Loading and transporting machinery safety intelligent auxiliary system and method

Abstract

The invention discloses a loading and transporting machinery safety intelligent auxiliary system and method, and relates to the field of intelligent safety systems.The method comprises the steps that a deep learning model is used for fusing preprocessed environment data, an optimal path from a current position to a destination is calculated through a search algorithm, and based on the optimal path, the optimal path is calculated; according to the method, the deep learning technology is utilized to calculate the potential safety hazard on the optimal path, the calculated potential safety hazard is prompted to an operator, and the operator checks the operation state and provides countermeasures so as to collect response data of the loading and transporting machinery. The output prediction value of the potential safety hazard is calculated by using the decision tree model, so that the potential safety hazard on the path is predicted, the system can identify potential risk points in advance, the calculated potential safety hazard is prompted to an operator, response data of loading and transporting machinery is collected, and the coping capacity of the operator to the potential safety hazard is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent security systems, and in particular to a safety intelligent assistance system and method for shipping machinery. Background Art

[0002] With the continuous improvement of industrial automation level, shipping machinery is more and more widely used in industries such as logistics and construction. Traditional shipping machinery mainly relies on manual operation, and it is easy to cause safety accidents due to human errors. The progress of sensor technology makes it more convenient and accurate to collect information about the environment around the machinery.

[0003] However, the existing safety intelligent assistance systems for shipping machinery still have deficiencies. First, most systems rely on a single type of sensor for data collection, resulting in incomplete environmental perception. Second, in terms of path planning, although existing research has optimized the path by combining deep learning models, these methods ignore the assessment of potential safety hazards on the path and cannot fully consider unexpected situations encountered in the actual operation process. In addition, the existing systems are also relatively weak in the operator feedback mechanism, lacking an effective closed-loop control strategy to dynamically adjust the action strategy of the machinery, which limits the adaptability and flexibility of the system. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a safety intelligent assistance system and method for shipping machinery, which solves the problems of incomplete environmental perception of shipping machinery and ignoring the assessment of potential safety hazards.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a safety intelligent assistance method for shipping machinery, which includes collecting environmental data around the started shipping machinery using a variety of sensors and performing preprocessing;

[0008] Fusing the preprocessed environmental data using a deep learning model, and calculating the best path from the current position to the destination through a search algorithm;

[0009] Based on the best path, calculating potential safety hazards on the best path using deep learning technology;

[0010] Prompting the calculated potential safety hazards to the operator, the operator checks the running state and provides countermeasures, thereby collecting response data of the shipping machinery;

[0011] Adjusting the running state of the shipping machinery according to the response data of the shipping machinery.

[0012] As a preferred embodiment of the safety intelligent assistance method for shipping machinery according to the present invention, among which: the multiple sensors include a high-definition camera, a radar, and a GPS receiver;

[0013] The environmental data includes distance, speed, and specific geographical location;

[0014] The preprocessed data includes converting the environmental data into a unified format by removing noise in the environmental data through filtering techniques and deleting outliers.

[0015] As a preferred embodiment of the safety intelligent assistance method for shipping machinery according to the present invention, among which: using a deep learning model to fuse the preprocessed environmental data and calculating the best path from the current location to the destination through a search algorithm, including the following steps,

[0016] Select a convolutional neural network model;

[0017] Extract the environmental features of the position, shape of obstacles, and distance, speed, and direction of the target object from the preprocessed environmental data, and fuse the extracted environmental features in a weighted manner;

[0018] Use the greedy best-first search algorithm to calculate the comprehensive score of each path from the current location to the destination and number them, and select the path with the highest comprehensive score as the best path. The expression is:

[0019]

[0020] F(p i )=C(p i )+α×∑ j≠i C(p j );

[0021]

