Packaging carton forming control system and control method

Abstract

The invention relates to the technical field of carton forming control, in particular to a packaging carton forming control system and method, and the system comprises a structure analysis module, a sequence judgment module, a state matching module, an action constraint module and an instruction generation module. According to the method, the paperboard size and the fold line position are subjected to structured analysis, the section sequence relation is established, the forming sequence can be adaptively adjusted along with the change of the carton structure, the folding process is mapped into the state evolution relation, and action constraint judgment is conducted in combination with the real-time position of an execution mechanism; the broken line actions form a mutual exclusion coordination relation in space and time, so that mechanical interference and sequence conflicts among different forming stages are avoided, forming instructions have continuity and consistency, stable operation is kept under the conditions of specification switching and complex structures, manual intervention requirements are reduced, and the overall forming reliability is improved.

Description

Technical Field

[0001] This invention relates to the field of carton forming control technology, and in particular to a packaging carton forming control system and control method. Background Technology

[0002] The field of carton forming control technology refers to the related technical fields of equipment operation coordination and motion control involved in the process of corrugated cardboard or cartons transforming from a flat state to a three-dimensional box on the packaging production line. Its core issues include cardboard conveying, positioning, folding lines, creasing, side panel folding, box unfolding, bottom or side shaping, and the coordinated action sequence and timing of the driving components. This technical field usually relies on the packaging machinery structure combined with transmission mechanisms such as conveyor belts, push plates, connecting rods, cylinders, and motors. By uniformly controlling the action sequence, positional relationship, and timing relationship of each actuator, it can adapt to the continuous forming requirements of cartons of different specifications.

[0003] Traditional packaging carton forming control systems and methods refer to control methods used to issue commands and manage the sequence of mechanical actions in carton forming equipment. The technical issue it addresses is the coordinated operation of multiple processes during the carton forming process. Typically, through a pre-set forming process, after the cardboard enters the forming station, the conveying action, positioning action, folding action, and shaping action are triggered in sequence. The specific method includes setting a fixed sequence of actions according to the carton specifications, using limit switches or position sensors to detect the cardboard's position and then driving the corresponding cylinder or motor to complete the folding or pushing operation, and outputting control signals through the controller according to a predetermined logical sequence to complete the entire carton forming process.

[0004] Traditional carton forming control relies on preset fixed processes and mechanical triggering conditions. The sequence of actions and timing relationships are fixed during the equipment debugging stage. During operation, folding and shaping operations are mainly performed step by step based on the cardboard arrival signal. It lacks the ability to dynamically analyze the differences in cardboard structure and changes in fold line distribution. When faced with complex structures with multiple fold lines or frequent changes in specifications, it is easy to have unreasonable folding sequence relationships, overlapping or mutual interference of actuator actions, resulting in decreased forming stability, extended adjustment time, increased reliance on human experience, and increased risk of equipment idle travel and malfunction. Summary of the Invention

[0005] To address the technical problems existing in the prior art, embodiments of the present invention provide a packaging carton forming control system and control method. The technical solution is as follows: On the one hand, a packaging carton forming control system is provided, the system comprising: The structure analysis module collects the length, width, height and fold line positions of the packaging cardboard, calculates the corresponding relationship between the cardboard length and fold line positions, divides the fold line segments and sorts them, generates a structural segment sequence and passes it to the sequence determination module; The sequence determination module inputs the structural segment sequence into the topological sorting algorithm to determine the sequential relationship of the multi-fold line segments in carton forming, arranges the bottom plate, side wall and folding edge segments in sequence, generates forming sequence data and transmits it to the state matching module. The state matching module inputs the forming sequence data into a finite state machine, matches the forming sequence with the folding and pressing state conditions in the state machine, analyzes the corresponding states of multiple forming segments, generates an action stage chain, and transmits it to the action constraint module. The motion constraint module detects the position of the actuator based on the motion stage chain, calculates the start and end positions of the motion by combining the fold line position and the carton height, performs mutual exclusion judgment on adjacent fold line positions, generates motion constraint data and transmits it to the instruction generation module. The instruction generation module obtains the current position of the actuator, compares it with the action constraint data to determine whether the action start and action termination conditions are met, and generates a formed instruction sequence by combining the corresponding states of the action stage chain.

