A smart stacking device, stacker crane control method, control system, and medium
Through intelligent stacking devices and control systems, image recognition and sensor modules are used to monitor cargo status, predict stacker operation risks, and adjust shock absorption parameters, solving the problems of stacker shaking and cargo overturning, and improving stacking efficiency and safety sex.
Patent Information
- Application Number
- CN202311703601.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-12
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2043-12-12
AI Technical Summary
The existing stacker has not been effectively solved due to acceleration and deceleration during the working process, and it is impossible to monitor the tilt and frequent shaking of the goods to be stacked in time, resulting in economic losses and safety hazards caused by overturning the goods, and has a single function. The size and defects of the goods cannot be sensed.
Using intelligent stacking devices and control systems, including image recognition modules, sensor modules, adjustable shock absorbers and control systems, to predict stacker operation risks by obtaining cargo data and operating data, and adjust adjustable shock absorbers according to risks. shock absorbing parameters to reduce cargo overturning risks and improve stacking efficiency.
Effectively reduce the risk of cargo overturning, improve stacking efficiency, ensure the safety of goods and staff, timely sense and handle cargo status abnormalities, and improve the stability and safety of cargo stacking.
Smart Images

Figure CN117819104B_ABST
Abstract
Description
Technical Field
[0001] The present specification relates to the field of computer information processing, and in particular to an intelligent stacking device and a stacker control method, a control system and a medium. Background Art
[0002] With the gradual development of society, the use of stackers to replace manual stacking is a major trend in the current warehousing system. In order to solve the shaking problem caused by acceleration and deceleration of the stacker during operation, the existing technology usually adopts the method of adding reverse torque during movement to prevent the stacker from swinging; using high-rigidity components to increase the stability and rigidity of the stacker. However, the impact of the goods to be stacked on the shaking is not considered. When the goods to be stacked are tilted and shake frequently, the existing stackers cannot monitor the movement status of the goods in time. Once the goods overturn, it will bring a lot of economic losses and threaten the staff on site. In addition, the existing stackers often have a single function and cannot perceive the size, appearance, deformation and other defects of the goods to be stacked. There is a lack of monitoring and adjustment of the changes in the state of the goods during the stacking process. When the state of the goods is abnormal, it is often not perceived and handled in time.
[0003] Therefore, it is necessary to propose an intelligent stacking device and control system to reduce the risk of cargo overturning, improve stacking efficiency, and ensure the safety of cargo and on-site personnel. Summary of the invention
[0004] One or more embodiments of the present specification provide a stacker control method. The stacker control method includes: acquiring first cargo data, second cargo data, and operation data; predicting the operation risk of the stacker based on the first cargo data, the second cargo data, and the operation data; and determining an adjustment instruction based on the operation risk, the adjustment instruction including an adjustment amount of a shock absorbing parameter of an adjustable shock absorber of the stacker.
[0005] One or more embodiments of the present specification provide a stacker control system. The stacker control system includes: a stacker, an image recognition module, a sensor module, an adjustable shock absorber and a control system, wherein the image recognition module, the sensor module and the adjustable shock absorber are mechanically connected to the stacker and are communicatively connected to the control system; the image recognition module is configured to identify the first cargo data on the stacker and the second cargo data on the shelf; the sensor module is configured to obtain the operation data of the stacker; the adjustable shock absorber is deployed at the bottom of the pallet of the stacker and is configured to dampen the stacker; the control system is configured to: predict the operation risk of the stacker based on the first cargo data, the second cargo data and the operation data; determine an adjustment instruction based on the operation risk, and the adjustment instruction includes the adjustment amount of the damping parameter of the adjustable shock absorber.
[0006] One or more embodiments of the present specification provide an intelligent stacking device, including a processor, wherein the processor is used to execute the stacker control method described in the above embodiments.
[0007] One or more embodiments of the present specification provide a computer-readable storage medium, wherein the storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the stacker control method described in the above embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:
[0009] Figure 1 is an exemplary schematic diagram of an intelligent stacking device according to some embodiments of this specification;
[0010] Figure 2 is an exemplary flow chart of an intelligent stacking method according to some embodiments of this specification;
[0011] Figure 3 is an exemplary schematic diagram of operation risk prediction according to some embodiments of this specification;
[0012] Figure 4 It is an exemplary schematic diagram of determining the adjustment instruction process according to some embodiments of this specification. DETAILED DESCRIPTION
[0013] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.
[0014] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.
[0015] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.
[0016] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.
[0017] Figure 1 is an exemplary schematic diagram of an intelligent stacking device according to some embodiments of the present specification.
[0018] In some embodiments, the intelligent stacking device 100 includes a stacker 110, an adjustable shock absorber 120, a control system 130, an image recognition module 140, and a sensor module 150. The image recognition module, the sensor module, and the adjustable shock absorber are mechanically connected to the stacker and are communicatively connected to the control system.
[0019] The stacker 110 refers to a lifting and transporting device used for stacking, transporting, and storing goods in a storage system. The stacker 110 can receive instructions from the control system 130 to transport target goods to a target shelf, or take out goods from a target shelf and transport them to a designated location.
[0020] The adjustable shock absorber 120 is configured as a functional module for shock absorbing the stacker 110. In some embodiments, the adjustable shock absorber 120 may include a hydraulic shock absorber, an air spring, and the like.
[0021] In some embodiments, one or more adjustable shock absorbers 120 may be installed under the pallet of the stacker 110 and mechanically connected to the stacker. The mechanical connection includes but is not limited to bolt connection, riveting connection, latch connection, welding connection, hinge connection, etc. In some embodiments, the adjustable shock absorber 120 may also be connected to the control system 130 through hardware connection and / or software communication.
[0022] In some embodiments, the adjustable shock absorber 120 can adjust the shock absorption coefficient based on the adjustment instruction generated by the control system 130 to achieve different shock absorption levels. For more information about the adjustment instruction, shock absorption coefficient and shock absorption level, please refer to Figure 2 and its related description.
[0023] The control system 130 is configured to be a system for coordinating and coordinating the connection and cooperation between the various functional modules, and to gather all the information of the stacking device to provide control and management functions for the stacker.
[0024] In some embodiments, the control system 130 may receive cargo-related data collected by the image recognition module 140 and sensor data from the sensor module 150. The control system 130 may also send an adjustment instruction including an adjustment amount of a damping parameter of the adjustable damper to the adjustable damper 120.
[0025] In some embodiments, the control system 130 can evaluate the risks in the stacker operation process based on the cargo data and the operation data and generate adjustment instructions. For more information about cargo data, operation data and adjustment instructions, please refer to Figure 2 and its related description.
[0026] The image recognition module 140 is configured as a functional module for capturing images and performing image recognition, and the image recognition module may include a camera, a rangefinder, etc. In some embodiments, the image recognition modules may be installed on the stacker 110 and the shelf, respectively, to obtain image data on the stacker and the shelf, respectively.
[0027] In some embodiments, the image recognition module 140 can perform image recognition on the goods in the captured image to obtain relevant data of the goods in the image. For more information about image recognition, please refer to Figure 2 and its related description.
