Container identifying and grabbing system based on visual identification

Abstract

The invention relates to the field of container identification and grabbing, and particularly discloses a container identification and grabbing system based on visual identification, which comprises an image acquisition module used for capturing a container image in a working environment in real time; the visual processing module is used for receiving the image transmitted by the image acquisition module, the visual processing module comprises a Yov5 algorithm and a U-net algorithm, the position of the lockhole of the container is quickly positioned through the Yov5 algorithm, and then the lockhole is finely segmented by using the U-net algorithm so as to determine the specific area and the center point of the lockhole; by integrating advanced technologies of image acquisition, visual processing, navigation positioning and the like, the system realizes automatic identification and grabbing of containers, so that the automation degree of container carrying is remarkably improved, manual intervention is reduced, and the working efficiency is improved; through cooperative use of the Yov5 algorithm and the U-net algorithm, the system can quickly and accurately position the position of the lockhole of the container and perform fine segmentation.

Description

Technical Field

[0001] The present invention belongs to the field of container identification and grasping, and specifically relates to a container identification and grasping system based on visual recognition. Background Art

[0002] Container identification technology is an important part of the field of logistics automation. With the rapid development of the logistics industry, the quantity and variety of containers are increasing continuously, which poses higher requirements for the identification and grasping of containers. Traditional container identification methods mainly rely on manual operation. The operator observes the position of the container with the naked eye and then manually operates the grasping device to perform the operation.

[0003] However, the traditional method of relying on a large number of workers for container identification and grasping operations not only has low efficiency but also high costs. In addition, there are certain safety risks in manual operation, such as equipment damage or personal injury caused by misoperation. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention provides a container identification and grasping system based on visual recognition to solve the problems of low efficiency and high costs in the prior art, where a large number of workers are relied on for container identification and grasping operations.

[0005] A container identification and grasping system based on visual recognition includes:

[0006] An image acquisition module for capturing container images in the working environment in real time;

[0007] A visual processing module for receiving the images transmitted by the image acquisition module. The visual processing module includes the Yolov5 algorithm and the U-net algorithm. The Yolov5 algorithm is used to quickly locate the position of the container keyhole, and then the U-net algorithm is used to perform fine segmentation on the keyhole to determine the specific area and center point of the keyhole.

[0008] A navigation and positioning module and a visual servo module for enabling the straddle carrier to move from the initial position to accurately grasp the container. The navigation and positioning module plans the initial driving path of the straddle carrier according to the keyhole position information provided by the visual processing module.

[0009] The visual servo module calculates the deviation between the real-time captured image and the keyhole image output by the visual processing module, and adjusts the motion parameters of the straddle carrier in real time according to these deviations until the accurate alignment and grasping of the container are achieved.

[0010] A remote control module for remotely monitoring and operating the straddle carrier to increase the flexibility and safety of the system.

[0011] The data analysis and monitoring module is used to monitor the operating status and working environment of the straddle carrier in real time, conduct data analysis and optimization, and improve the overall performance and reliability of the system.

[0012] Preferably, the image acquisition module adopts an OpenMV camera module, which can capture and process images in real time, providing high-quality image data to meet the requirements of the vision processing module for image quality.

[0013] Preferably, the vision processing module further includes an image preprocessing unit for preprocessing the acquired images, including denoising and enhancing contrast, to improve the image quality and subsequent recognition accuracy.

[0014] Preferably, the vision processing module also has the function of evaluating the vision recognition accuracy. By calculating the ratio of the number of keyholes detected by the system within the evaluation time T to the actual number of keyholes, and combining the accuracy of keyhole area segmentation, the vision recognition accuracy P is comprehensively obtained recog , ensuring the accuracy of system recognition;

[0015] Among them, the vision recognition accuracy evaluation formula is as follows:

[0016]

[0017] Among them, P recog is the average accuracy of the system for detecting and segmenting the container keyholes within time T;

[0018] C det is the number of keyholes detected by the system within time t;

[0019] C true is the actual number of existing keyholes;

[0020] A seg，i is the segmentation area of the i-th keyhole;

[0021] A true，i is the actual area of the i-th keyhole;

[0022] N is the total number of detected keyholes;

[0023] T is the evaluation time;

[0024] Value range: P recog ∈[0,1], the closer the value is to 1, the higher the vision recognition accuracy.