[0022] Among them, C(p i ) represents the environmental feature score of the position point p i , Z represents the normalization constant, d(p i ) represents the distance of the target object at the position p i , η represents the attenuation coefficient of speed, v(p i ) represents the speed of the target object at the position p i , s(p i ) represents the size of the obstacle at the position p i , θ(p i ) represents the direction of the target object at the position p i , ||p i || represents the Euclidean norm of the position p i , F(p iIndicates the position point p i The comprehensive environmental characteristic score of, C(p j ) Indicates the position point p j The environmental characteristic score of, α indicates adjusting the position point p j The influence degree of the score on p i The weighted coefficient of the comprehensive score, j represents the index variable for traversing all position points in the path, i represents the index variable for identifying the position point p i Indicates the number of the best path with the highest comprehensive score, Indicates the number of the path that obtains the maximum value in all path sets, Indicates the number of one of the paths from the starting point to the destination, Y represents the number of all path sets, p i Indicates the path One of the position points in, g represents the destination, H(p i , g) Represents the estimated distance from the position p i To the destination g, o represents the total number of obstacles, Indicates the position p i The number of obstacles around, d(p i , o) Represents the distance from the position p i To the obstacle o.

[0023] As a preferred solution of the safety intelligent auxiliary method for the shipping machinery described in the present invention, wherein: based on the best path, the output prediction value of potential safety hazards on the best path is calculated by using deep learning technology, including the following steps,

[0024] Based on the best path with the highest comprehensive score, a decision tree model is constructed by using deep learning technology, and at the same time, the decision tree model is deployed to the actual environment by using an integrated development environment;

[0025] The geographical location is input into the decision tree model, and the output prediction value of potential safety hazards on the best path is calculated. The expression is:

[0026]

[0027] Among them, y represents the output prediction value of potential safety hazards, n represents the number of decision tree models, f represents the index of the decision tree, w i Represents the weight of the f-th decision tree, g f (x) Represents the prediction function of the f-th decision tree.

[0028] As a preferred solution of the safety intelligent auxiliary method for the shipping machinery described in the present invention, wherein: a threshold is set to divide the degree of potential safety hazards, including the following steps,

[0029] ​Calculate the mean and standard deviation of the output prediction values of potential safety hazards, and set low and high thresholds based on the mean and standard deviation to classify the levels of potential safety hazards;

[0030] When the output prediction value of a potential safety hazard is less than or equal to the low threshold, it is determined that the level of the potential safety hazard on the optimal path is a low risk;

[0031] When the output prediction value of a potential safety hazard is greater than the high threshold, it is determined that the level of the potential safety hazard on the optimal path is a high risk;

[0032] When the output prediction value of a potential safety hazard is greater than the low threshold and less than or equal to the high threshold, it is determined that the level of the potential safety hazard on the optimal path is a medium risk.

[0033] As a preferred solution of the safety intelligent assistance method for the shipping machinery described in the present invention, the potential safety hazards identified are prompted to the operator, and the operator checks the operating status and provides countermeasures, thereby collecting response data of the shipping machinery, including the following steps,

[0034] Set the display information for the level of potential safety hazards with low risk to green, medium risk to yellow, and high risk to red, and prompt the display information to the operator;

[0035] When the level of a potential safety hazard is a low risk, listen to the sounds in the surrounding environment and use a high-definition camera to monitor the environment, and save the monitoring data in a log;

[0036] When the level of a potential safety hazard is a medium risk, the shipping machinery switches to manual mode and reduces the operating speed, and reports the situation of medium risk to the relevant department;

[0037] When the level of a potential safety hazard is a high risk, stop the operation of all shipping machinery, arrange for all staff to evacuate to a safe area, and immediately contact team members to plan an alternative action path;

[0038] Collect the response data of the shipping machinery according to the countermeasures under the different risks described.

[0039] As a preferred solution of the safety intelligent assistance method for the shipping machinery described in the present invention, adjust the operating status of the shipping machinery according to the response data of the shipping machinery, including the following steps,

[0040] Collect response data including the speed and distance of the shipping machinery after the countermeasures;

[0041] If there is no deviation between the response data and the target speed and distance, operate the shipping machinery normally;

[0042] If there are deviations between the response data and the target speed and distance, immediately stop the operation of the shipping machinery, check the working status of the control unit and sensors of the machinery, repair the faulty components, and wait for the machinery to resume normal operation before continuing the operation;

[0043] If there is only a deviation in speed, remind the operator to pay attention and adjust the speed of the shipping machinery to return it to the target speed;

[0044] If there is only a deviation in distance, re-plan the route and notify the operator to drive according to the new route.