[0028] In some embodiments, before image recognition, the image recognition module may pre-process the captured image, for example, the pre-processing may include adjusting the size or resolution.
[0029] In some embodiments, the image recognition module 140 may be connected to the control system 130 through hardware connection and / or software communication, and send the data related to the goods in the recognized image to the control system 130 .
[0030] The sensor module 150 is configured as a functional module for detecting the operating state of the stacker and collecting corresponding operating data. In some embodiments, the sensor modules can be installed on the driving wheels and other power devices of the stacker 110 to ensure the accuracy of the collected data.
[0031] In some embodiments, the sensor module 150 may include multiple sensors. For example, the sensor module may include one or more of a speed sensor, an acceleration sensor, an angular velocity sensor, a position sensor, an inclination sensor, a vibration sensor, etc. Accordingly, the sensor data may include one or more of the speed data, acceleration data, angular velocity data, position data, inclination data, vibration data, etc. of the stacker.
[0032] In some embodiments, the sensor module 150 may be communicatively connected to the control system 130 via hardware connection and / or software communication, and transmit the sensor data collected from the stacker 110 to the control system 130 .
[0033] For more information about the above modules and systems, please refer to the following Figures 2 to 4 Related description.
[0034] Figure 2 is an exemplary flow chart of the intelligent stacking method according to some embodiments of this specification. Figure 2 As shown, the process 200 includes the following steps. In some embodiments, the process 200 may be executed by the control system 130 .
[0035] Step 210, obtaining first cargo data, second cargo data and operation data.
[0036] The first cargo data refers to characteristic data of the cargo on the stacker. The first cargo data may include at least one indicator, such as cargo quantity, positioning data, offset size, deformation data, appearance data, etc. When the cargo state on the stacker changes (for example, cargo is added or reduced), the first cargo data also changes accordingly.
[0037] The quantity of goods refers to the number of pieces of goods stacked on the stacker; the positioning data refers to the specific location information of the goods on the stacker. For example, the geometric center position of the goods can be determined as the positioning data, and the center of gravity position of the goods can also be determined as the positioning data. The positioning data can be represented by a two-dimensional or three-dimensional coordinate system.
[0038] The offset dimension refers to the difference between the preset position of the goods and the actual position of the goods. The offset dimension can be expressed by distance difference or coordinate difference. For example, the offset dimension can be expressed by two-dimensional or three-dimensional coordinate difference. The preset position of the goods refers to the preset position of the goods on the stacker, which can be preset based on prior experience; deformation data refers to the difference between the external contour shape of the goods and the original contour shape of the goods after the handling and transportation process. The deformation data may include the degree of deformation, distortion, damage, etc. of the goods; appearance data refers to data describing the appearance of the goods, such as appearance size, shape, color, etc.
[0039] When there are multiple goods on the stacker at the same time, the first goods data can represent a single goods or a combination of multiple goods. For example, the positioning data in the first goods data can describe the center of gravity position of a single goods or the center of gravity position of a combination of multiple goods on the stacker.
[0040] The second cargo data refers to the characteristic data of the cargo located on the shelf. When the status of the cargo on the shelf changes (for example, the cargo is increased or decreased), the second cargo data also changes accordingly. The indicators included in the second cargo data are the same as those in the first cargo data, and the relevant description of the first cargo data can be referred to.
[0041] Operation data refers to the motion data of the stacker itself. In some embodiments, the operation data may include one or more of the stacker speed, acceleration, angular velocity, position, inclination, swing frequency and amplitude, etc.
[0042] In some embodiments, the control system 130 may obtain first cargo data through an image recognition module 140 fixed on the stacker 110 , obtain second cargo data through an image recognition module 140 fixed on the shelf, and obtain operation data through a sensor module 150 fixed on the stacker 110 .
[0043] In some embodiments, the corresponding image recognition module 140 can perform image recognition in a variety of ways. For example, the image recognition module 140 can take images of goods on a stacker or a shelf at a certain frequency, and perform image recognition based on a neural network model to obtain the first goods data or the second goods data. The parameters of the neural network model can be obtained through training.
[0044] The training sample may include several groups of training data, each of which may be obtained based on the cargo data of historical cargo. For example, the captured images of historical cargo are input into the initial neural network model as training samples, and a loss function is constructed based on the output and label of the initial neural network model. The parameters of the initial neural network model are iteratively updated according to the loss function until the preset conditions are met, the training ends, and the trained neural network model is obtained. The preset conditions may include but are not limited to the convergence of the loss function and the reaching of the training cycle. The labels of the training samples may be the cargo quantity, positioning data, offset size, deformation data, and appearance data of the cargo in the manually annotated image.
[0045] Step 220 , predicting the operation risk of the stacker based on the first cargo data, the second cargo data and the operation data.
[0046] Operational risk refers to the risk of cargo tipping over and being deformed and damaged during the handling and storage and retrieval process of the stacker. In some embodiments, the operational risk may be positively correlated to the deformation data of the cargo, the quantity of cargo, the speed and acceleration of the stacker, the swing amplitude and frequency, etc. The operational risk may be expressed in levels or values.
[0047] The control system 130 can determine the operation risk in a variety of ways. For example, the control system 130 can construct a first feature vector based on the first cargo data, the second cargo data, and the operation data. The control system 130 can search in the first vector database based on the first feature vector to obtain a first reference vector whose vector distance with the first feature vector satisfies the distance condition. The operation risk corresponding to the first reference vector is determined as the current operation risk. Among them, the vector database is used to store a number of historical vectors and their corresponding operation risks. The historical vector is constructed based on the historical first cargo data, the historical second cargo data, and the historical operation data. The distance condition can be the minimum vector distance.
[0048] In some embodiments, the control system 130 may also predict the first risk through the first risk prediction model based on the first cargo data and the operation data; and predict the second risk through the second risk prediction model based on the second cargo data and the cargo to be piled data. For the prediction process of the first risk and the second risk, please refer to Figure 3 and its related description.
[0049] Step 230 : determining an adjustment instruction based on the operation risk, wherein the adjustment instruction includes an adjustment amount of a damping parameter of an adjustable damper of the stacker.
[0050] The adjustment instruction refers to an instruction to control and adjust the stacker operation parameters and the damping parameters of the adjustable shock absorber. For example, the adjustment instruction to the stacker may be to reduce the speed of the stacker, reduce the amount of goods on the stacker, adjust the inclination angle of the stacker to keep it level, etc. The control system 130 may send an adjustment instruction to the stacker 110 and the adjustable shock absorber 120 to adjust the working state of the stacker and the adjustable shock absorber.
[0051] The damping parameter of the adjustable shock absorber refers to the parameter that controls the damping degree of the shock absorber. In some embodiments, the damping parameter may include adjusting the damping force, elastic coefficient, etc. of the adjustable shock absorber. The damping parameter may be represented by a numerical value. For example, the larger the numerical value, the stronger the damping capacity of the current adjustable shock absorber is, and the lower the possibility of the stacker shaking is.
[0052] The adjustment amount refers to the change range of the shock absorbing parameter. The larger the adjustment amount, the larger the change range of the shock absorbing parameter. In some embodiments, the adjustment amount may include the adjustment amount of the damping force of the adjustable shock absorber and the elastic coefficient of the adjustable shock absorber. The adjustment amount may be represented by a numerical value. When the numerical value is positive, the corresponding shock absorbing parameter is increased, and when it is negative, the corresponding shock absorbing parameter is decreased.