[0006] As a further embodiment of the present invention, the structural segment sequence includes fold line segments, fold line positions, and cardboard length; the forming sequence data includes bottom plate folding sequence, side wall folding sequence, edge folding sequence, and segment association sequence; the action stage chain includes folding stage state, pressing stage state, holding stage state, and releasing stage state; the action constraint data includes action start position value, action end position value, segment mutual exclusion identifier, and position valid range; and the forming instruction sequence includes start instruction code, end instruction code, and stage switching instruction.

[0007] As a further aspect of the present invention, the structure analysis module specifically comprises: The size acquisition submodule collects data on the length, width, height, and fold line positions of the packaging cardboard. It performs consistency checks on the collected values, performs boundary checks on the fold line positions based on the cardboard length baseline, records the calibration coordinates of multiple fold lines in the unfolding direction, and generates a set of unfolded cardboard dimensions. The fold line segment submodule performs position calculations based on the cardboard unfolding size set, the cardboard length value, and the multi-fold line calibration coordinates. It then divides the fold lines into segments based on their relative positions in the unfolding direction, analyzes the continuous segment identifiers based on the spacing between adjacent fold lines and the boundary positions, and generates a fold line segment distribution table. The segment generation submodule calls the polyline segment distribution table, performs sequential arrangement for multi-polyline segments, sorts the segments according to their starting coordinates in the unfolding direction of the cardboard, eliminates overlapping segments and retains segment identifiers, and generates a structural segment sequence.

[0008] As a further aspect of the present invention, the reference value for the cardboard length is determined by collecting multiple sets of cardboard length data, sorting all the collected cardboard length values, and analyzing the median value of the sorted cardboard length data.

[0009] As a further aspect of the present invention, the sequence determination module specifically comprises: The segment sequence submodule inputs the structural segment sequence into the topology sorting algorithm to calculate the forming order, collects the multi-structural segment numbers, connection relationship identifiers and adjacent constraint markers, expands nodes and records edges according to the pointing relationship between segments, and generates segment association information. The sequential relationship submodule, based on the segment association information, records the folding trigger order identifier for multi-fold line segments in the carton forming process, compares the execution order of the bottom plate folding segment, the side wall folding segment, and the flange folding segment, and performs a sequential comparison based on the constraint pointing between segments to obtain the segment sequential constraint sequence. The sequence arrangement submodule, based on the sequence of constraints of the segments, calls the multi-segment folding identifier and constraint release condition to perform sequential arrangement operation, outputs the segment number stream according to the constraint resolution order, performs integrity verification on the arrangement result, and generates the formed sequence data.

[0010] As a further aspect of the present invention, the state matching module specifically comprises: The sequential receiving submodule, based on the forming sequence data as the starting input, collects the segment number, folding identifier and sequence index corresponding to each forming sequence item, calculates the difference between adjacent indices to determine whether there is a sequence breakpoint, and generates a forming sequence index set. The state determination submodule, based on the forming sequence index set, calls the preset folding state condition table and pressing state condition table in the finite state machine, performs a state condition comparison for the segment number corresponding to each sequence index, determines the state affiliation of the current sequence item, and generates an action state mapping sequence. The stage chain generation submodule, based on the action state mapping sequence, performs continuity judgment on the state identifiers of adjacent sequential items, detects state identifier change points and records stage switching positions, and reassembles multi-state segments according to the sequential index to generate an action stage chain.

[0011] As a further embodiment of the present invention, the finite state machine consists of a set of states, a set of input conditions, state transition relationships, an initial state, and a set of decision states.

[0012] As a further aspect of the present invention, the action constraint module specifically comprises: The position detection submodule detects the position signal of the actuator based on the action stage chain, collects the stroke value and stage identifier value corresponding to the current position, performs a consistency judgment on the stage identifier, records the coordinate sequence corresponding to the current position in multiple action stages, and generates the actuator position sequence. The boundary calculation submodule obtains the broken line position and carton height based on the position sequence of the actuator, compares the current position with the broken line position and calculates the action start position condition value, compares the current position with the carton height and calculates the action end position condition, and generates an action position boundary parameter group. The mutual exclusion determination submodule, based on the action position boundary parameter group, obtains the adjacent broken line positions in the structural segment sequence and performs mutual exclusion determination, marks the boundary intervals where actions cannot be triggered simultaneously, analyzes the dataset including the constraint relationship between the start condition and the termination condition, and generates action constraint data.