[0025] Preferably, the navigation and positioning module adopts two-dimensional code vision navigation technology or GPS / INS integrated navigation technology. By identifying two-dimensional code tags in the environment or using satellite signals and inertial sensor data, it provides accurate positioning and navigation information for the straddle carrier.

[0026] Preferably, the navigation and positioning module also has the function of evaluating the navigation and positioning error. By calculating the average value E of the sum of the squares of the Euclidean distances between the estimated position and the actual position of the straddle carrier nav , the accuracy of the straddle carrier's navigation and positioning is ensured;

[0027] Among them, the navigation and positioning error evaluation formula is as follows:

[0028]

[0029] Among them, E nav is the average error in M times of navigation and positioning of the system;

[0030] x est,j and y est,j are respectively the estimated position coordinates of the straddle carrier in the jth navigation and positioning;

[0031] x true,j and y true,j are respectively the true position coordinates of the straddle carrier in the jth navigation and positioning;

[0032] Value range: E nav ≥0, the smaller the value, the higher the navigation and positioning accuracy.

[0033] Preferably, the remote control module adopts a LoRa wireless transmission module to ensure stable data transmission, and realizes remote control and real-time monitoring through a user-friendly interface.

[0034] Preferably, the data analysis and monitoring module establishes a big data analysis platform, which can real-time monitor the operation status and working environment of the straddle carrier, conduct data analysis and optimization, including analyzing the change trends of visual recognition accuracy and navigation and positioning errors, so as to improve the overall performance and reliability of the system.

[0035] Preferably, it further includes safety monitoring devices, including cameras and sensors, which are used to real-time monitor potential risks in the working environment, including personnel intrusion and equipment failures, to ensure the safety of operations.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] By integrating advanced technologies such as image acquisition, visual processing, and navigation and positioning, this system realizes the automatic recognition and grasping of container keyholes, significantly improves the automation degree of container handling, reduces manual intervention, and improves the operation efficiency;

[0038] By using the Yolov5 algorithm in conjunction with the U-net algorithm, the system can quickly and accurately locate the position of the container keyhole, perform fine segmentation, and ensure the precise determination of the specific area and center point of the keyhole. This high-precision recognition ability provides a strong guarantee for the precise control and grasping of the straddle carrier. Brief Description of the Drawings

[0039] Figure 1 It is a schematic diagram of the system of the present invention. Detailed Embodiments

[0040] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0041] As Figure 1 shown:

[0042] Embodiment 1: The present invention provides a container recognition and grasping system based on visual recognition, including:

[0043] An image acquisition module for capturing container images in the working environment in real time;

[0044] A visual processing module for receiving the images transmitted by the image acquisition module. The visual processing module includes the Yolov5 algorithm and the U-net algorithm. The Yolov5 algorithm is used to quickly locate the position of the container keyhole, and then the U-net algorithm is used to perform fine segmentation on the keyhole to determine the specific area and center point of the keyhole;

[0045] Among them, the specific application process of the Yolov5 algorithm is as follows:

[0046] Image reception: The Yolov5 algorithm first receives the container image data from the image acquisition module.

[0047] Quick positioning: Using the efficient object detection ability of the Yolov5 algorithm, the container image is quickly scanned to locate the approximate position of the container keyhole. The algorithm divides the input image into multiple grids, each grid is responsible for detecting the keyhole therein, and predicts multiple bounding boxes, each bounding box contains a keyhole. For each bounding box, the algorithm will predict the keyhole it contains and the confidence level.

[0048] Feature Extraction and Fusion: The Yolov5 algorithm extracts image features through a backbone network (such as CSPDarknet53) to generate multiple feature maps. These feature maps are then processed through multi-scale feature fusion techniques to enhance the model's detection ability for targets of different sizes.

[0049] Bounding Box Prediction and Optimization: The algorithm predicts the bounding box coordinates and confidence levels of the targets in each grid cell, and removes the bounding boxes with large overlaps through non-maximum suppression (NMS), retaining the bounding box with the highest confidence level. This step further refines the position information of the keyhole.