[0053] In some embodiments, the control system 130 may determine the adjustment instruction in a variety of ways. For example, the processor may determine at least one corresponding adjustment instruction by querying the first preset table based on the current operation risk.
[0054] In some embodiments, the first preset table may include multiple groups of historical operation risks and corresponding adjustment instructions. In some embodiments, the first preset table may be preset based on historical data or prior knowledge.
[0055] By judging the relevant information of the goods themselves and the operating status of the stacker to judge the operational risks during transportation, the accuracy and credibility of the operational risk prediction can be improved. By obtaining adjustment instructions based on the operational risks and controlling the stacking device, the risk of cargo overturning and damage can be reduced in a timely and effective manner, reducing the property losses caused by cargo overturning accidents.
[0056] Figure 3 It is a schematic diagram of predicting operational risks according to some embodiments of this specification.
[0057] In some embodiments, the operational risk includes a first risk, and the control system can predict the first risk 330-1 through the first risk prediction model 320 based on the first cargo data 310-1 and the operation data 310-2. For relevant content about the first cargo data and the operation data, see Figure 2 and its related contents.
[0058] The first risk refers to the risk of the goods on the stacker tipping over when the stacker is heading to the target shelf. The first risk can be expressed as a value from 0 to 10, and the larger the value, the greater the risk of the goods tipping over.
[0059] The first risk prediction model is a model for predicting the first risk. The first risk prediction model may be a machine learning model, such as a deep neural network (DNN) model or a neural network (NN) model.
[0060] In some embodiments, the input of the first risk prediction model may include first cargo data and operation data; the output may be the first risk.
[0061] The first risk prediction model is trained using the swing data of the first sample cargo data in the historical data as training data. The swing data of the first sample cargo data refers to the swing data corresponding to the first sample cargo under the first sample cargo data and the sample operation data.
[0062] In some embodiments, the control system may train a first risk prediction model based on a large number of first samples with first labels. The first sample may include first sample cargo data and sample operation data. The first label may be a first sample risk corresponding to the first sample cargo under the first sample cargo data and the sample operation data. The first sample risk may be determined based on the swing data corresponding to the first sample cargo under the first sample cargo data and the sample operation data.
[0063] The first sample cargo is historical cargo that serves as the data source of the training sample of the first risk prediction model. The first sample cargo data is the first cargo data corresponding to the first sample cargo. The sample operation data is the operation data when the stacker transports the first sample cargo.
[0064] The swing data of the first sample cargo data refers to the swing of the first sample cargo during the process of the stacker transporting the first sample cargo to the target shelf. For example, the swing data of the first sample cargo data may include the swing amplitude and swing frequency of the first sample cargo.
[0065] The swing amplitude of the first sample goods refers to the deviation amplitude of the first sample goods in the process of going to the target shelf. For example, the amplitude of the center of gravity of the first sample goods deviating from the support point or the maximum swing amplitude of the same point of the first sample goods during the entire transportation process. The support point refers to the contact point or the midpoint of the contact surface between the stacker and the first sample goods. The swing amplitude can be expressed in angle or distance.
[0066] The swing frequency of the first sample goods refers to the number of swings per unit time of the first sample goods in the process of moving to the target shelf. When the swing amplitude is greater than the preset amplitude, the movement of the first sample goods can be recorded as one swing. The preset amplitude can be manually preset based on the goods size of the first sample goods, for example, the preset amplitude can be 3% of the goods size.
[0067] In some embodiments, a plurality of different sample combinations may be formed based on different first sample cargo data and different sample operation data, and the first risks corresponding to each sample combination may be manually marked based on the swing data of the first sample cargo data corresponding to each sample combination.
[0068] For example, for the first cargo data that actually capsized in history, the first label of the first sample corresponding to the data may be marked with a risk value of 10. For another example, when the swing amplitude is not greater than 5% of the size of the first sample cargo, and the swing frequency is not greater than 3 times per minute, and the maximum number of swings is not greater than 6 times per minute, the first label corresponding to the first sample cargo data may be marked with a risk value of 3. For another example, when the swing amplitude is not less than 20% of the size of the first sample cargo, and the swing frequency is not less than 7 times per minute, and the maximum number of swings is not less than 10 times per minute, the first label corresponding to the first sample cargo data may be marked with a risk value of 9. The above description of the value of the first label is only an example.
[0069] Some embodiments of the present specification determine the first risk through a first risk prediction model, taking into account the condition of the cargo itself and the swinging condition during movement, as well as the impact on the capsizing of the cargo. By learning historical data, a more accurate first risk can be obtained.
[0070] In some embodiments, the operation risk includes a second risk. The control system can predict the second risk 330-2 through a second risk prediction model 350 based on the second cargo data 340-1 and the cargo data to be piled 340-2. For more information about the second cargo data, see Figure 2 and its related contents.
[0071] The second risk refers to the risk of goods tipping over when being stacked on the shelves. The second risk can be expressed as a value from 0 to 10, and the larger the value, the greater the risk of goods tipping over.
[0072] The second risk prediction model is a model for determining the second risk, and may be a machine learning model, such as a deep neural network (DNN) model.
[0073] The input of the second risk prediction model may include the second cargo data and the cargo to be piled data; the output may be the second risk.
[0074] The data of goods to be piled refers to data related to the goods to be piled on the shelf. The data of goods to be piled may include the quantity, size and shape of the goods to be piled on the shelf next time. The method for obtaining the data of goods to be piled can refer to the method for obtaining the first goods data.
[0075] The second risk prediction model is trained using the swing data of the second sample cargo data in the historical data as training data. The swing data of the second sample cargo data refers to the swing data corresponding to the second sample cargo data and the sample cargo data to be piled.
[0076] In some embodiments, the control system can train a second risk prediction model based on a large number of second samples with second labels. The second samples include second sample cargo data and sample cargo data to be stacked. The second label can be the second sample risk corresponding to the second sample cargo under the second sample cargo data and the sample cargo data to be stacked. The second sample risk can be determined based on the swing data corresponding to the second sample cargo under the second sample cargo data and the sample cargo data to be stacked.
[0077] The second sample goods are historical goods that serve as the data source of the training samples of the second risk prediction model. The second sample goods data are the second goods data corresponding to the second sample goods. The sample to-be-stacked goods data are the to-be-stacked goods data of the goods that are ready to be stacked on the shelves in the next stacking after the second sample goods are stacked on the shelves.
[0078] The swing data of the second sample cargo data refers to the swing of the second sample cargo during the process of the stacker stacking the second sample cargo to the target shelf. For example, the swing data of the second sample cargo data may include the swing amplitude and swing frequency of the second sample cargo.
[0079] The swing amplitude of the second sample goods refers to the deviation amplitude of the second sample goods during the process of stacking to the target shelf. For example, the amplitude of the center of gravity of the second sample goods deviating from the support point or the maximum swing amplitude of the same point of the second sample goods during the entire stacking process. Please refer to the previous text for the description of the support point and the swing amplitude.