[0013] As a further aspect of the present invention, the instruction generation module specifically comprises: The position acquisition submodule acquires the current position of the actuator, collects the stroke value and timestamp identifier corresponding to the current position, performs a consistency judgment on the stroke value interval, records the coordinate status of the current position within the current sampling period, and generates the current position status quantity of the actuator. The constraint determination submodule, based on the current position state of the actuator, and combined with the action start boundary and action end boundary in the action constraint data, compares the current position value with multiple boundary values ​​to determine whether it has entered the start interval or the end interval, and generates an action command trigger flag. The sequence generation submodule, based on the action instruction trigger flag, calls the corresponding stage state value of the action stage chain, performs a matching judgment between the stage state and the execution order of the trigger flag, arranges the instruction codes corresponding to each stage, and generates a formed instruction sequence.

[0014] On the other hand, a method for controlling the forming of a packaging carton, which is based on the aforementioned packaging carton forming control system, includes the following steps: S1: Collect the length, width, height and fold line positions of the packaging cardboard, calculate the corresponding relationship between the cardboard length and the fold line positions, divide the fold line segments and sort them to generate a structural segment sequence; S2: Input the structural segment sequence into the topological sorting algorithm to determine the sequential relationship of the multi-fold line segments in the carton forming process, and arrange the bottom plate, side wall and flange folding segments in sequence to generate forming sequence data; S3: Input the molding sequence data into a finite state machine, match the molding sequence with the folding and pressing state conditions in the state machine, analyze the corresponding states of multiple molding segments, and generate an action stage chain. S4: Based on the current position of the actuator detected by the action stage chain, the start and end positions of the action are calculated by combining the fold line position and the carton height. Mutual exclusion judgment is performed on adjacent fold line positions to generate action constraint data. S5: Obtain the current position of the actuator, compare it with the action constraint data to determine whether the action start and action termination conditions are met, and generate a formed instruction sequence by combining the corresponding state of the action stage chain.

[0015] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: By structurally analyzing the cardboard size and fold line positions and establishing the sequence relationship of segments, the forming sequence can be adaptively adjusted according to changes in the carton structure. The folding process is mapped as a state evolution relationship and action constraint judgment is made in combination with the real-time position of the actuator. This enables each fold line action to form a mutually exclusive and coordinated relationship in space and time, thereby avoiding mechanical interference and sequence conflicts between different forming stages. This ensures that the forming instructions are continuous and consistent, maintains stable operation under specification switching and complex structural conditions, reduces the need for manual intervention, and improves the overall forming reliability. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a system schematic diagram of the present invention; Figure 2 This is a schematic diagram of the system framework of the present invention; Figure 3 This is a flowchart of the structure analysis module in this invention; Figure 4 This is a flowchart of the sequence determination module in this invention; Figure 5 This is a flowchart of the state matching module in this invention; Figure 6 This is a flowchart of the action constraint module in this invention; Figure 7 This is a flowchart of the instruction generation module in this invention; Figure 8 This is a flowchart of the method of the present invention. Detailed Implementation

[0018] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0019] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0020] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0021] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0022] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0023] This invention provides a packaging carton forming control system, such as... Figure 1-2 The diagram shown illustrates a packaging carton forming control system, which includes: The structure analysis module collects the length, width, height and fold line positions of the packaging cardboard, calculates the corresponding relationship between the cardboard length and fold line positions, divides the fold line segments and sorts them, generates a structural segment sequence and passes it to the sequence determination module; The sequence determination module inputs the structural segment sequence into the topological sorting algorithm to determine the sequential relationship of the multi-fold line segments in carton forming, arranges the bottom plate, side wall and folding edge segments in sequence, generates forming sequence data and transmits it to the state matching module. The state matching module inputs the molding sequence data into the finite state machine, matches the molding sequence with the folding and pressing state conditions in the state machine, analyzes the corresponding states of multiple molding segments, generates an action stage chain, and passes it to the action constraint module. The motion constraint module detects the position of the actuator based on the motion stage chain, calculates the start and end positions of the motion by combining the fold line position and the carton height, performs mutual exclusion judgment on adjacent fold line positions, generates motion constraint data and transmits it to the instruction generation module. The instruction generation module obtains the current position of the actuator, compares the action constraint data to determine whether the action start and termination conditions are met, and generates a complete instruction sequence by combining the corresponding states of the action stage chain.