[0050] Result Output: Finally, the Yolov5 algorithm outputs the specific position information of the keyhole for use by the navigation and positioning module.

[0051] Among them, the specific application process of the U-net algorithm

[0052] Fine Segmentation Preparation: After the Yolov5 algorithm locates the approximate position of the keyhole, the U-net algorithm receives the image data at these positions (possibly cropped image patches) for fine segmentation.

[0053] Encoder-Decoder Processing: The U-net algorithm adopts an encoder-decoder structure. The encoder part extracts image features through convolutional layers and pooling layers to form low-resolution feature maps. The decoder part then restores the feature maps to the original image size through transposed convolutional or upsampling layers, and gradually restores the detailed information of the image during this process.

[0054] Skip Connections: To enhance the feature representation ability of the decoder part, the U-net algorithm establishes skip connections between the encoder and the decoder. These connections directly transfer the feature maps from the encoder part to the decoder part, helping to retain more image details and edge information.

[0055] Fine Segmentation: After encoder-decoder processing and skip connections, the U-net algorithm outputs a segmentation mask with the same size as the input image. This mask precisely represents the area and shape of the keyhole.

[0056] Center Point Determination: By further processing the segmentation mask (such as calculating the centroid, etc.), the center point position of the keyhole can be determined.

[0057] The navigation and positioning module and the visual servo module are used to enable the straddle carrier to move from the initial position to precisely grasp the container; the navigation and positioning module plans the initial driving path of the straddle carrier according to the keyhole position information provided by the visual processing module;

[0058] The visual servo module calculates the deviation by using the real-time captured image and the keyhole image output by the visual processing module, and adjusts the motion parameters of the straddle carrier in real time according to these deviations until the precise alignment and grasping of the container are achieved;

[0059] The remote control module is used to remotely monitor and operate the straddle carrier, increasing the flexibility and safety of the system;

[0060] The data analysis and monitoring module is used to monitor the operating status and working environment of the straddle carrier in real time, perform data analysis and optimization, and improve the overall performance and reliability of the system.

[0061] As can be seen from the above, this system captures the container image in real time through the image acquisition module. The visual processing module uses the Yolov5 algorithm to quickly locate the keyhole position and combines the U-net algorithm for fine segmentation to determine the specific area and center point of the keyhole. The navigation and positioning module plans the driving path of the straddle carrier based on this information, while the visual servo module precisely adjusts the motion parameters of the straddle carrier through real-time image deviation calculation to achieve the precise alignment and grasping of the container. The remote control module increases the flexibility and safety of the system, and the data analysis and monitoring module monitors the status of the straddle carrier and the working environment in real time to optimize the system performance. The entire system is highly automated and intelligent, significantly improving the efficiency and accuracy of container handling, and providing strong support for the intelligent development of the logistics industry.

[0062] Embodiment 2: This embodiment is basically the same as the previous embodiment, except that the image acquisition module uses an OpenMV camera module, which can capture and process images in real time, providing high-quality image data to meet the requirements of the visual processing module for image quality.

[0063] Specifically, the visual processing module further includes an image preprocessing unit for preprocessing the collected images, including denoising and enhancing contrast, to improve the image quality and subsequent recognition accuracy.

[0064] Specifically, the visual processing module also has the function of evaluating the visual recognition accuracy. By calculating the ratio of the number of keyholes detected by the system within the evaluation time T to the actual number of keyholes, and combining the accuracy of keyhole area segmentation, the visual recognition accuracy P is comprehensively obtained recog , ensuring the accuracy of system recognition;

[0065] Among them, the visual recognition accuracy evaluation formula is as follows:

[0066]

[0067] Among them, P recog is the average accuracy of the system in detecting and segmenting the container keyhole within time T;

[0068] C det is the number of keyholes detected by the system within time t;

[0069] C true is the actual number of keyholes;

[0070] A seg，i is the segmentation area of the i-th keyhole;

[0071] A true，i is the actual area of the i-th keyhole;

[0072] N is the total number of detected keyholes;

[0073] T is the evaluation time;

[0074] Value range: P recog ∈[0, 1], the closer the value is to 1, the higher the visual recognition accuracy;

[0075] Among them, the specific application process of the above formula is as follows:

[0076] Data collection:

[0077] Collect the number of keyholes (C det ) detected by the system within the evaluation time T and the time stamp of each detection.