[0080] The swing frequency of the second sample goods refers to the number of swings per unit time of the second sample goods in the process of being stacked on the target shelf. For instructions on how to determine the number of swings, see the previous text.
[0081] The method of determining the second sample risk value corresponding to the second label based on the swing data of the second sample cargo data is similar to the method of determining the first sample risk value corresponding to the first label based on the swing data of the first sample cargo data. Please refer to the corresponding content in the previous text for details.
[0082] In some embodiments, the control system may further determine future stacking parameters of the stacker.
[0083] The future stacking parameters refer to the stacking parameters corresponding to the subsequent stacking of goods by the stacker. For example, the stacking parameters may include parameters such as stacking height, stacking depth, stacking width, and stacking direction.
[0084] In some embodiments, the process of the control system determining the future stacking parameters of the stacker may include the following steps:
[0085] S1. In response to the second risk exceeding a second risk threshold: determining a plurality of process radar charts corresponding to different stacking processes based on values of the second cargo data in a plurality of feature dimensions during the stacking process.
[0086] The second risk threshold refers to the maximum value of the second risk when the goods need to be adjusted, which can be based on manual preset.
[0087] Feature dimensions may include the quantity, height, degree of deformation, degree of tilt, and other dimensions of the cargo.
[0088] A radar chart is a visual chart that can display data of multiple dimensions in the same chart. In some embodiments, the control system can draw a radar chart based on the values of different characteristic dimensions of the second cargo data.
[0089] For example, during the stacking process, as goods are continuously stacked, characteristics such as the height, quantity, tilt, and deformation of the goods on the shelf will change. Therefore, during the stacking process, the values of different characteristic dimensions of the second goods data will also change accordingly.
[0090] The process radar chart is a plurality of radar charts drawn according to different values of the second cargo data in different feature dimensions in different stacking processes.
[0091] The stacking process may refer to the entire process from the start of stacking to the completion of stacking. In some embodiments, the control system may divide the entire process from the start of stacking to the completion of stacking into a plurality of sub-processes based on a preset unit time period, each sub-process corresponds to a unit time period, and a process radar chart corresponds to the second cargo data being drawn at different characteristic dimensions after the completion of a sub-process. Based on the entire stacking process, a plurality of process radar charts corresponding to different stacking processes may be obtained.
[0092] As an example only, a second process radar chart corresponding to the second unit time period can be drawn based on the second cargo data obtained in the second unit time period, in which the values of the four characteristic dimensions of cargo quantity, height, deformation degree, and tilt degree are 3 pieces, 1.5m, 4cm, and 6° respectively.
[0093] S2: Determine actual change characteristics corresponding to the second cargo data based on the multiple process radar charts.
[0094] The actual change feature refers to the change feature of the second cargo data in one or more feature dimensions during the stacking process, described based on the process radar chart. The actual change feature may include multiple sub-change features.
[0095] The sub-change feature includes the change value of the second cargo data in one or more feature dimensions in two adjacent process radar charts.
[0096] In some embodiments, the sub-change feature can be represented based on a feature sequence, and each element in the feature sequence represents the difference in the values of the second cargo data in each feature dimension in two adjacent process radar charts. For example, the first sub-change feature can be determined based on the difference between the values of the second cargo data in each feature dimension in the second process radar chart and the values of the second cargo data in each feature dimension in the first process radar chart.
[0097] As an example only, the first sub-change feature can be expressed as a feature sequence, wherein a1=A2-A1, A2 is the value of the second cargo data in the second process radar chart in terms of the cargo quantity (such as 3 pieces in the aforementioned example), and A1 is the value of the second cargo data in the first process radar chart in terms of the cargo quantity. Correspondingly, b1, c1, and d1 correspond to the differences in height, deformation degree, and tilt degree between the two process radar charts, respectively.
[0098] S3. Determine a final radar chart based on the values of the second cargo data in multiple feature dimensions after the stacking is completed.
[0099] The final radar chart refers to a radar chart drawn based on the values of the second cargo data in multiple feature dimensions when the stacking is completed.
[0100] S4. Determine the actual shape feature corresponding to the second cargo data based on the process radar chart and the final radar chart.
[0101] The actual shape feature refers to the shape feature corresponding to the stacking radar chart, which can be represented based on the values of each feature dimension in the stacking radar chart. For example, the actual shape feature can be represented as {A, B, C, D}. A, B, C, and D are the values of the stacking radar chart in the four feature dimensions of cargo quantity, height, deformation degree, and tilt degree.
[0102] The stacking radar chart can be determined based on the process radar chart and the final radar chart. As an example only, the stacking radar chart also includes four characteristic dimensions: the quantity of goods, height, degree of deformation, and degree of inclination. The value of the quantity of goods is the standard deviation or variance of the value of the quantity of goods in each process radar chart and the final radar chart, and the values of the height, degree of deformation, and degree of inclination are also determined in a similar manner. Based on the process radar chart and the final radar chart, the control system can draw the actual shape features corresponding to the second cargo data.
[0103] S5. Determine a preset shape feature and a preset change feature based on the second cargo data.
[0104] The preset shape feature and the preset change feature refer to the standard shape feature and the standard change feature corresponding to the second cargo data. In some embodiments, the preset change feature may include a plurality of preset sub-change features. For the description of the sub-change features, please refer to the corresponding content above.
[0105] In some embodiments, standard shape characteristics and standard change characteristics corresponding to different second cargo data can be preset based on historical experience or test data. Under the standard shape characteristics and standard change characteristics, the risk of overturning, deformation and damage of the second cargo corresponding to the second cargo data is the lowest or lower than a preset threshold.
[0106] In some embodiments, the control system may preset multiple standard sub-process radar charts and standard final radar charts corresponding to different second cargo data based on historical experience or test data. The preset sub-change characteristics, preset change characteristics, and preset shape characteristics may also be determined based on multiple standard sub-process radar charts and standard final radar charts. The specific determination method is similar to the aforementioned process of determining the actual shape characteristics and actual change characteristics corresponding to the second cargo data based on the process radar chart and the final radar chart. For specific descriptions, please refer to the aforementioned description of determining the actual change characteristics and actual shape characteristics.
[0107] S6. Determine a first distance between the actual shape feature and the preset shape feature, and a second distance between the actual change feature and the preset change feature.
[0108] In some embodiments, the control system may calculate the first distance between the actual shape feature and the preset shape feature in a variety of ways, for example, based on a Euclidean distance algorithm or a Manhattan distance algorithm.
[0109] In some embodiments, the control system may calculate the second distance between the actual change feature and the preset change feature in a variety of ways. For example, the control system may determine the similarity between two sets of feature sequences with time information corresponding to the actual change feature and the preset change feature by a dynamic time warping method, and determine the second distance based on the similarity, for example, the greater the similarity, the smaller the second distance.
[0110] Among them, the actual change feature corresponds to a group of actual sub-change features (including multiple actual sub-change features), each actual sub-change feature corresponds to a feature sequence, and the preset change feature corresponds to a group of preset sub-change features (including multiple preset sub-change features), each preset sub-change feature corresponds to a feature sequence.
