threshold value, the collision analysis unit immediately triggers a collision protection mechanism. The anti-collision protection mechanism comprises an alarm signal sending part and an automatic braking part. When the system sends out an alarm signal, the operator is reminded of the potential danger and can take corresponding measures to intervene. If the risk of collision continues to increase or has approached an unavoidable level, the system automatically controls the monorail crane to make an emergency brake, slowing or stopping the movement of the monorail crane, thereby avoiding the occurrence of collisions. The innovation of the collision analysis unit is its highly intelligent and automated risk assessment and protection mechanism. Compared with the traditional single sensor or simple rule driven system, the invention adopts multi-source data fusion, advanced path planning algorithm and probability model, so that collision analysis is more accurate and reliable. The multi-sensor data fusion provides a more comprehensive and accurate environment perception, the path planning algorithm ensures the accurate identification of collision risks in a complex environment, the probability model introduces uncertainty analysis for risk assessment, and the robustness and reliability of prediction are improved.

Example 2: the operation data at least comprises: the speed components of the monorail crane in the X axis, the Y axis and the Z axis respectively, the angular speed components of the monorail crane around the X axis, the Y axis and the Z axis respectively, and the rolling angle, the pitch angle and the yaw angle of the monorail crane; the environmental data includes at least: the components of wind speed in the X-axis, Y-axis and Z-axis directions and the initial three-dimensional coordinates of each person, respectively.

Specifically, in the system, the running data of the monorail crane are acquired in real time through a series of high-precision sensors. The linear velocity component, which is the movement velocity of the monorail crane in three directions X, Y, Z in space, is provided by an accelerometer and a GPS module. The sensors monitor the movement of the monorail crane in all directions in real time, so that the system can accurately predict the future position of the monorail crane, and the occurrence of collision events is effectively prevented. At the same time, the angular velocity component, i.e. the rotational velocity of the monorail crane about its own X, Y, Z axes, is measured by means of a gyroscope. These data are critical to understanding the rotational motion of the monorail, as any change in angular velocity affects the overall trajectory of the monorail. In order to fully grasp the attitude of the monorail crane, the system also monitors roll angle, pitch angle and yaw angle, which are provided by attitude sensors and Inertial Measurement Units (IMUs), ensuring accurate positioning and control of the monorail crane in three dimensions. The acquisition of environmental data is also critical, especially in complex operating environments. The components of wind speed in three directions X, Y, Z are measured by means of anemometers mounted around the monorail crane. These data provide detailed information about the external wind forces to which the monorail crane is subjected, enabling the system to evaluate the effect of wind forces on the stability of the monorail crane and to make motion compensation and adjustments as necessary. In addition, the initial three-dimensional coordinates of each person are provided by environmental sensors such as lidar and cameras. The sensors generate three-dimensional coordinate data of each person by detecting and analyzing the positions of the persons in the environment, so that the system can accurately identify and track the motion trail of the person. Once this detailed operational and environmental data is collected, the system will be processed and analyzed by advanced algorithms. Firstly, the system fuses the linear velocity data and the angular velocity data of the monorail crane to form a complete motion model. The model considers the movement speed and the rotation speed of the monorail crane in different directions, and accurately describes the dynamic behavior of the monorail crane in space by combining the gesture data. By real-time analysis of this data, the system is able to predict future positions and attitudes of the monorail crane, thereby preventing possible collisions. In the aspect of environmental data processing, the system combines wind speed data with a motion model of the monorail crane to evaluate the influence of external wind power on the operation of the monorail crane. The variation of the wind velocity component may cause deflection or instability of the monorail crane, so that the system needs to make the necessary adjustments to ensure smooth running of the monorail crane based on the wind velocity data. The initial three-dimensional coordinate data of the person is used for tracking and predicting the moving track of the person. By analyzing the personnel position data, the system can generate the motion trail of each personnel and predict the future moving path of each personnel. These predicted trajectories are compared to the motion trajectories of the monorail crane to identify potential collision risks. The collision analysis unit uses the processed data to calculate the relative position and speed between the monorail crane and the person, and evaluates the collision risk. When the system detects that the collision risk value exceeds a preset safety threshold, an anti-collision protection mechanism is triggered immediately, wherein the anti-collision protection mechanism comprises an alarm signal and automatic braking, and the safety of personnel and equipment is ensured.

Example 3: let multisource data set be; Wherein, Representing a three-dimensional coordinate set consisting of initial three-dimensional coordinates of each person, which takes the position of the monorail crane as an origin,Is the firstThe X-axis coordinate values of the individual person,Is the firstThe Y-axis coordinate values of the individual persons,Is the firstZ-axis coordinate values of the individuals; Representing a speed set of the monorail crane, 、AndThe speed components of the monorail crane in the X axis, the Y axis and the Z axis are respectively represented; representing a set of wind speeds, 、AndComponents of wind speed in X-axis, Y-axis and Z-axis directions are respectively represented; A set of angular velocities is represented and, 、AndRespectively representing angular velocity components of the monorail crane around an X axis, a Y axis and a Z axis; Represents a deflection angle set of the monorail crane, 、AndRespectively representing the roll angle, pitch angle and yaw angle of the monorail crane.