[0078] Record the actual number of keyholes (C true ), which is usually obtained through manual annotation or known data sets.

[0079] For each detected keyhole, obtain its segmentation area (A seg，i ) and record the corresponding actual keyhole area (A true，i ).

[0080] Calculate the integral part:

[0081] According to the integral part in the formula, calculate the average accuracy of the system's keyhole detection and segmentation within time T. This includes two parts: one is the proportion of the time when the system continuously and correctly detects keyholes within time T (calculated by subtracting the proportion of the time of false detection from 1); the other is the accuracy of keyhole area segmentation (calculated by comparing the difference between the segmentation area and the actual area).

[0082] Summation part:

[0083] For all detected keyholes, calculate the sum of the squares of the area segmentation errors and divide it by the total number of keyholes N to obtain the average error of area segmentation.

[0084] Comprehensive calculation:

[0085] Substitute the results of the integral part and the summation part into the formula to calculate the visual recognition accuracy (P recog ).

[0086] Processing flow for non - compliance:

[0087] Result analysis:

[0088] Check whether the P recog value falls within the range of [0, 1]. If it is not within this range, it indicates that there is an error in the calculation process.

[0089] Analyze the calculation results of the integral part and the summation part to determine which part causes the P recog value not to meet the requirements.

[0090] Error location:

[0091] If the integral part does not meet the requirements, check the division of time T, the accuracy of the number of keyhole detections, and the recording of timestamps.

[0092] If the summation part does not meet the requirements, check the accuracy of the keyhole area segmentation, including the performance and parameter settings of the segmentation algorithm.

[0093] Corrective measures:

[0094] According to the results of the error location, adjust the data collection method, optimize the segmentation algorithm, or adjust the algorithm parameters.

[0095] Recalculate until the P recog value meets the formula requirements.

[0096] Specifically, the navigation and positioning module adopts two - dimensional code visual navigation technology or GPS / INS integrated navigation technology, and provides accurate positioning and navigation information for the straddle carrier by identifying two - dimensional code tags in the environment or using satellite signals and inertial sensor data.

[0097] Specifically, the navigation and positioning module also has the function of evaluating navigation and positioning errors. By calculating the average value E of the sum of the squares of the Euclidean distances between the estimated position and the actual position of the straddle carrier nav , ensure the accuracy of the straddle carrier's navigation and positioning;

[0098] Among them, the navigation and positioning error evaluation formula is as follows:

[0099]

[0100] Among them, E nav is the average error of the system in M navigation and positioning;

[0101] x est,j and y est,jThey are the estimated position coordinates of the straddle carrier in the j-th navigation and positioning respectively;

[0102] x true,j and y true,j They are the true position coordinates of the straddle carrier in the j-th navigation and positioning respectively;

[0103] Value range: E nav ≥0. The smaller the value, the higher the navigation and positioning accuracy;

[0104] Among them, the specific application process of the above formula is as follows:

[0105] Data collection:

[0106] During the M times of navigation and positioning, record the estimated position coordinates (x est,j and y est,j ) of the straddle carrier at each navigation and positioning.

[0107] Obtain the true position coordinates (x true,j and y true,j ) of the straddle carrier at each navigation and positioning through a high-precision positioning device or a known data set.

[0108] Calculate the error:

[0109] According to the formula, calculate the error in each navigation and positioning, that is, the square of the Euclidean distance between the estimated position coordinates and the true position coordinates.

[0110] Calculate the average error:

[0111] Sum up the errors in the M times of navigation and positioning, and then divide by M to obtain the average error E nav .

[0112] Processing flow for non-compliance:

[0113] Result analysis:

[0114] Check whether the value of E nav is greater than or equal to 0. If it is less than 0, it means there is an error in the calculation process.

[0115] Analyze the error of each navigation and positioning to determine whether there are outliers or systematic errors.

[0116] Error location:

[0117] If there are outliers, check whether there are errors or interference factors in the data collection process.