[0024] The structural segment sequence includes fold line segments, fold line positions, and cardboard lengths. The forming sequence data includes the bottom plate folding sequence, side wall folding sequence, edge folding sequence, and segment association sequence. The action stage chain includes the folding stage state, pressing stage state, holding stage state, and releasing stage state. The action constraint data includes the action start position value, action end position value, segment mutual exclusion identifier, and valid position range. The forming instruction sequence includes the start instruction code, end instruction code, and stage switching instruction.

[0025] Specifically, such as Figure 2 , 3 As shown, the structure parsing module specifically consists of: The size acquisition submodule collects data on the length, width, height, and fold line positions of the packaging cardboard. It performs consistency checks on the collected values, performs boundary checks on the fold line positions based on the cardboard length baseline, records the calibration coordinates of multiple fold lines in the unfolding direction, and generates a set of unfolded cardboard dimensions. In the initial inspection station of the automated cardboard production line, a high-frequency laser displacement sensor array is first activated to perform a full-coverage scan of the corrugated cardboard to be inspected, which is laid flat on the conveyor belt. The sensor is controlled to continuously capture data along the unfolding direction of the cardboard at a sampling frequency of 500 Hz, acquiring information on the physical edge contour and surface indentation depth of the cardboard. The raw point cloud data fed back by the sensor is read, filtered, and the cardboard length is extracted as 1200.50 mm and the cardboard width as 800.00 mm. The standard height of the cardboard corresponding to the current batch is retrieved according to the production task sheet, which is 400.00 mm. At the same time, three obvious transverse fold lines are identified on the surface of the cardboard, and their position data in the acquisition coordinate system is read. For the collected values, a consistency check was immediately performed. Cardboard width was selected as the key verification indicator. Width data was collected at five evenly distributed cross-sectional locations perpendicular to the cardboard length. The obtained measurement sets were 799.80 mm, 800.10 mm, 800.00 mm, 799.90 mm, and 800.20 mm, respectively. An arithmetic mean was calculated, adding the five values ​​to obtain a total of 4000.00 mm. This sum was then divided by the number of sampling points (5) to obtain an average width of 800.00 mm. The absolute value of the deviation between each sampling point and the average value was then calculated, with the maximum absolute deviation being 0.20 mm. A preset consistency deviation threshold of 0.50 mm was applied. This threshold is set based on the allowable tolerance range of the cardboard cutting process, i.e., ±0.50 mm of cutting accuracy. The calculated maximum deviation of 0.20 mm was compared with the threshold of 0.50 mm. Since 0.20 mm was less than 0.50 mm, the consistency of the cardboard width data was confirmed to meet the quality requirements. Subsequently, the collected fold line positions were checked against the baseline value of 1200.50 mm for the cardboard length. The position readings of the three fold lines in the original coordinate system were 300.30 mm, 600.50 mm, and 900.80 mm, respectively. The check process compared each fold line position value with the baseline length value to verify whether it was within the valid physical range of greater than 0.00 mm and less than 1200.50 mm. After comparison, all three values ​​met the boundary conditions. After confirming the data's validity, the starting point of the cardboard's physical edge was defined as the zero point coordinate of 0.00 mm in the unfolding direction, and the ending point was defined as 1200.50 mm. The positions of the three fold lines that passed the check were then locked as the calibrated coordinates in the unfolding direction. Finally, the starting coordinates, the coordinates of the three broken lines, and the coordinates of the ending point are integrated in sequence to generate a set of unfolded cardboard dimensions containing five key nodes. The data items in this set are 0.00 mm, 300.30 mm, 600.50 mm, 900.80 mm, and 1200.50 mm, respectively, thus completing the digital calibration of the cardboard's planar structure.