Specifically, the system collects operation data and environmental data in real time through various sensors. Initial three-dimensional coordinate set of personnelThe position of each person in the space is included, the position of the monorail crane is taken as an origin, and the coordinate values of each person in the directions of the X axis, the Y axis and the Z axis are recorded. These data are acquired by sensors such as lidar, cameras, etc., ensuring that the system is able to accurately locate and track the position of each person. These initial three-dimensional coordinates are the basis for the system to perform personnel trajectory prediction and collision risk assessment, providing accurate data support for subsequent trajectory analysis. Meanwhile, the system collects the speed data of the monorail craneIncluding velocity components in three directions, the X-axis, the Y-axis, and the Z-axis. The speed data are monitored in real time through the high-precision accelerometer and the GPS module, so that the system can accurately acquire the movement speed of the monorail crane in all directions. The velocity component data are critical to predicting the future position and trajectory of the monorail as they directly affect the motion state and trajectory planning of the monorail. In addition, the system also collects wind speed dataIncluding the components of wind speed in three directions, the X-axis, the Y-axis and the Z-axis. These data are obtained by means of anemometers mounted around the monorail crane, and the system evaluates the effect of the wind on the stability of the monorail crane by analysing the wind speed data and, if necessary, adjusting the movement of the monorail crane to ensure its smooth running. Accurate acquisition and processing of wind speed data are helpful for maintaining stability and safety of the monorail crane under complex environmental conditions. Angular velocity dataAlso one of the key data of the system includes the angular velocity components of the monorail crane about the X, Y and Z axes. These data are measured by gyroscopes and angular velocity data are critical to understanding the rotational motion of the monorail, as any change in angular velocity affects the overall trajectory of the monorail. By monitoring and analyzing the angular velocity data in real time, the system can accurately predict the rotation motion of the monorail crane, and ensure the accurate control and stable operation of the monorail crane in a three-dimensional space. Finally, the system collects deflection angle data of the monorail craneIncluding roll angle, pitch angle, and yaw angle. These data are provided by attitude sensors and Inertial Measurement Units (IMUs) to ensure accurate positioning and control of the monorail crane in three dimensions. The deflection angle data reflects the posture change of the monorail crane, and has important significance for preventing the monorail crane from tilting or deviating from a track in the running process. After the detailed operation data and the environment data are acquired and processed, the system fuses the data through a complex algorithm to form a comprehensive and multidimensional environment perception model. The model combines the speed, angular speed and deflection angle data of the monorail crane and the initial three-dimensional coordinate and wind speed data of personnel, and can accurately describe the dynamic changes of the monorail crane and the surrounding environment thereof. By analyzing and processing the data in real time, the system can predict the future position and track of the monorail crane and evaluate the collision risk by combining the motion track of the personnel. When the system detects a potential collision risk, the protection mechanism is triggered immediately, including alarm generation and automatic braking, so as to ensure the safety of personnel and equipment.

Example 4: the multi-source data acquisition and fusion unit uses an improved unscented Kalman filter to carry out nonlinear fusion on a multi-source data set, and the specific process comprises the following steps: a prediction step, an updating step and an adaptive factor updating step; wherein, the formula of the prediction step is as follows:

；

the core of the prediction step is the generation of a sigma point set . These sigma points are state estimates from the last instant by an unscented transformationSum covariance matrixThe purpose of the extraction is to capture the mean and variance of the state distribution. In particular, the manner in which sigma points are generated ensures that the nonlinear nature of the state distribution is preserved, which is critical to accurately predicting the motion state of the monorail crane. The sigma point generation process includes estimating in a stateFor the center, add and subtract a scaling parameterMultiplying by covariance matrixThereby forming a set of points that can represent the state distribution. Next, a sigma point setBy state transfer functionPropagating to generate sigma point set of predicted time. State transfer functionIs a nonlinear function incorporating a control inputAnd dynamic parameters of the systemAnd simulating the movement process of the monorail crane. This function describes the evolution of the parameters of the monorail crane, including speed, position and attitude, in terms of state change under controlled conditions. By propagating sigma points, the system is able to simulate the possible state distribution of the monorail crane at future times, which is critical for predicting the motion trajectory of the monorail crane and identifying potential collision risks. After the propagated sigma point set is obtainedThe system then performs a weighted average of these sigma points to calculatePriori state estimation of time of day. The process of weighted averaging uses specific weightsThe weights are set according to the distribution characteristics of sigma points, so that the average result can accurately reflect the average value of the state. The purpose of this step is to extract a most representative predicted state from a plurality of possible state points for use in a subsequent updating step.