[0045] In a second aspect, the present invention provides a safety intelligent auxiliary system for shipping machinery, including an environmental data collection module, which is responsible for collecting the environmental data around the started shipping machinery using a variety of sensors and performing preprocessing;

[0046] An environmental data fusion and path planning module, which is responsible for fusing the preprocessed environmental data using a deep learning model and calculating the best path from the current location to the destination through a search algorithm;

[0047] A potential safety hazard identification module, which is responsible for calculating the potential safety hazards on the best path based on the best path using deep learning technology;

[0048] A safety prompt and countermeasure module, which is responsible for prompting the calculated potential safety hazards to the operator, and the operator checks the operating status and provides countermeasures;

[0049] A response data collection module, which is responsible for collecting the response data of the shipping machinery;

[0050] A dynamic adjustment module, which is responsible for adjusting the operating status of the shipping machinery according to the response data of the shipping machinery.

[0051] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the safety intelligent auxiliary method for shipping machinery as described in the first aspect of the present invention is implemented.

[0052] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the safety intelligent auxiliary method for shipping machinery as described in the first aspect of the present invention is implemented.

[0053] The beneficial effects of the present invention are as follows: By using a variety of sensors to collect environmental data and perform preprocessing, the comprehensive capture of environmental information and noise filtering are realized, and the quality of data and the accuracy of environmental perception are improved. Using deep learning technology to construct a decision tree model to calculate the output prediction value of potential safety hazards, the prediction of potential safety hazards on the path is realized, enabling the system to identify potential risk points in advance, and prompting the calculated potential safety hazards to the operator, collecting the response data of the shipping machinery, and improving the operator's ability to respond to potential safety hazards. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0055] Figure 1 It is a flowchart of the safety intelligent assistance method for shipping machinery in Embodiment 1.

[0056] Figure 2 It is a decision diagram of the safety intelligent assistance method for shipping machinery in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention with reference to the accompanying drawings of the specification.

[0058] In the following description, many specific details are set forth to facilitate a thorough understanding of the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0059] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that mutually excludes other embodiments.

[0060] Embodiment 1, referring to Figure 1 and Figure 2 , is the first embodiment of the present invention. This embodiment provides a safety intelligent assistance method for shipping machinery, including the following steps:

[0061] S1. Use a variety of sensors to collect the environmental data around the started shipping machinery and perform preprocessing, including the following steps.

[0062] Multiple sensors include high-definition cameras, radars, and GPS receivers;

[0063] Environmental data includes distance, speed, and specific geographical location;

[0064] Preprocessed data includes converting environmental data into a unified format after removing noise from the environmental data through filtering techniques and deleting outliers.

[0065] S2. Use a deep learning model to fuse the preprocessed environmental data and calculate the best path from the current location to the destination through a search algorithm, including the following steps:

[0066] Select a convolutional neural network model;

[0067] Extract the environmental features of the positions, shapes of obstacles, and distances, speeds, and directions of target objects from the preprocessed environmental data, and fuse the extracted environmental features through a weighted method.

[0068] Weighting refers to a processing method that assigns different degrees of importance to different data or variables in data analysis, statistical calculations, and models. The core idea of weighting is to give different weights according to the importance and credibility of each element. A weight is a numerical value that can reflect the relative importance of a specific data or factor in the final result. In mathematics and statistics, weighting is usually used to calculate weighted averages, weighted sums, or weighted fusions.

[0069] The greedy best-first search algorithm is a heuristic search algorithm. Its basic idea is to choose the currently seemingly optimal path to advance at each step, that is, always choose the node closest to the target node for expansion. If the target node is not reached, all unvisited neighbor nodes of the current node are added to the priority queue, and the values of each neighbor node are calculated.