[0118] If there are systematic errors, check the performance, parameter settings or environmental factors of the navigation and positioning system.

[0119] Corrective measures:

[0120] According to the result of error localization, adjust the data collection method, optimize the navigation and positioning system, or adjust the system parameters.

[0121] Re - perform the calculation until the nav E value meets the requirements of the formula.

[0122] As can be seen from the above, the image acquisition module of this embodiment adopts an OpenMV camera module, which can capture and process high - quality images in real time, meeting the high - standard requirements of the vision processing module for images; the vision processing module not only includes Yolov5 and U - net algorithms, but also adds an image pre - processing unit to improve image quality and recognition accuracy, and introduces a vision recognition accuracy evaluation function, which comprehensively evaluates by calculating the ratio of the number of detected keyholes to the actual number and the accuracy of keyhole area segmentation to ensure the accuracy of system recognition; in addition, the navigation and positioning module adopts two - dimensional code vision navigation or GPS / INS combined navigation technology to provide accurate positioning and navigation information, and has a function of evaluating navigation and positioning errors, ensuring the accuracy of navigation and positioning by calculating the average value of the sum of squares of the Euclidean distances between the estimated position and the actual position; these improvement measures together improve the overall performance and reliability of the system, making it more suitable for complex container handling operation environments.

[0123] Embodiment 3: This embodiment is basically the same as the previous one, except that the remote control module adopts a LoRa wireless transmission module to ensure stable data transmission, and realizes remote control and real - time monitoring through a user - friendly interface.

[0124] Specifically, the data analysis and monitoring module establishes a big data analysis platform, which can monitor the running state and operation environment of the straddle carrier in real time, conduct data analysis and optimization, including analyzing the change trends of vision recognition accuracy and navigation and positioning errors, so as to improve the overall performance and reliability of the system.

[0125] Specifically, it also includes safety monitoring equipment, including cameras and sensors, which are used to monitor potential risks in the operation environment in real time, including personnel intrusion and equipment failures, to ensure the safety of operations.

[0126] As can be seen from the above, in this embodiment, the remote control module adopts a LoRa wireless transmission module, ensuring the stability of data transmission, and realizing remote control and real-time monitoring through a user-friendly interface, improving the flexibility and operation convenience of the system; the data analysis and monitoring module establishes a big data analysis platform, which can monitor the running status and working environment of the straddle carrier in real time, deeply analyze the changing trends of visual recognition accuracy and navigation positioning error, and provide strong support for the optimization of system performance; in addition, the system is also equipped with safety monitoring devices, including cameras and sensors, to monitor potential risks in the working environment in real time, such as personnel intrusion and equipment failures, effectively ensuring the safety of operations; these improvement measures jointly improve the overall performance and reliability of the system, making it more adaptable to complex and changeable working environments.

[0127] The standard parts used in the present invention can all be purchased from the market. The special-shaped parts can be customized according to the description in the specification and the drawings. The specific connection methods of each part all adopt conventional means such as bolts, rivets, and welding that are mature in the prior art. The machines, parts, and equipment all adopt conventional models in the prior art. In addition, the circuit connection adopts a conventional connection method in the prior art, which will not be elaborated here. The content not described in detail in this specification belongs to the prior art well-known to those skilled in the art.

[0128] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. The meaning of "a plurality" is two or more, unless otherwise specifically defined.

[0129] In the present invention, unless otherwise clearly defined and limited, the terms "installed", "connected", "connected", "fixed", etc. should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0130] In the present invention, unless otherwise clearly specified or limited, a first feature being "on" or "under" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "over" and "on top of" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the horizontal height of the first feature is higher than that of the second feature. A first feature being "under", "below" and "beneath" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the horizontal height of the first feature is lower than that of the second feature.

[0131] In the description of this specification, the description with reference to terms such as "an embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not have to refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0132] In the drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments are involved, and other structures can refer to the general design. Without conflict, the same embodiment and different embodiments of the present invention can be combined with each other.