At the same time, the system also calculates a priori covariance matrix. The covariance matrix reflects the uncertainty of the state estimate by weighted averaging the variance of sigma points and taking into account the process noise covariance matrixThe system is able to quantify the uncertainty of the predicted state. Process noise covariance matrixIs based on process noise adaptation factorAnd the dynamic adjustment reflects the state change of the system under different noise environments. This step ensures that the system can adapt to uncertainty factors in the environment, improving the robustness and accuracy of the predictions. The core of the unscented Kalman filter is the application of unscented transforms throughout the prediction process. Conventional kalman filters perform well when dealing with linear systems, but their performance tends to be limited when faced with nonlinear systems. And the unscented Kalman filter can more accurately handle the nonlinear problem by introducing unscented transforms. The unscented transformation approximates the mean and covariance of the state distribution with a carefully selected set of sigma points, thereby avoiding approximation errors in the linearization process. This allows unscented Kalman filters to offer advantages over Extended Kalman Filters (EKFs) in handling complex nonlinear systems, providing greater accuracy and stability. In an intelligent control system for approaching protection and anti-collision protection of monorail crane personnel, the high precision and the real-time performance of the prediction step are important to the safety and the reliability of the system. By accurately predicting the motion state of the monorail crane, the system can identify potential collision risks in advance, and necessary preventive measures are taken to ensure the safe operation of the monorail crane. For example, when the system predicts that the monorail crane will approach a person at some time in the future, an alarm can be given in time and braking measures can be taken to prevent collision accidents. The efficient prediction and protection mechanism not only improves the operation safety of the monorail crane, but also remarkably improves the intelligent level of the system. In addition, the introduction of the self-adaptive factor in the prediction step enables the system to dynamically adjust the covariance matrix of the process noise and the observation noise, and further enhances the adaptability and the robustness of the filter. In complex and changeable environments, the noise level may change at any time, and the adjustment of the adaptive factor ensures that the system can maintain the optimal performance under different noise conditions, and improves the accuracy and stability of prediction. The adaptive factor is used to make the process noise covariance matrixAnd observed noise covariance matrixCan be dynamically adjusted according to real-time data to ensure that the filter maintains optimal performance in the face of different noise levels. Process noise adaptation factorAnd observation noise adaptation factorThe dynamic adjustment of (3) enables the system to flexibly cope with environmental changes and provides more accurate and reliable state estimation. Through complex calculation and accurate adjustment of the prediction steps, the intelligent control system for the approach protection and the anti-collision protection of monorail crane personnel can realize efficient and reliable multi-source data fusion and state prediction. This not only improves the safety and reliability of the system, but also provides valuable experience and reference for future development of similar systems. The successful application of the prediction step shows the strong capability of the unscented Kalman filter in processing a nonlinear system, and provides a solid technical guarantee for intelligent safety control of the monorail crane.

The formula of the updating step is as follows:

；

the core of the updating step is to combine the prior state estimation obtained in the predicting step with the actual observation data and map the sigma points from the state space to the observation space by a nonlinear observation function. Observation function Is a nonlinear function describing the observation process of the sensor and reflecting how the system acquires observation data from the actual environment. The input of the observation function is sigma point set after being propagated through the state transfer functionThe output is the set of observed sigma points. These observed sigma point setsBy reflecting various aspects of the monorail crane operating conditions, including position, speed, angle and environmental factors, a comprehensive observation data set is provided. Next, a predicted observation is calculated by weighted averaging the set of observed sigma points. Weight of weighted averageSetting is carried out according to the distribution characteristics of sigma points, so that the average result can accurately reflect the average value of the observed value. Predicting observationsIs an estimate of the actual observations that reflects the observations expected by the system in a given state. After obtaining the predicted observations, the system needs to calculate the covariance matrix of the predicted observationsAnd a cross covariance matrix between states and observations. The calculation of these covariance matrices takes into account the observed noise covariance matricesThe matrix is adaptive to the observed noiseDynamically adjusts to reflect uncertainty in the observation process. The adjustment of the covariance matrix of the observation noise enables the system to adapt to different observation environments and noise levels, and accuracy and reliability of observation data are ensured.

Next, the system calculates a kalman gainThis is the key formula in the update step. Cross covariance matrix between Kalman gain pass state and observationAnd a covariance matrix of predicted observationsIs multiplied by the inverse matrix of (c). The effect of the kalman gain is to measure the difference between the predicted state and the actual observed value and use this difference to correct the prior state estimate. By means of the Kalman gain, the system can observe errorsThe correction amount is converted into the correction amount of the state estimation, so that the accuracy of the state estimation is improved. After computing the Kalman gain, the system uses the gain to estimate the prior stateCorrecting to obtain posterior state estimation. Specifically, posterior state estimationIs obtained by adding the product of the Kalman gain and the observed error to the prior state estimate. This process ensures that the system is able to dynamically adjust the state estimate based on the most current observations, providing a more accurate and reliable state description. Posterior state estimationAs an updated best state estimate, it will be used in the prediction step at the next moment. At the same time, the system also updates the posterior covariance matrixTo reflect the uncertainty of the updated state estimate. The updating of the posterior covariance matrix includes subtracting the product of the Kalman gain and the covariance matrix of the predicted observations and the Kalman gain. Through this updating step, the system is able to quantify the uncertainty of the state estimate, ensuring that high accuracy state predictions and risk assessments can continue to be provided at future times. In an intelligent control system for approaching protection and anti-collision protection of monorail crane personnel, the high precision and the real-time performance of the updating step are critical to the safety and the reliability of the system. By accurately calculating the Kalman gain and dynamically adjusting the state estimation, the system can timely respond to environmental changes and fluctuation of observed data, so that potential collision risks can be effectively prevented. For example, when the system detects that the monorail crane approaches a person during operation, the state estimation is quickly corrected through the updating step, an alarm is sent out, braking measures are taken, and collision accidents are prevented. The efficient updating mechanism not only improves the operation safety of the monorail crane, but also remarkably improves the intelligent level of the system.