[0070] Use the greedy best-first search algorithm to calculate the comprehensive score of each path from the current location to the destination and number them, and select the path with the highest comprehensive score as the best path. The expression is:

[0071]

[0072] F(p i ) = C(p i ) + α × ∑ j≠i C(p j );

[0073]

[0074] Among them, C(p i ) represents the position point p iThe environmental feature score, Z represents the normalization constant, which is used to normalize the scores of all location points, d(p i ) represents the position p i The closer the distance, the greater the distance. i ) is smaller, The larger the value, the more important the position is. η represents the attenuation coefficient of speed, which is used to adjust the influence of speed on the score. i ) represents the position p i The speed of the target object, the greater the speed, v(p i ) 2 The larger the value, the smaller the speed impact score of the position. i ) represents the position p i The size of the upper obstacle, p i (p i ) is larger, indicating that the complexity of the position has a smaller impact on the score, θ(p i ) represents the position p i The direction of the target object. The change of direction is reflected by the cosine function. The more drastic the change of direction, the greater the cosine function. i )) is smaller, indicating that the direction change of the position has less impact on the score, ||p i || indicates position p i The Euclidean norm of , the larger the norm, The smaller the position, the smaller the relative position influence score. i ) represents the position point p i The comprehensive environmental characteristic score, C(p j ) represents the position point p j The environmental feature score, α represents the adjustment position point p j The score of p i The weighted coefficient of the influence of the comprehensive score, which determines the influence of the scores of the surrounding points on the current point p i How much influence does α have on the score? If α is large, the influence of the surrounding points is greater. Otherwise, the influence of the surrounding points is smaller. j represents the index variable of all points in the traversal path. i Indicates the marked position point p i The index variable, Indicates the number of the best path with the highest comprehensive score, Indicates the number of the path that achieves the maximum value in all path sets. represents the number of one of the paths from the starting point to the destination, Y represents the number of the set of all paths, and p i Indicates the path A location point in , g represents the destination, H(p i ,g) indicates the position pi The estimated distance to the destination g, where o represents the total number of obstacles, represents the position p i The number of obstacles around it, d(p i , o) represents the position p i The distance to the obstacle o;

[0075] The derivation process of the expression for calculating the optimal path is as follows:

[0076] First, initialize an empty priority queue, add the starting point to the queue, and at the same time initialize an infinite comprehensive score for each position point p i and set the comprehensive score of the starting point to zero;

[0077] For each position point p i , calculate its comprehensive score:

[0078] F(p i ) = C(p i ) + α × ∑ j≠i C(p j );

[0079] Here, C(p i ) is the environmental feature score of the position point p i , which has been calculated through the previous formula. The comprehensive score F(p i ) includes the environmental feature score of the position point itself and the influence of the surrounding position points;

[0080] For each path, calculate its comprehensive score:

[0081]

[0082] Take out the position point p with the highest current score from the priority queue i , mark it as visited, and update the comprehensive scores of its neighboring position points. If the comprehensive score of a neighboring position point is smaller than before, update its comprehensive score and add it to the queue;

[0083] If the currently expanded position point is the end point, stop the search and return the current path with the maximum value as the optimal path. The expression is:

[0084]

[0085] If the currently expanded position point is not the end point, re-expand the path until the end point is found to be empty.

[0086] S3. Based on the optimal path, use deep learning technology to calculate the output prediction value of potential safety hazards on the optimal path, including the following steps,

[0087] Based on the best path with the highest comprehensive score, a decision tree model is constructed using deep learning technology, and at the same time, the decision tree model is deployed into the actual environment using an integrated development environment;

[0088] An integrated development environment is a software application for software development that integrates multiple development tools and functions in a single graphical user interface so that developers can efficiently write, debug, run, and manage software code;

[0089] The geographical location is input into the decision tree model, and the output prediction value of potential safety hazards on the best path is calculated. The expression is:

[0090]

[0091] where y represents the output prediction value of potential safety hazards, n represents the number of decision tree models, which is determined according to experiments, f represents the index of the decision tree, used to distinguish different decision trees, and is sequentially assigned according to the training order of the decision trees, w i represents the weight of the f-th decision tree, and this weight reflects the contribution degree of the f-th decision tree in the overall model prediction, g f (x) represents the prediction function of the f-th decision tree, which is used to accept the geographical location as input and output the potential safety hazard prediction value, obtained by training the decision tree model. For each decision tree, the training data is used to fit the model to obtain a function that accepts the input geographical location and outputs the prediction value.