[0111] In some embodiments, the second distance between the actual change feature and the preset change feature can also be determined based on multiple second sub-distances between multiple groups of actual sub-change features and corresponding preset sub-change features. For example, the second distance is the mean or weighted sum of multiple second sub-distances. A second sub-distance is represented based on the Euclidean distance or cosine distance between an actual sub-change feature and its corresponding (e.g., both corresponding to the same unit time period) preset sub-change feature.
[0112] S7. Determine future stacking parameters based on the first distance and the second distance.
[0113] Determining the future stacking parameters may include determining characteristic dimensions and adjustment values that need to be adjusted in the current stacking parameters.
[0114] In some embodiments, when the first distance and the second distance are both smaller than the first distance threshold and the second distance threshold, respectively, it can be determined that the current stacking parameters do not need to be adjusted, that is, the current stacking parameters can be directly used for future stacking parameters.
[0115] In some embodiments, in response to the first distance being greater than the first distance threshold, the control system may use a feature dimension whose difference between the corresponding feature dimension of the actual shape feature and the preset shape feature is greater than the first distance threshold as a feature dimension that needs to be adjusted, and use the absolute value of the difference between the corresponding feature dimensions as the adjustment value of the feature dimension.
[0116] For example, when the first distance threshold is 0.5, the feature dimension whose corresponding feature dimension difference between the actual shape feature (3, 7, 4, 1.5) and the preset shape feature (3, 6, 3.9, 1.5) is greater than the first distance threshold is the second feature dimension, and the adjustment value of the feature dimension is |7-6|=1.
[0117] In some embodiments, in response to the second distance being greater than a second distance threshold, the control system may take a feature dimension whose difference between a corresponding feature dimension in an actual change feature and a preset change feature is greater than the second distance threshold as a feature dimension that needs to be adjusted.
[0118] The larger the second distance is, the greater the change in the second cargo data in certain characteristic dimensions is, and the stacking parameters need to be adjusted significantly. For example, a comparison table of the comparison relationship between the second distance values of each characteristic dimension and the parameter adjustment values of the stacking parameters can be preset based on historical experience, and the parameter adjustment values of the corresponding stacking parameters can be determined by querying the comparison table based on the second distance value.
[0119] For more information on adjusting stacking parameters, please refer to the relevant instructions below.
[0120] Some embodiments of the present specification compare the shape characteristics and change characteristics of preset goods with the shape characteristics and change characteristics of actual goods in history, and adjust the stacking parameters of the goods to be stacked based on the differences to reduce the risk of cargo overturning.
[0121] Some embodiments of the present specification determine the second risk through a second risk prediction model, taking into account the risk that subsequent goods stacked on the shelf may cause to the entire shelf during the stacking process, and a more accurate second risk can be obtained by learning historical data.
[0122] In some embodiments, the control system may determine the target vector 360 based on the feature values of the second cargo data in multiple feature dimensions; and determine the third risk 330 - 3 based on the matching result of the target vector 360 in the target database 370 .
[0123] The third risk refers to the risk of goods tipping over after being stacked on the shelves. The third risk can be expressed as a value from 0 to 10. The larger the value, the greater the risk of goods tipping over.
[0124] In some embodiments, the control system may determine the third risk by performing vector matching in the target database based on the feature values of the second cargo data in multiple feature dimensions.
[0125] The target vector refers to the feature vector corresponding to the current second cargo data. The target vector can be constructed based on the second cargo data to characterize the second cargo data, and the elements of the target vector can include feature values of multiple dimensions of the second cargo data.
[0126] The target database may include multiple reference vectors and the risk value corresponding to each reference vector. The reference vector is constructed based on the feature values of the historical second cargo data after historical stacking in multiple feature dimensions. The control system can determine the risk value corresponding to each reference vector based on the feature values of one or more feature dimensions of the historical second cargo data in a variety of ways. For example, the control system can determine the risk value corresponding to each reference vector based on the rule that the height, weight, tilt degree and deformation degree of the cargo are respectively positively correlated with the risk value, and the load-bearing capacity and shape regularity of the shelf are respectively negatively correlated with the risk value.
[0127] By way of example only, the control system may determine the risk value based on the following formula (1):
[0128] Risk value = (h × m × t × S) ÷ (b × k) (1)
[0129] Among them, h is the height of the goods; k is the regularity of the shape of the goods, which is a value between 0 and 1, and the larger the value, the more regular the shape; m is the weight of the goods, which can be obtained based on the number of pieces of goods and the weight of a single piece of goods (wherein the weight of a single piece of goods can be determined based on manual input); t is the degree of inclination of the goods; b is the load-bearing capacity of the shelf; S is the degree of deformation of the goods, which is a value between 0 and 1, and the larger the value, the greater the degree of deformation of the goods.
[0130] In some embodiments, the control system can construct a target vector based on the second cargo data and a reference vector based on the second cargo data after historical stacking. The control system calculates the vector distance between the target vector and multiple reference vectors, uses the reference vector whose vector distance is less than the distance threshold as the control vector, and uses the risk value corresponding to the control vector as the current third risk. The distance threshold refers to the maximum value of the vector distance at which two vectors are considered similar, and can be based on manual preset.
[0131] Some embodiments of this specification determine the third risk by means of vector matching. Based on a large amount of data, a predicted value of the third risk that is closer to the current situation can be obtained, thereby reducing the risk of cargo capsizing.
[0132] Figure 4 It is a schematic diagram of determining the adjustment instruction according to some embodiments of this specification.
[0133] In some embodiments, the control system may determine a target adjustment degree 410 based on the operational risk 330 ; and determine an adjustment instruction 420 based on the target adjustment degree 410 .
[0134] In some embodiments, the intelligent stacking device further includes one or more laser ranging units, which may include devices such as a laser transmitter, a laser reflector, and a timer. In some embodiments, the laser ranging unit may be mechanically connected to the lifting structure of the stacker and communicatively connected to the control system. For example, one or more laser ranging units may be installed on the lifting device of the stacker to obtain the second cargo data on the shelf in real time during the stacking process.
[0135] The target adjustment refers to the characteristic dimensions and adjustment values of various parameters that need to be adjusted when the intelligent stacking device is adjusted to a suitable working state. The target adjustment may include at least one of the adjustment of the operating parameters 410-1, the adjustment of the stacking parameters 410-2, the adjustment of the damping parameters 410-3, and the adjustment of the laser ranging unit 410-4. For detailed descriptions of the operating parameters, stacking parameters, and damping parameters, please refer to Figure 2 and its related contents.
[0136] In some embodiments, the target adjustment degree can be determined by querying a first preset table based on the operation risk, standard parameters, and current parameters. Standard parameters refer to various parameters corresponding to the optimal operation state of the intelligent stacking device, which can be based on manual presets. Current parameters refer to various parameters of the current intelligent stacking device. The first preset table includes the correspondence between various historical operation risks, historical standard parameters, historical current parameters, and historical target adjustment degrees.
[0137] The adjustment degree of the operating parameters refers to the dimension and adjustment value that the intelligent stacking device needs to adjust when it runs next time.
[0138] The adjustment degree of the shock absorption parameters refers to the degree of shock absorption adjustment of the adjustable shock absorber.