In addition, the introduction of the self-adaptive factor in the updating step enables the system to dynamically adjust the observed noise covariance matrix, and further enhances the adaptability and the robustness of the filter. In complex and changeable environments, the noise level may change at any time, and the adjustment of the adaptive factor ensures that the system can maintain the optimal performance under different noise conditions, and improves the accuracy and stability of the updating step. The adaptive factor is used to make the observed noise covariance matrixCan be dynamically adjusted according to real-time data to ensure that the filter maintains optimal performance in the face of different noise levels. By means of complex calculation and accurate adjustment of updating steps, the intelligent control system for the approach protection and the anti-collision protection of monorail crane personnel can achieve efficient and reliable multi-source data fusion and state updating. This not only improves the safety and reliability of the system, but also provides valuable experience and reference for future development of similar systems. The successful application of the updating step shows the strong capability of the unscented Kalman filter in processing a nonlinear system, and provides a solid technical guarantee for intelligent safety control of the monorail crane.

Example 5: the formula of the adaptive factor updating step is as follows:

；

Specifically, the first formula of the adaptive factor update step is used to adjust the process noise adaptive factor. The formula dynamically adjusts this factor by the error between the actual observations and the predicted observations. When the error is large, the process noise adaptation factor should be increased to reflect the higher uncertainty. Specifically, when the error exceeds a certain threshold, the system will rapidly increase the process noise adaptive factor to make the process noise covariance matrixAnd increases, thereby increasing tolerance to uncertainty in the predictions. This design ensures that the system can quickly adjust the predictive model as the noise level changes, maintaining a high accuracy state estimate. For example, during monorail crane operation, if environmental data measured by the sensor is subject to large errors due to external disturbances, the system can recognize this change and adjust the process noise adaptation factor accordingly. When the wind speed changes severely, the motion state of the monorail crane can be changed obviously, and the system can enhance the adaptability to the changes by increasing the process noise self-adaptive factor, so that the stability and the reliability of state estimation are ensured. The second formula is used to adjust the observed noise adaptation factor. This formula adjusts the observation noise adaptation factor by squaring the observation error. When the observation error is large, the observation noise adaptation factor is reduced to reflect the increase in the observation noise. The design concept is to improve the credibility of the observed data by reducing the influence of the observed noise. When the observation error is smaller, the adjustment amplitude of the observation noise adaptive factor can be reduced, and the relative stability of the observation noise covariance matrix is maintained. In an intelligent control system for approaching protection and anti-collision protection of monorail crane personnel, the real-time adjustment of the adaptive factor of observation noise is particularly important. For example, when the system detects that the fluctuation of the position data of the personnel in the surrounding environment is large, the influence of the fluctuation on the state estimation can be reduced by adjusting the observation noise adaptive factor, and the recognition accuracy of the system on the position of the actual personnel is improved. This is particularly critical in a scene where the personnel are intensive or the environment is complex, and can significantly improve the safety and reliability of the system. Through the application of the two formulas, the intelligent control system for approaching protection and anti-collision protection of monorail crane personnel can adaptively adjust the levels of process noise and observation noise, so that the optimal performance is maintained under different environmental conditions. The dynamic adjustment mechanism of the self-adaptive factor enables the system to respond to the change of the observation error in real time, and the robustness and the precision of the filter are improved. Specifically, when the system is in operation and the sensor data generates a large error due to external interference, the adaptive factor updating step can rapidly identify the change and correspondingly adjust the noise covariance matrix. For example, when the monorail crane is operated in a strong wind environment, sudden changes in wind speed may cause increased sensor measurement errors, and the system may reflect this uncertainty by increasing the process noise adaptation factor, improving the robustness of state prediction. Likewise, when the observation noise of the sensor increases, the system increases the importance of the observation data by reducing the observation noise adaptive factor, thereby improving the accuracy of the state update. In the intelligent control system for the approach protection and the anti-collision protection of monorail crane personnel, the implementation of the self-adaptive factor updating step obviously improves the adaptability and the reliability of the system in complex and changeable environments. The dynamic adjustment mechanism ensures that the system can adjust the levels of process noise and observation noise in real time according to the change of actual observation data, thereby improving the precision and stability of state estimation. Through the mechanism, the system can keep high-efficiency and safe operation in various complex environments, and can timely identify and prevent potential collision risks, so that the safety of personnel and equipment is protected. In addition, the formula design of the self-adaptive factor updating step enables the system to have high intelligent and automatic level. By means of real-time analysis and dynamic adjustment of the observation errors, the system can adaptively optimize the performance of the filter without manual intervention. The intelligent adjustment mechanism not only improves the operation convenience of the system, but also obviously improves the reliability and stability of the system, and provides a solid technical guarantee for intelligent control of the monorail crane.