[0092] For the index of each decision tree, the input geographical location information is used, and the prediction value of each decision tree is multiplied by its corresponding weight. The weighted prediction values of all decision trees are summed, and the final y obtained is the output prediction value of potential safety hazards after integrating the prediction results of all decision trees;

[0093] The mean and standard deviation of the output prediction value of potential safety hazards are calculated, and low and high thresholds are set based on the mean and standard deviation to divide the degree of potential safety hazards;

[0094] When the output prediction value of potential safety hazards is less than or equal to the low threshold, it is determined that the degree of potential safety hazards on the best path is low risk;

[0095] When the output prediction value of potential safety hazards is greater than the high threshold, it is determined that the degree of potential safety hazards on the best path is high risk;

[0096] When the output prediction value of potential safety hazards is greater than the low threshold and less than or equal to the high threshold, it is determined that the degree of potential safety hazards on the best path is medium risk.

[0097] S4. The identified safety hazard is notified to the operator, and the operator checks the operating status and provides countermeasures, thereby collecting response data of the shipping machinery, including the following steps:

[0098] Set the display information of low-risk safety hazards to green, medium-risk to yellow, and high-risk to red. At the same time, set three types of speaker voice prompts, namely, the current path is in a low-risk state, please note that the current path has a medium risk, please operate with caution, and warning that the current path has a high risk, please take immediate measures, and all display information will be prompted to the operator;

[0099] When the safety hazard level is low risk, listen to the sounds of the surrounding environment, use high-definition cameras to set cycles to scan environmental changes around the shipping machinery, use dashboards to monitor the environment, and save monitoring data in logs;

[0100] When the safety hazard level is medium risk, the loading machinery switches to manual mode and reduces the running speed. Magnetometers and vibration sensors are used for auxiliary monitoring. The sensor readings and monitoring data of image materials are recorded and reported to the relevant departments.

[0101] When the safety hazard level is high risk, stop the operation of all loading machinery and arrange for all personnel to evacuate to a safe area. At the same time, immediately contact team members to plan alternative action routes;

[0102] Collect response data of shipping machinery based on the response measures under different risks.

[0103] S5. Adjusting the operating state of the shipping machinery according to the response data of the shipping machinery comprises the following steps:

[0104] Collect response data including speed and distance of loading machinery following countermeasures;

[0105] If the response data does not deviate from the target speed and distance, operate the loading machinery normally and monitor the system status regularly;

[0106] If the response data deviates from the target speed and distance, stop the operation of the loading machinery immediately, find out the cause of the deviation, check the working status of the control unit and sensor of the machinery and repair the faulty parts. If there is electromagnetic interference, use shielded cables to isolate sensitive equipment to prevent electromagnetic interference from entering the control system, and wait for the machinery to resume normal operation before continuing the operation;

[0107] If only the speed shows a deviation, check whether there are faults in the motor system, such as overload, wear or electrical faults. Clean the motor surface with compressed air and a special cleaning agent and add lubricating oil to the bearings. At the same time, remind the operator to pay attention and adjust the speed of the shipping machinery to return it to the target speed;

[0108] If only the distance shows a deviation, check whether there are wear, looseness or other physical problems in the main structure of the shipping machinery. Tighten the loose screws, nuts and other fasteners, and adjust the belt tension, gear meshing clearance and other mechanical components. After the processing is completed, re-plan the route and notify the operator to drive according to the new route.

[0109] This embodiment also provides a safety intelligent auxiliary system for a shipping machinery, including: an environmental data collection module, which is responsible for collecting the environmental data around the started shipping machinery using a variety of sensors and performing preprocessing;

[0110] An environmental data fusion and path planning module, which is responsible for fusing the preprocessed environmental data using a deep learning model and calculating the best path from the current position to the destination through a search algorithm;

[0111] A potential safety hazard identification module, which is responsible for calculating the potential safety hazards on the best path based on the best path using deep learning technology;

[0112] A safety prompt and countermeasure module, which is responsible for prompting the calculated potential safety hazards to the operator. The operator checks the operating status and provides countermeasures;

[0113] A response data collection module, which is responsible for collecting the response data of the shipping machinery;

[0114] A dynamic adjustment module, which is responsible for adjusting the operating status of the shipping machinery according to the response data of the shipping machinery.