[0139] The adjustment degree of the stacking parameters refers to the dimension and adjustment value of the stacking parameters that need to be adjusted during the next stacking.
[0140] The adjustment degree of the laser ranging unit includes the number of laser ranging units enabled and the irradiation intensity. In some embodiments, the laser ranging unit can measure the second cargo data of multiple cargoes on the shelf. Multiple laser ranging units do not need to be turned on at the same time. For example, when the second risk or the third risk is small, a small number of laser ranging units can be turned on to reduce the irradiation intensity and save energy.
[0141] In some embodiments, the control system can adaptively adjust the operating data and the damping parameters in response to the first risk exceeding the first risk threshold; including determining the adjustment direction of the operating parameters based on the swing data of the first cargo data; and determining the adjustment degree of the operating parameters and the adjustment degree of the damping parameters based on the first risk. For a detailed description of the first risk, the first distance threshold, and the swing data of the first cargo data, please refer to Figure 3and its related contents.
[0142] The adjustment direction of the operating parameters refers to the dimension of the operating parameters to be adjusted and the direction of the increase or decrease of the parameter value. The dimension of the operating parameters may include the speed and acceleration of the stacker. For example, the adjustment direction of the operating parameters may be to reduce the speed of the stacker.
[0143] The first risk threshold refers to the minimum value of the first risk when adjustment of operating data and shock absorption parameters is required, and can be preset based on historical experience.
[0144] In some embodiments, the control system may determine the adjustment direction of the operating parameter based on the swing data of the first cargo data. When the value of a certain dimension in the swing data of the first cargo data exceeds the corresponding swing threshold, the value of the corresponding dimension needs to be adjusted.
[0145] The swing threshold refers to the minimum value when the swing data of the first cargo data needs to be adjusted. The swing threshold may include a swing amplitude threshold and a swing frequency threshold. For example, when the swing amplitude exceeds the swing amplitude threshold, the acceleration of the stacker needs to be adjusted; when the swing frequency is greater than the swing frequency threshold, the speed of the stacker needs to be adjusted.
[0146] In some embodiments, the control system may adjust the speed variation characteristics of the stacker based on the swing stability.
[0147] Swing stability refers to the stability of the swing of goods. Swing stability can be expressed by swing speed and swing frequency. Swing speed and swing frequency are negatively correlated with swing stability. Swing speed and swing frequency can be obtained based on manual statistics.
[0148] The speed change characteristic may include a speed change amplitude and a speed change frequency of the stacker. The speed change characteristic may characterize the stability of the speed of the stacker. In some embodiments, the control system may determine the speed change characteristic of the stacker based on the swing data of the first cargo data. The control system may query a second preset table based on the current swing speed and swing frequency to adjust the speed change characteristic of the current stacker. The second preset table includes the relationship between the historical swing speed and swing frequency and the historical speed change characteristic under stable operation.
[0149] In some embodiments, the control system may determine the adjustment degree of the operating parameters and the damping parameters based on the first risk. For example, the control system may determine the adjustment degree of the stacker speed and acceleration in the operating parameters based on the first risk.
[0150] In response to the first risk being greater than the first risk threshold, the control system may obtain the stacker speed at that time. For example, the stacker speed at that time is v1, and the stacker operating speed corresponding to the first threshold is the safe speed v2.
[0151] In some embodiments, the control system can adjust the stacker speed by using a proportional integral differential control algorithm (PID). The PID control algorithm is an algorithm that combines a proportional algorithm, an integral algorithm, and a differential algorithm to control the output result and obtain an accurate result.
[0152] The proportional term, integral term and differential term in the PID control algorithm are positively correlated with the adjustment degree. For example, the control system can determine the adjustment degree of the stacker's operating speed through the following formula (2):
[0153] Adjustment degree = Kp×e(t)+Ki×∫e(t)dt+Kd×de(t) / dt (2)
[0154] Among them, Kp, Ki and Kd are proportional coefficient, integral coefficient and differential coefficient respectively, e(t) is the current speed deviation, and e(t)=v1-v2, de(t) / dt is the rate of change of speed deviation.
[0155] Kp×e(t) is the proportional term in the PID control algorithm, which can adjust the size of the speed deviation. Ki×∫e(t)dt is the integral term in the PID control algorithm, which can be used to accumulate the historical value of the speed deviation and eliminate the steady-state error in the calculation. Kd×de(t) / dt is the differential term in the PID control algorithm, which can predict the future trend of the speed deviation.
[0156] The acceleration of the stacker and the adjustment degree of the acceleration can also be determined by the above algorithm, which will not be described in detail.
[0157] In some embodiments, the proportional coefficient Kp, the integral coefficient Ki and the differential coefficient Kd are respectively related to the weight and / or height of the currently stacked goods. For example, the proportional coefficient Kp and the differential coefficient Kd are respectively positively correlated with the weight and height of the goods, and the integral coefficient Ki is negatively correlated with the weight and height of the goods.
[0158] Some embodiments of this specification take into account the impact of the weight and height of the cargo on the coefficients of the PID control algorithm, adjust the coefficients to maintain the balance of the algorithm, offset the impact of the weight and height of the cargo, and reduce the risk of cargo capsizing.
[0159] In some embodiments, the control system may determine the adjustment degree of the damping parameter in a variety of ways. For example, the control system may determine the adjustment degree of the damping parameter by a fuzzy control algorithm.
[0160] The control system can first set the input of the fuzzy variable to the first risk and the degree of shock absorption; the output of the fuzzy variable is set to the degree of shock absorption adjustment. A fuzzy set is constructed based on the first risk and the degree of shock absorption of the adjustable shock absorber. Then, through the membership function, the fuzzy variables are mapped to the fuzzy set, and the values of the input variables are converted into the membership values of each fuzzy set. Then, a fuzzy rule database is constructed to map the input variables (first risk, degree of shock absorption) to the output variables (degree of shock absorption adjustment). Then, the input variables are mapped to the corresponding fuzzy sets, and the fuzzy rule database is used for inference calculation to obtain the output membership. Finally, the output membership of different rules is mapped to the actual shock absorption adjustment amount to obtain the adjustment degree of the shock absorption parameters.
[0161] Exemplarily, the control system can divide the first risk into three levels of "low", "medium" and "high", divide the damping degree of the adjustable shock absorber into three levels of "small", "medium" and "large", and divide the damping adjustment degree into "small", "medium", "large" and "maximum". And set the input fuzzy variables to the first risk and the damping degree, and set the output fuzzy variables to the damping adjustment degree. Taking the triangular membership function as an example, the membership value of each fuzzy set is calculated separately by the triangular membership function calculation formula. For example, the membership values of the three sets corresponding to the first risk can be obtained, which are μA1, μA2 and μA3 respectively; and the membership values of the three sets corresponding to the damping degree can be obtained, which are μB1, μB2 and μB3 respectively.
[0162] Exemplarily, the triangular membership function is shown in the following formula, where formula (3-1) represents the fuzzy set of the first risk being "low", formula (3-2) represents the fuzzy set of the first risk being "medium", and formula (3-3) represents the fuzzy set of the first risk being "high":
[0163]
[0164]
[0165]
[0166] Among them, x is the current actual first risk, a is the minimum value of the independent variable x, b is the average value of the independent variable x, and c is the maximum value of the independent variable x.