Example 6: the personnel tracking and predicting unit tracks personnel trajectories, predicts future trajectories of each personnel based on the personnel trajectories, and the process of obtaining predicted trajectories of each personnel is represented by using the following formula:

；

Specifically, in the formula,Expressed in timeA set of three-dimensional coordinates of the person at the moment. This function captures and records the position data of the person in space by means of high-precision sensors, forming a continuous track curve. The distance between the curve and the three-dimensional coordinate point of the person is minimum, so that the accuracy and consistency of the track are ensured. This means that at any given moment the system is able to provide a most likely trajectory of the person's position and make predictions of future positions based thereon. Caputo fractional derivative operatorIs one of the core components of the formula. Fractional derivatives are generalized differential operators that can handle more complex dynamic system characteristics. Here, the Caputo fractional derivative is used to describe the change in the motion profile of the person. By differentiating fractional orders of past motion data, the system is able to better understand and predict future motion trends of a person. The formula uses Gaussian noise matrixIt reflects random disturbances in the environment and measurement errors. In the real world, any measurement and movement is inevitably affected by various random factors. The introduction of the Gaussian noise matrix allows the model to take these uncertainties into account, thereby improving the robustness and reliability of the prediction results. By introducing noise terms, the system can simulate uncertainty in the real environment and make corresponding adjustments in the prediction process. To achieve prediction of future trajectories of persons, the formula is passed through a time stepAnd predicting time intervalsTo perform the calculation. Time stepDetermining how often the system performs trajectory computation within a predicted time intervalThe predicted time window length is determined. The system calculates the changes in the position of the person within each time step in a stepwise iterative manner and predicts the future track position based on these changes. In practical application, the system captures three-dimensional coordinate data of a person at different time points through a sensor, and inputs the data into a track tracking model. These time series data are then processed using the Caputo fractional derivative operator. In the formulaDescribing the change rate of the personnel position at the current moment, based on which the system can predict the future personnel position. Meanwhile, the transfer matrix in the formulaPlays a key role. The transfer matrix incorporates historical position data to help predict future motion trajectories by analyzing past motion trajectories. Gaussian noise matrixRandom disturbance is introduced in the calculation process to simulate uncertainty factors in the real environment. The method not only considers the inertia and the historical track of the personnel movement, but also fully considers the randomness and the uncertainty possibly existing in the environment, so that the prediction result has more practical significance. For example, in one practical scenario, as the system detects that a person is moving about the monorail crane, the sensor will continuously track the trajectory data of that person. By means of the above formula, the system can process this data, predicting the possible position of the person within a few seconds of the future. If the prediction shows that the person will enter the working area of the monorail crane, the system will immediately give an alarm and take corresponding precautions, such as slowing down or emergency braking, to prevent a possible collision accident. The application of the Caputo fractional derivative enables the system to handle more complex trajectory variations, while the introduction of gaussian noise matrices increases the robustness and reliability of the predictions. This combination enables the system to maintain a high degree of accuracy in a complex and versatile environment. The system can provide accurate prediction results through dynamic adjustment and real-time calculation no matter the random movement of personnel or sudden interference in the environment. In summary, the personnel tracking and predicting unit achieves high-precision prediction of future trajectories of personnel by using complex mathematical models of the Caputo fractional derivatives and gaussian noise matrices. The process not only improves the understanding and predicting capability of the system to the motion trail of the personnel, but also enhances the adaptability and the robustness of the system in a complex environment. Through real-time tracking and dynamic prediction, the system can timely identify potential collision risks and take effective preventive measures to ensure safe operation of the monorail crane. The highly intelligent and automatic track prediction mechanism provides a solid technical guarantee for intelligent control of the monorail crane, and simultaneously shows great potential and value of a modern intelligent control system in industrial application.

Example 7: the Caputo fractional derivative transfer matrix is expressed using the following formula:

;

Wherein, Is a gamma function.

Specifically, the core of the Caputo fractional derivative transfer matrix is that it can correlate historical data with current state, thereby leveraging past information in predicting future trajectories. This is particularly important when dealing with human motion trajectories, as human motion tends to have memory and inertial characteristics. By applying fractional derivatives, the system can not only consider the state at the current moment, but also comprehensively consider the state changes at a plurality of moments in the past, thereby realizing more accurate prediction. Specifically, the order of the fractional order derivativeThe sensitivity of the system to historical data is determined. When (when)When the system is close to 1, the dependency of the system on past data is strong, more historical information can be captured, and otherwise, the dependency is weak. Gamma functionThe effect in this formula is to weight each historical data point within a time step so that the effect of each data point on the current state is reasonably quantified. Transfer matrixBy weighted summation of the data at the past instants. This process involves a segmentation of the time, i.e. the time is divided into several stepsAnd processes the data in each time step. The introduction of the gamma function enables the system to accurately weight the data in each time step, ensuring that the influence of each data point is reasonably considered. In a specific formula calculation, the matrix is transferred by summing the historical dataThe influence of the past data on the current state can be accurately reflected, so that the prediction accuracy is improved. In practice, when the sensor captures the movement of a person in the vicinity of the monorail crane, this data is first recorded and entered into a trajectory tracking model. The system processes the historical data through the Caputo fractional derivative transfer matrix to predict the future motion trail of the personnel. The prediction process not only considers the current position and speed of the personnel, but also combines the past movement trend, thereby providing a more comprehensive and accurate prediction result.