[0115] This embodiment also provides a computer device, which is applicable to the situation of the safety intelligent auxiliary method for a shipping machinery, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the safety intelligent auxiliary method for a shipping machinery as proposed in the above embodiment.

[0116] The computer device may be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads provided on the outer shell of the computer device. It can also be an external keyboard, touchpad, or mouse, etc.

[0117] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for realizing the safety intelligent assistance of the shipping machinery as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (abbreviated as SRAM), electrically erasable programmable read-only memory (abbreviated as EEPROM), erasable programmable read-only memory (abbreviated as EPROM), programmable read-only memory (abbreviated as PROM), read-only memory (abbreviated as ROM), magnetic memory, flash memory, magnetic disk, or optical disc.

[0118] In summary, through the use of multiple sensors to collect environmental data and perform preprocessing, the present invention realizes the comprehensive capture of environmental information and noise filtering, improving the quality of data and the accuracy of environmental perception. By using deep learning technology to build a decision tree model and calculate the output prediction value of potential safety hazards, the prediction of potential safety hazards on the path is realized, enabling the system to identify potential risk points in advance, prompt the calculated potential safety hazards to the operator, and collect the response data of the shipping machinery, improving the operator's ability to respond to potential safety hazards.

[0119] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for intelligent assistance of loading machinery safety, characterized in that: include, Use a variety of sensors to collect and pre-process the data of the surrounding environment of the loading and unloading machinery after it is started; The pre-processed environmental data is integrated using a deep learning model, and the optimal path from the current location to the destination is calculated using a search algorithm. Based on the optimal path, deep learning technology is used to calculate the safety hazards on the optimal path; The calculated safety hazards are notified to the operator, who then checks the operating status and provides countermeasures, thereby collecting response data of the loading machinery; The operating status of the loading machinery is adjusted according to the response data of the loading machinery.

2. The intelligent safety assistance method for loading machinery according to claim 1, characterized in that: The various sensors include high-definition cameras, radars, and GPS receivers; The environmental data includes distance, speed and specific geographical location; The data preprocessing includes removing noise in the environmental data through filtering technology, deleting abnormal values, and then converting the environmental data into a unified format.

3. The intelligent safety assistance method for loading machinery according to claim 2, characterized in that: The pre-processed environmental data is fused using a deep learning model, and the optimal path from the current location to the destination is calculated through a search algorithm, including the following steps: Select a convolutional neural network model; Extract the environmental features of the position and shape of obstacles and the distance, speed and direction of the target object from the preprocessed environmental data, and fuse the extracted environmental features in a weighted manner; Use the greedy best-first search algorithm to calculate the comprehensive score of each path from the current location to the destination and number them, and select the path with the highest comprehensive score as the best path. The expression is: F(p i )=C(p i )+α×∑ j≠i C(p j ); Among them, C(p i ) represents the position point p i The environmental characteristic score, Z represents the normalization constant, d(p i ) represents the position p i The distance to the target object, η represents the velocity attenuation coefficient, v(p i ) represents the position p i The speed of the target object, s(p i ) represents the position p i The size of the upper obstacle, θ(p i ) represents the position p i The direction of the target object, ||p i || indicates position p i The Euclidean norm of i ) represents the position point p i The comprehensive environmental characteristic score, C(p j ) represents the position point p j The environmental feature score, α represents the adjustment position point p j The score of p i The weighted coefficient of the influence of the comprehensive score, j represents the index variable of all the position points in the traversal path, and i represents the identification position point p i The index variable, Indicates the number of the best path with the highest comprehensive score, Indicates the number of the path that achieves the maximum value in all path sets. represents the number of one of the paths from the starting point to the destination, Y represents the number of the set of all paths, and p i Indicates the path A location point in , g represents the destination, H(p i ,g) indicates the position p i The estimated distance to the destination g, o represents the number of all obstacles, Indicates position p i The number of surrounding obstacles, d(p i ,o) indicates position p i The distance to the obstacle o.