[0133] Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A container recognition and grasping system based on visual recognition, characterized in that Including: An image acquisition module for capturing container images in the working environment in real time; A vision processing module for receiving the images transmitted by the image acquisition module. The vision processing module includes the Yolov5 algorithm and the U-net algorithm. The position of the container keyhole is quickly located through the Yolov5 algorithm, and then the U-net algorithm is used to finely segment the keyhole to determine the specific area and center point of the keyhole; A navigation and positioning module and a vision servo module for enabling the straddle carrier to move from the initial position to accurately grasp the container; the navigation and positioning module plans the initial driving path of the straddle carrier according to the keyhole position information provided by the vision processing module; The vision servo module calculates the deviation by using the real-time captured image and the keyhole image output by the vision processing module, and adjusts the motion parameters of the straddle carrier in real time according to these deviations until the accurate alignment and grasping of the container are achieved; A remote control module for realizing remote monitoring and operation of the straddle carrier, increasing the flexibility and safety of the system; A data analysis and monitoring module for monitoring the operating state of the straddle carrier and the working environment in real time, performing data analysis and optimization to improve the overall performance and reliability of the system.

2. The container recognition and grasping system based on visual recognition according to claim 1, characterized in that The image acquisition module uses an OpenMV camera module, which can capture and process images in real time, providing high-quality image data to meet the requirements of the vision processing module for image quality.

3. The container recognition and grasping system based on visual recognition according to claim 2, wherein, The vision processing module further includes an image preprocessing unit for preprocessing the acquired images, including denoising and enhancing the contrast to improve the image quality and subsequent recognition accuracy.

4. The container recognition and grasping system based on visual recognition according to claim 3, characterized in that, The visual processing module also has the function of evaluating the visual recognition accuracy. By calculating the ratio of the number of keyholes detected by the system within the evaluation time T to the actual number of keyholes, and combining the accuracy of keyhole area segmentation, the visual recognition accuracy P is comprehensively obtained recog , ensuring the accuracy of system recognition; Among them, the vision recognition accuracy evaluation formula is as follows: Among them, P recog is the average accuracy of the system for detecting and segmenting the container keyhole within time T; C det is the number of keyholes detected by the system within time t; C true is the number of actually existing keyholes; A seg，i is the divided area of the i-th keyhole; A true，i is the actual area of the i-th keyhole; N is the total number of detected keyholes; T is the evaluation time; Value range: P recog ∈[0, 1], the closer the value is to 1, the higher the visual recognition accuracy.

5. The container recognition and grasping system based on visual recognition according to claim 4, characterized in that The navigation and positioning module adopts two-dimensional code vision navigation technology or GPS / INS integrated navigation technology, and provides accurate positioning and navigation information for the straddle carrier by identifying two-dimensional code tags in the environment or using satellite signals and inertial sensor data.

6. The container recognition and grasping system based on visual recognition according to claim 5, wherein The navigation and positioning module also has the function of evaluating the navigation and positioning error, by calculating the average value E of the sum of the squares of the Euclidean distances between the estimated position and the actual position of the straddle carrier nav , to ensure the accuracy of the straddle carrier's navigation and positioning; Among them, the navigation and positioning error evaluation formula is as follows: Among them, E nav is the average error of the system in M navigation and positioning times; x est,j and y est,j are respectively the estimated position coordinates of the straddle carrier in the j-th navigation and positioning; x true,j and y true,j are respectively the true position coordinates of the straddle carrier in the j-th navigation and positioning; Value range: E nav ≥0, the smaller the value, the higher the navigation and positioning accuracy.

7. The container recognition and grasping system based on visual recognition according to claim 1, characterized in that The remote control module adopts a LoRa wireless transmission module to ensure stable data transmission, and realizes remote control and real-time monitoring through a user-friendly interface.

8. The container recognition and grasping system based on visual recognition according to claim 1, characterized in that, The data analysis and monitoring module establishes a big data analysis platform, which can monitor the operating state of the straddle carrier and the working environment in real time, perform data analysis and optimization, including analyzing the change trends of vision recognition accuracy and navigation and positioning error, to improve the overall performance and reliability of the system.

9. The container recognition and grasping system based on visual recognition according to claim 1, characterized in that It also includes safety monitoring devices, including cameras and sensors, for monitoring potential risks in the working environment in real time, including personnel intrusion and equipment failures, to ensure the safety of operations.

Images