[0167] When a is 3, b is 6, c is 9, and x is 4, that is, when the first risk input is 4, the membership value corresponding to the "low" risk set The membership value corresponding to the “medium” risk set The membership value μA3 corresponding to the "high" risk set is 0. According to the maximum membership principle, the fuzzy set corresponding to the value with the largest membership is the set to which the current input value belongs. That is, when the first risk is 4, the risk level corresponding to the first risk is "low".
[0168] Similarly, μB1, μB2 and μB3 can be calculated through the triangular membership function.
[0169] The control system can build a fuzzy rule database based on manual presets. Fuzzy rules refer to the mapping relationship between input fuzzy variables and output fuzzy variables. For example, when the first risk is "low" and the damping degree is "small", the damping adjustment amount is "fine tuning", etc.
[0170] The control system maps the input fuzzy variables to the output fuzzy variables through the fuzzy rule database to obtain the membership values corresponding to the output variables. For example, the four levels of "small", "medium", "large" and "maximum" of the damping adjustment degree can correspond to μY1, μY2, μY3 and μY4 respectively. The control system can calculate the output membership values of different fuzzy rules respectively, that is, different damping adjustment amounts can be obtained under different first risks and different damping degrees.
[0171] Exemplarily, when the first risk is 4, the membership corresponding to the first risk being "low" is μA1=0.6; the membership corresponding to the shock absorption degree being "small" is μB1=0.8, and the membership corresponding to the adjustment amount being "fine-tuning" is μY1=0.9; the fuzzy rule is (μA1(AND)μB1)×μY1, that is, through a union operation, the smaller value of the membership corresponding to the risk level and the shock absorption degree is selected and multiplied by the membership corresponding to the adjustment amount to obtain the output membership value of the fuzzy rule; then the corresponding membership of the fuzzy rule that is finally output is μ=(μA1(AND)μB1)×μY1=0.6×0.9=0.54.
[0172] Finally, the control system can map the membership of the output corresponding to different fuzzy rules to the actual damping adjustment degree in a variety of ways. For example, the control system can defuzzify by the average method, the centroid method or the maximum membership method. For example, in the above example, the membership calculated by the fuzzy rule (μA1(AND)μB1)×μY1 is 0.54, and the actual damping adjustment degree corresponding to 0.54 can be the damping force adjustment p and the elastic coefficient adjustment q, where p and q are natural numbers.
[0173] Some embodiments of this specification use PID control algorithms to determine the adjustment degree of operating parameters, which can buffer and adjust the speed and acceleration so that they fluctuate around the safe value and gradually stabilize. This can avoid the instability and shaking of goods caused by sudden changes in speed / acceleration.
[0174] Other embodiments of the present specification use a fuzzy control algorithm based on the first risk to calculate the degree of shock absorption adjustment. The degree of shock absorption can be dynamically adjusted based on different first risks and the current degree of shock absorption, thereby saving electricity consumption caused by always using maximum shock absorption.
[0175] In some embodiments, the control system can determine the stacking parameters and the opening degree of the laser ranging unit in response to the second risk exceeding the second risk threshold and / or the third risk exceeding the third risk threshold. For details about the second risk, the second risk threshold and the third risk, please refer to Figure 3 and its related contents.
[0176] The third risk threshold refers to the minimum value of the third risk when the intelligent stacking device needs to be adjusted, and can be based on manual preset.
[0177] In some embodiments, the control system may determine the adjustment degree of the stacking parameter and the activation degree of the laser ranging unit in response to the combined risk being greater than the combined risk threshold.
[0178] The opening degree refers to the working condition of the laser ranging unit, which can include the number of laser ranging units turned on and the intensity of the laser.
[0179] The combined risk refers to the risk value calculated by adding the second risk and the third risk according to a certain weight. The combined risk threshold is the preset minimum combined risk value when the stacking parameters and the laser ranging unit need to be adjusted.
[0180] The weights of the second and third risks are related to the different dimensions of the stacking parameters and the degree of activation of the laser ranging unit. For example, for the adjustment degree of the stacking direction, the second risk has a greater impact, and its corresponding weight is larger (for example, 0.9), while the third risk has a smaller weight (for example, 0.1).
[0181] The control system can determine different dimensions of the stacking parameters and the opening degree of the laser ranging unit based on a third preset table. The third preset table contains different dimensions of the stacking parameters, the opening degree of the laser ranging unit, and the relationship between the weights corresponding to the second risk and the third risk. The third preset table can be based on manual preset.
[0182] In some embodiments, the control system can determine the adjustment degree of different dimensions of the stacking parameters by querying the fourth preset table based on the relationship between the joint risk threshold and the multiple of the joint risk. The fourth preset table contains the relationship between the joint risks of various stacking parameters in different dimensions and the multiple of the corresponding joint risk threshold. For example, when the joint risk corresponding to the stacking height is twice the joint risk threshold, the adjustment degree corresponding to the stacking height at this time is to reduce it to half of the original height, and increase the stacking width to twice the original.
[0183] In some embodiments, the control system can determine the opening degree of the laser ranging unit based on the joint risk threshold and the fuzzy control rules. The control system can divide the joint risk into three levels of "low", "medium" and "high"; divide the number of laser ranging units turned on into "small", "large" and "all"; and divide the laser intensity into "weak", "medium" and "strong". The control system can map different joint risks to different opening degrees through fuzzy reasoning based on preset fuzzy control rules, and then obtain the actual opening degree through defuzzification. For a specific description of the fuzzy control algorithm, please refer to the above description of determining the adjustment degree of the shock absorption parameters.
[0184] Some embodiments of the present specification determine the combined risk based on the second risk and the third risk, and based on the combined risk, judge the dimensions and adjustment values of the stacking parameters that need to be adjusted and the adjustment degree of the laser ranging unit, consider the impact of potential risks during and after the stacking process on the target adjustment degree, and use weights to measure their impact on different elements in the target adjustment degree, so as to obtain a more accurate target adjustment degree.
[0185] In some embodiments, the dimension of the stacking parameter that needs to be adjusted is related to the second cargo data. For a detailed description of the second cargo data, see Figure 2 and its related contents.
[0186] The control system can determine the adjustment degree of the stacking parameters through the stacking parameter adjustment model. The stacking parameter adjustment model is a model used to determine the adjustment degree of the stacking parameters. The stacking parameters can be a machine learning model, for example, a deep neural network (DNN) model, etc. The input of the stacking parameter adjustment model may include the second cargo data after a stacking (such as the current second cargo data), the first distance and the second distance; the output may be the adjustment degree of the stacking parameters in different dimensions. For a detailed description of the first distance and the second distance, please refer to Figure 3 and its related contents.
[0187] In some embodiments, the control system can train the stacking parameter adjustment model based on a large number of third samples with third labels. The third samples include sample second cargo data without overturning after each historical stacking, and the corresponding sample first distance and sample second distance. The third label can be the adjustment degree of the stacking parameters in different dimensions without overturning after each historical stacking.