Example 8: the collision analysis unit calculates a dangerous area according to the fused multi-source data set by the following formula:

;

Wherein, The area of the dangerous area is round with the monorail crane as the center; The radius of the base of the monorail crane is the radius; The longest radial length of the monorail crane; The deviation angle of the monorail crane relative to the horizontal plane; for a second element of the fused multisource dataset, corresponding to a monorail crane velocity set in the fused multisource dataset; for a third element of the fused multisource dataset, corresponding to a wind speed set in the fused multisource dataset; for a fourth element of the fused multisource dataset, corresponding to an angular velocity set in the fused multisource dataset; The current load of the monorail crane is carried; the operation of vector modulus is obtained; Is the maximum allowable speed; is the maximum allowable wind speed; Is the maximum allowable load; ， And The roll angle, the pitch angle and the yaw angle respectively correspond to the fusion multisource data set; To adjust the coefficients.

Specifically, the core part of the formula isIt represents the area of the circular hazard zone around the monorail crane centered around it. Calculating this area requires a comprehensive consideration of the physical dimensions, motion state, environmental impact of the monorail crane and the current loading situation. Firstly, the basic shape of the hazard zone is a circle with a radius defined by the radius of the base of the monorail craneAnd longest radial lengthIs determined by the sinusoidal components of (a). Specifically, in the formulaShowing the area of a circular region centered on a monorail crane, whereinIs the deviation angle of the monorail crane from the horizontal plane. This part calculates the basic geometry of the monorail crane in space, taking into account its physical dimensions and attitude. Radius of baseAnd radial lengthAre critical parameters because they directly affect the working coverage of the monorail crane. As the boom of the monorail crane increases or the angle to the horizontal becomes larger, the radius of the hazardous area will correspondingly increase and vice versa. On the basis, the further correction factor is obtained by an exponential functionThe function combines the effects of speed, wind speed and load capacity of the monorail crane. Here the number of the elements is the number,Representing monorail crane velocity vectors in the fused multisource dataset,The wind velocity vector is represented as such,The angular velocity vector is represented by a vector of angular velocities,Representing the current load capacity. The normalization factors of these parameters are respectively the maximum allowable speedsMaximum allowable wind speedAnd maximum allowable load capacity. By normalizing these parameters, the formula can accommodate different operating conditions and environmental changes. The effect of the exponential function is to amplify the effect of these variables, ensuring that the hazard zone can be properly enlarged at high speeds, high wind speeds or high loads, thereby providing a greater safety buffer. In particular, velocity vectorThe impact on the hazardous area is particularly pronounced. As the speed of the monorail crane increases, the uncertainty in its path of movement increases, thus requiring a greater hazard area to ensure safety. Also, wind speed vectorThe ambient wind speed represented has an important influence on the stability of the monorail crane. The larger the wind speed is, the more remarkable the stress change of the monorail crane is, and the deviation of the movement track is increased. Thus, in high wind conditions, the system may properly enlarge the hazard zone to cope with possible motion deviations.