4. The intelligent safety assistance method for loading machinery according to claim 3, characterized in that: Based on the optimal path, deep learning technology is used to calculate the output prediction value of safety hazards on the optimal path. The following steps are included: Based on the best path with the highest comprehensive score, a decision tree model is built using deep learning technology, and the decision tree model is deployed in the actual environment using an integrated development environment; The geographic location is input into the decision tree model to calculate the output prediction value of the safety hazard on the best path. The expression is: Among them, y represents the output prediction value of the safety hazard, n represents the number of decision tree models, f represents the index of the decision tree, and w f represents the weight of the f-th decision tree, g f (x) represents the prediction function of the f-th decision tree.

5. The intelligent safety assistance method for loading machinery according to claim 4, characterized in that: Setting thresholds to classify the degree of potential safety hazards includes the following steps: Calculate the mean and standard deviation of the output prediction value of the safety hazard, and set the low threshold and high threshold according to the mean and standard deviation to classify the degree of the safety hazard; When the output prediction value of the safety hazard is less than or equal to the low threshold, the safety hazard level on the optimal path is judged to be low risk; When the output prediction value of the safety hazard is greater than the high threshold, the safety hazard level on the optimal path is judged to be high risk; When the output prediction value of the safety hazard is greater than the low threshold and less than or equal to the high threshold, the degree of the safety hazard on the optimal path is judged to be medium risk.

6. The intelligent safety assistance method for loading machinery according to claim 5, characterized in that: The identified safety hazards are notified to the operator, who then checks the operating status and provides countermeasures, thereby collecting response data of the loading and unloading machinery. The following steps are included: Set the display information of low-risk safety hazards to green, medium-risk to yellow, and high-risk to red, and prompt the operator with the displayed information; When the potential safety hazard is low risk, listen to the surrounding sounds and use high-definition cameras to monitor the environment, while saving the monitoring data in the log; When the safety hazard level is medium risk, the loading machinery will switch to manual mode and reduce the operating speed, and report the medium risk situation to the relevant departments; When the safety hazard level is high risk, stop the operation of all loading machinery and arrange for all personnel to evacuate to a safe area. At the same time, immediately contact team members to plan alternative action routes; Collect response data of shipping machinery based on the response measures under different risks.

7. The intelligent safety assistance method for loading machinery according to claim 6, characterized in that: Adjusting the operating state of the shipping machinery according to the response data of the shipping machinery includes the following steps: Collect response data including speed and distance of loading machinery following countermeasures; If the response data has no deviation from the target speed and distance, the loading and unloading machinery is operated normally; If the response data deviates from the target speed and distance, stop the operation of the loading machinery immediately, check the working status of the control unit and sensors of the machinery, repair the faulty parts, and wait for the machinery to resume normal operation before continuing the operation; If only the speed is deviated, the operator is reminded to pay attention and adjust the speed of the loading machinery to return it to the target speed; If only the distance is off, re-plan the route and notify the operator to follow the new route.

8. A safe intelligent assistance system for shipping machinery, based on the safe intelligent assistance method for shipping machinery according to any one of claims 1 to 7, characterized in that: include, The environmental data collection module is responsible for using a variety of sensors to collect and pre-process the environmental data surrounding the loading machinery after it is started; The environmental data fusion and path planning module is responsible for fusing the pre-processed environmental data using a deep learning model and calculating the best path from the current location to the destination through a search algorithm; The safety hazard identification module is responsible for calculating the safety hazards on the optimal path based on the optimal path using deep learning technology; The safety prompt and countermeasure module is responsible for notifying the operator of the calculated safety hazards, and the operator checks the operating status and provides countermeasures; Response data collection module, responsible for collecting response data of shipping machinery; The dynamic adjustment module is responsible for adjusting the operating status of the loading machinery according to the response data of the loading machinery.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the loading machinery safety intelligent assistance method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the loading machinery safety intelligent assistance method according to any one of claims 1 to 7 are implemented.

Images