[0188] Some embodiments of the present specification determine the adjustment degree of stacking parameters through a machine learning model. Based on the difference between the current stacking situation and the preset stacking situation, the stacking height, depth, width, stacking direction and other parameters for the next stacking can be determined and adjusted, thereby obtaining a more adaptive adjustment degree of stacking parameters.
[0189] In some embodiments, the control system may determine an adjustment instruction based on the target adjustment degree. For example, the parameters to be adjusted in the target adjustment degree and their corresponding adjustment values are used as adjustment instructions to adjust each parameter.
[0190] Some embodiments of the present specification determine the target adjustment degree 410 based on the operation risk, and then determine the adjustment instruction to adjust the parameters. Through various algorithms, for different operation stages of the intelligent stacking device, such as before stacking, during stacking, and after stacking, the impact on the future of each parameter is considered, and each parameter is dynamically adjusted to ensure the normal operation of the intelligent stacking device and prevent accidents such as cargo overturning.
[0191] Some embodiments of the present specification provide an intelligent stacking device, including at least one memory and at least one processor, wherein the memory is used to store computer instructions; and the processor is used to execute the above-mentioned intelligent stacking method.
[0192] Some embodiments of the present specification provide a computer-readable storage medium, which stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the above-mentioned intelligent stacking method.
[0193] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.
[0194] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.
[0195] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.
[0196] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.
[0197] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values is as accurate as possible within the feasible range.
[0198] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, documents, etc., cited in this specification are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the contents of this specification, documents that limit the broadest scope of the claims of this specification (currently or later attached to this specification) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the materials attached to this specification are inconsistent or conflicting with the contents described in this specification, the descriptions, definitions, and / or use of terms in this specification shall prevail.
[0199] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.
Claims
1. An intelligent stacking device, comprising a stacker, an image recognition module, a sensor module, an adjustable shock absorber and a control system, wherein the image recognition module, the sensor module and the adjustable shock absorber are mechanically connected to the stacker and are communicatively connected to the control system; The image recognition module is configured to recognize first cargo data on the stacker and second cargo data on the shelf; The sensor module is configured to obtain operation data of the stacker; The adjustable shock absorber is disposed at the bottom of the pallet of the stacker and is configured to reduce shock to the stacker; The control system is configured to: predicting an operation risk of the stacker based on the first cargo data, the second cargo data, and the operation data; Based on the operation risk, an adjustment instruction is determined, the adjustment instruction including the adjustment amount of the damping parameter of the adjustable shock absorber; the operation risk includes a first risk, a second risk and a third risk; the first risk is the risk of the goods on the stacker tipping over when the stacker is heading to the target shelf; the second risk is the risk of the goods being stacked tipping over when the goods are being stacked on the shelf; the third risk is the risk of the goods tipping over after being stacked on the shelf; The control system is further configured to: Predicting the first risk by using a first risk prediction model based on the first cargo data and the operation data; The first risk prediction model is a machine learning model, and the first risk prediction model is trained using the swing data of the first sample cargo data in the historical data as training data; The intelligent stacking device further comprises one or more laser distance measuring units, the laser distance measuring units being mechanically connected to the lifting structure of the stacker and being communicatively connected to the control system; The control system is further configured to: Based on the operation risk, determining a target adjustment degree of the stacker, the target adjustment degree comprising at least one of an adjustment degree of an operating parameter, an adjustment degree of a stacking parameter, an adjustment degree of a damping parameter, and an adjustment degree of the laser ranging unit; Based on the target adjustment degree, determining the adjustment instruction; In response to the first risk exceeding a first risk threshold, adaptively adjusting the operation data and shock absorption parameters of the stacker, including: determining an adjustment direction of the stacker operation parameters based on the swing data of the first cargo data; determining an adjustment degree of the stacker operation parameters and an adjustment degree of the shock absorption parameters based on the first risk; In response to the second risk exceeding the second risk threshold, and / or the third risk exceeding the third risk threshold, stacking parameters and the activation degree of the laser distance measuring unit are determined.
2. The device according to claim 1, The control system is further configured to: Predicting the second risk by using a second risk prediction model based on the second cargo data and the cargo to be piled data; The second risk prediction model is a machine learning model, and the second risk prediction model is trained using the swing data of the second sample cargo data in the historical data as training data.
3. The device according to claim 1 The control system is further configured to: Determining a target vector based on feature values of the second cargo data in multiple feature dimensions; The third risk is determined based on a matching result of the target vector in a target database.
4. A stacker control method, It is characterized in that Based on the control system of the intelligent stacking device according to claim 1, the method comprises: Acquiring first cargo data, second cargo data, and operation data; predicting an operation risk of a stacker crane based on the first cargo data, the second cargo data, and the operation data; Based on the operation risk, determining an adjustment instruction, the adjustment instruction including an adjustment amount of a damping parameter of an adjustable damper of the stacker; The operation risks include the first risk, the second risk and the third risk; the first risk is the risk of the goods on the stacker tipping over when the stacker is heading to the target shelf; the second risk is the risk of the goods being stacked tipping over when the goods are being stacked on the shelf; the third risk is the risk of the goods tipping over after being stacked on the shelf; The predicting the operation risk of the stacker based on the first cargo data, the second cargo data and the operation data includes: Predicting the first risk by using a first risk prediction model based on the first cargo data and the operation data; The first risk prediction model is a machine learning model, and the first risk prediction model is trained using the swing data of the first sample cargo data in the historical data as training data; The intelligent stacking device further comprises one or more laser distance measuring units, the laser distance measuring units being mechanically connected to the lifting structure of the stacker and being communicatively connected to the control system; The determining of the adjustment instruction based on the operational risk includes: Based on the operation risk, determining a target adjustment degree of the stacker, the target adjustment degree comprising at least one of an adjustment degree of an operating parameter, an adjustment degree of a stacking parameter, an adjustment degree of a damping parameter, and an adjustment degree of the laser ranging unit; Based on the target adjustment degree, determining the adjustment instruction; In response to the first risk exceeding a first risk threshold, adaptively adjusting the operation data and shock absorption parameters of the stacker, including: determining an adjustment direction of the stacker operation parameters based on the swing data of the first cargo data; determining an adjustment degree of the stacker operation parameters and an adjustment degree of the shock absorption parameters based on the first risk; In response to the second risk exceeding the second risk threshold, and / or the third risk exceeding the third risk threshold, stacking parameters and the activation degree of the laser distance measuring unit are determined.
5. The method according to claim 4, It is characterized in that The operation risk includes a second risk, and the operation risk of the stacker is predicted based on the first cargo data, the second cargo data and the operation data, including: Predicting the second risk by using a second risk prediction model based on the second cargo data and the cargo to be piled data; The second risk prediction model is a machine learning model, and the second risk prediction model is trained using the swing data of the second sample cargo data in the historical data as training data.
6. A stacker control device, It is characterized in that The apparatus comprises at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is configured to execute at least part of the computer instructions to implement the stacker control method according to any one of claims 4 to 5.
7. A computer-readable storage medium storing computer instructions, It is characterized in that After the computer reads the computer instructions in the storage medium, the computer executes the stacker control method as claimed in any one of claims 4 to 5.