Load weightIs also a key parameter. With the increase of the carrying capacity, the inertia of the monorail crane is increased, the motion response time is prolonged, and the control precision is reduced. Thus, under heavy load conditions, the hazardous area needs to be properly enlarged to provide more reaction time and space to prevent accidental collisions. Another part of the formulaThe calculation of the hazard zone is further refined. Adjustment coefficientAnd modulus of angular velocity vectorThe sensitivity of the formula is adjusted, and the system is ensured to accurately reflect the influence of the angular speed on the dangerous area. This section also includes three orientation attitude angles (roll, pitch and yaw) that are respectively determined by fusion of the multisource datasets、AndAnd (3) representing. These angles are converted into the actual effect on the size of the hazard zone by means of an inverse cosine function. These attitude angles are critical to determining the precise position and direction of movement of the monorail in three dimensions. For example, the roll angle, pitch angle and yaw angle describe the rotation of the monorail about its three axes, respectively, and these angle changes directly affect the actual working range and path of motion of the monorail. Therefore, the system can more accurately predict dangerous areas of the monorail crane under different postures by accurately calculating the angles. In a combined view, the formula reflects the importance of multi-source data fusion in an intelligent control system. By combining real-time operation data of the monorail crane with environmental factors, the system can dynamically adjust the size of the dangerous area, thereby preventing collision risk more accurately. For example, as monorail crane travel speeds increase, the hazard zone may correspondingly expand to provide more reaction time and space. Likewise, when wind speed increases or monorail crane load increases, the system will adjust the hazardous area accordingly, ensuring safety under various operating conditions. In practice this means that the system can monitor the change in its surroundings in real time as the monorail crane moves on the track and adjust the hazard zone according to the current operating conditions. If a person or object is detected entering this dynamically adjusted hazard zone, the system will immediately alert and take emergency braking or other safety measures as needed to prevent a collision accident. By comprehensively considering the physical dimensions, motion state, environmental impact and current load capacity of the monorail crane, the system can provide high-precision dangerous area prediction in a complex environment. This approach not only improves the safety of the system, but also enhances its adaptability under different operating conditions. For example, in environments with large wind speeds, the wind speed vectorCan have a significant impact on the size of the hazardous area. By adjusting the weight of the wind speed, the system can accurately reflect the influence of the wind speed on the running safety of the monorail crane. Likewise, when the monorail crane is carrying a heavy object, the load weightThe increase in (c) will expand the hazardous area to ensure that the system remains safe to operate when the load is high. Furthermore, the angular velocity vectorAnd the introduction of the attitude angle enables the system to more comprehensively consider the dynamic behavior of the monorail crane in the three-dimensional space. Through careful analysis of the roll angle, the pitch angle and the yaw angle, the system can accurately predict dangerous areas of the monorail crane under different postures. Through the multi-dimensional and multi-factor comprehensive consideration, the collision analysis unit not only can calculate the dangerous area in real time, but also can dynamically adjust the size of the dangerous area according to the change of the environment and the operation condition, thereby providing optimal safety protection. This dynamic adjustment mechanism ensures efficient and safe operation of the system in a variety of complex environments.

In summary, the collision analysis unit calculates the dangerous area around the monorail crane through a complex mathematical formula, and comprehensively considers a plurality of factors such as geometric parameters, motion states, environmental influences, current carrying capacity and the like of the monorail crane. Through the size of dynamic adjustment danger area, the system can prevent potential collision risk in real time, ensures the safe operation of monorail crane. The highly intelligent and automatic dangerous area prediction mechanism provides a solid technical guarantee for intelligent control of the monorail crane, and also shows great potential and value of a modern intelligent control system in industrial application. The method is not only suitable for personnel access protection and anti-collision protection of the monorail crane, but also provides references for other similar industrial equipment. In various complex and dynamic industrial environments, the system can provide high-precision prediction and early warning through multi-source data fusion and complex mathematical modeling, so that the overall safety and the operation efficiency are improved. The intelligent control system can play an important role in mines, ports and other industrial places, and ensure the safety of equipment and personnel.

Example 9: the adjustment coefficient is calculated by the following formula:

；

Wherein, For the initial value, the value range is 0.5 to 1.3.

Specifically, the formula describes the adjustment coefficientWhereinIs an initial value ranging between 0.5 and 1.3. Initial valueThe basic adjustment level of the system in the initial state is determined, and the system can be set according to the safety requirements of specific application scenes. For example, in a high risk environment, a higher initial value may be selected to increase sensitivity in the hazardous area, thereby improving safety. Multi-source data set in a formula, multi-source data setVarious parameters related to the operation of the monorail crane are included, including speed, acceleration, angular velocity, attitude angle, etc. These parameters comprehensively reflect the real-time operating state and environmental conditions of the monorail crane. Multiple source data set modulo multiple source data set by normalization processingRepresenting the combined effect of the current stateThe maximum allowed value of the modulus of these multisource data sets is used to normalize the current state. The formula performs nonlinear transformation on the normalized multisource data set through application of the hyperbolic tangent function. The value of the hyperbolic tangent function ranges between-1 and 1, which enables the adjustment coefficient to react sensitively to changes in the multisource data set. When the modulus of the multi-source data set is close to zero, the value of the hyperbolic tangent function is close to zero; when the modulus of the multisource data set approaches its maximum allowable value, the value of the hyperbolic tangent function approaches 1. This nonlinear transformation ensures that the system is able to adjust reasonablyThereby adapting to different operating conditions. For example, when the speed or angular velocity of the monorail crane is high, the modulus of the multisource data set is large and the output of the hyperbolic tangent function is also large, thereby makingAccordingly, dangerous areas are increased, and safety is improved. In addition, the adjustment coefficientTime of passage variableAnd (5) performing index adjustment. The exponential function in the equation shows that, over time,Dynamically changes to accommodate changes in long-term operation. The design enables the system to maintain high sensitivity to dangerous areas during long-term operation, and ensures safety under different environmental conditions. For example, when the monorail overhead crane has long running time or the environment changes severely,And correspondingly, the number of the dangerous areas is increased to provide a larger dangerous area, so that the operation safety is ensured. In practical application, when the monorail crane runs on the track, the system monitors the surrounding environment and the running state in real time and collects multi-source data sets according to the multi-source data setsDynamic adjustment of variation of (a). For example, in a high wind speed environment, wind speed parameters can significantly affect the modulus of the multisource data set, and thus the hyperbolic tangent functionIs a value of (2). The greater the wind speed, the greater the modulus of the multisource data set,The value of (2) is correspondingly increased, expanding the dangerous area. This ensures that the system is able to provide a greater safety buffer against accidental collisions at high wind speeds. Likewise, when the speed of the monorail crane increases or the angular speed changes greatly, the system adjustsIncreasing the sensitivity of the hazardous area and thus providing greater safety protection. For example, when the monorail crane is accelerating, the system recognizes an increase in speed by adjustingThe hazardous area is extended, ensuring that there is enough reaction time and space to handle the potentially hazardous situation. To sum up, the adjustment coefficientThe calculation formula of the system ensures that the intelligent control system for approaching protection and anti-collision protection of monorail crane personnel can adapt to various complex and dynamic operation conditions by dynamically adjusting the size of the dangerous area. By reasonably setting initial valuesAnd nonlinear transformation of the multisource data set, the system can sensitively respond to real-time changes and provide optimal security protection. The highly intelligent adjustment mechanism not only improves the safety and reliability of the system, but also shows great potential and value of the modern intelligent control system in industrial application. By the method, the operation of the monorail crane becomes safer and more efficient, and the safety of personnel and equipment is ensured.

While specific embodiments of the present invention have been described above, it will be understood by those skilled in the art that these specific embodiments are by way of example only, and that various omissions, substitutions, and changes in the form and details of the methods and systems described above may be made by those skilled in the art without departing from the spirit and scope of the invention. For example, it is within the scope of the present invention to combine the above-described method steps to perform substantially the same function in substantially the same way to achieve substantially the same result.

Claims

1. Intelligent control system of monorail crane personnel proximity protection and anticollision protection, its characterized in that, the system includes: the multi-source data acquisition and fusion unit is used for acquiring the operation data and the surrounding environment data of the monorail crane through a plurality of sensors to form a multi-source data set; nonlinear fusion is carried out on the multi-source data set to obtain a fusion multi-source data set; the personnel tracking and predicting unit is used for tracking personnel trajectories, predicting future trajectories of each personnel based on the personnel trajectories, and obtaining predicted trajectories of each personnel; the collision analysis unit is used for calculating a dangerous area according to the fusion multisource data set, then calculating a collision risk value based on the predicted track, and controlling the monorail crane to brake when the collision risk value exceeds a set first risk threshold value;

The operation data at least comprises: the speed components of the monorail crane in the X axis, the Y axis and the Z axis respectively, the angular speed components of the monorail crane around the X axis, the Y axis and the Z axis respectively, and the rolling angle, the pitch angle and the yaw angle of the monorail crane; the environmental data includes at least: the components of wind speed in the directions of an X axis, a Y axis and a Z axis and the initial three-dimensional coordinates of each person respectively;

Let multisource data set be ; Wherein, Representing a three-dimensional coordinate set consisting of initial three-dimensional coordinates of each person, which takes the position of the monorail crane as an origin,Is the firstThe X-axis coordinate values of the individual person,Is the firstThe Y-axis coordinate values of the individual persons,Is the firstZ-axis coordinate values of the individuals;

Representing a speed set of the monorail crane, 、AndThe speed components of the monorail crane in the X axis, the Y axis and the Z axis are respectively represented; representing a set of wind speeds, 、AndComponents of wind speed in X-axis, Y-axis and Z-axis directions are respectively represented; A set of angular velocities is represented and, 、AndRespectively representing angular velocity components of the monorail crane around an X axis, a Y axis and a Z axis; Represents a deflection angle set of the monorail crane, 、AndRespectively representing the rolling angle, the pitch angle and the yaw angle of the monorail crane;

The multi-source data acquisition and fusion unit uses an improved unscented Kalman filter to carry out nonlinear fusion on a multi-source data set, and the specific process comprises the following steps: a prediction step, an updating step and an adaptive factor updating step; wherein, the formula of the prediction step is as follows:

；

The formula of the updating step is as follows:

；

2. The intelligent control system for proximity protection and collision protection of monorail crane personnel as claimed in claim 1, wherein the formula of the adaptive factor updating step is as follows:

；

3. The intelligent control system for protecting a monorail crane personnel from approaching and collision protection as set forth in claim 2, wherein the personnel tracking and predicting unit tracks personnel trajectories and predicts future trajectories for each personnel based on the personnel trajectories, and the process of obtaining predicted trajectories for each personnel is represented using the following formula:

；

4. The intelligent control system for monorail operator proximity protection and collision avoidance as claimed in claim 3, wherein the Caputo fractional derivative transfer matrix is expressed using the following formula:

;

Wherein, Is a gamma function.

5. The intelligent control system for the approaching protection and collision avoidance of a monorail crane personnel as defined in claim 4, wherein the collision analysis unit calculates the hazard zone from the fused multisource data set by the following formula:

;

6. The intelligent control system for protecting personnel approaching and anti-collision protection of monorail crane of claim 5, wherein the adjustment coefficientThe calculation is performed by the following formula:

；