[0026] The fold line segment submodule performs position calculations based on the cardboard length value and the multi-fold line calibration coordinates according to the cardboard unfolding size set. It divides the fold lines into segments based on their relative positions in the unfolding direction, analyzes the continuous segment identifiers based on the spacing between adjacent fold lines and the boundary position, and generates a fold line segment distribution table. The generated set of unfolded cardboard dimensions is called, and a rigorous position calculation is performed based on the cardboard length values ​​and the coordinates of the multi-fold lines. This step aims to transform discrete coordinate points into continuous physical segment lengths. The calculation logic is as follows: along the unfolding direction of the cardboard, each subsequent coordinate value is subtracted from the previous one to obtain the scalar distance between adjacent coordinate points. The specific calculation process is as follows: First, the length of the first segment is calculated using the second coordinate value 300.30 - 0.00 = 300.30 mm; second, the length of the second segment is calculated using the third coordinate value 600.50 - 300.30 = 300.20 mm; third, the length of the third segment is calculated using the fourth coordinate value 900.80 - 600.50 = 300.30 mm; finally, the length of the fourth segment is calculated using the endpoint coordinate value 1200.50 - 900.80 = 299.70 mm. Based on the relative positions of the fold lines in the unfolding direction, the cardboard is divided into four independent physical segments according to the above calculations, and segment identifiers ID-01, ID-02, ID-03, and ID-04 are assigned sequentially. Furthermore, based on the analysis of the spacing between adjacent fold lines and the boundary positions, continuous segment identifiers are determined, and a "minimum effective segment threshold" is introduced as the criterion for segment validity. This threshold is set with reference to the physical characteristics of corrugated cardboard, taking the average thickness of three-layer corrugated cardboard (5.00 mm) as the base, multiplied by a safety factor of 1.5 for the folding process, resulting in a threshold of 7.50 mm. Comparing the calculated lengths of the four segments with 7.50 mm: 300.30 mm, 300.20 mm, 300.30 mm, and 299.70 mm are all significantly greater than 7.50 mm, indicating that all divided segments have sufficient physical length to support folding and that there is no risk of structural interference due to excessively narrow spacing. Subsequently, combining the carton height of 400.00 mm and the length characteristics of each segment obtained in the previous steps, the segment attributes were classified, and ID-01 and ID-03 were determined to be long sides, while ID-02 and ID-04 were wide sides. Finally, the segment identifiers, corresponding start and end coordinates, calculated segment lengths, and attribute determination results were integrated to generate the polyline segment distribution table shown in Table 1 below.

[0027] Table 1 Distribution Parameters of Broken Line Sections As shown in Table 1, the table details the geometric parameters of each segment after calculation and verification, clarifies the precise positional distribution and physical properties of each segment in the unfolding direction of the cardboard, indicating that the structural division of the cardboard has been digitally mapped, and the length of each segment meets the process threshold requirements.

[0028] The segment generation submodule calls the polyline segment distribution table, performs sequential arrangement for multi-polyline segments, sorts the segments according to their starting coordinates in the unfolding direction of the cardboard, eliminates overlapping segments and retains segment identifiers, and generates a structural segment sequence. The data from the polyline segment distribution table is retrieved, and a sequential sorting operation is performed on the multi-polyline segments. Although the data in the input table already presents a preliminary logical order, a coordinate-based sorting algorithm is enforced to ensure the absolute orderliness of the data after multi-threaded processing. The "starting coordinates" field of all segments is extracted as the sorting key, and the segment objects are sorted in ascending order using quicksort logic. In this embodiment, the extracted key sequence is 0.00, 300.30, 600.50, and 900.80, respectively. The order is confirmed by comparing the numerical values, and the physical connection order of the segments is locked as ID-01 to ID-02, then to ID-03, and finally to ID-04. Subsequently, depth overlap detection is performed based on the starting coordinates of the segments in the unfolding direction of the cardboard to eliminate overlapping segments and retain the correct segment identification. An "overlap judgment threshold" is set, which is three times the sensor measurement accuracy of 0.05 mm, and is set to 0.15 mm after multiplication. To demonstrate the elimination logic, an abnormal redundant segment ID-Err was detected, with a recorded start coordinate of 300.25 mm and an end coordinate of 300.35 mm. A difference calculation was performed for judgment: first, the difference between the end coordinate of the normal segment ID-01 (300.30 mm) and the start coordinate of the next segment ID-02 (300.30 mm) was calculated, resulting in 0.00 mm, less than 0.15 mm, indicating a normal connection. Next, the absolute value of the difference between the abnormal segment ID-Err's start coordinate of 300.25 mm and ID-01's end coordinate of 300.30 mm was calculated, resulting in 0.05 mm. Since 0.05 mm is less than the set overlap judgment threshold of 0.15 mm, ID-Err was determined to have an illegal spatial overlap with the existing main frame segment. Further analysis revealed that the length of ID-Err was only 0.10 mm, far less than the effective segment threshold, confirming it as noise data generated by sensor jitter. The rejection instruction is executed immediately, removing ID-Err from the sequence and retaining only ID-01, ID-02, ID-03, and ID-04 as valid segments. After cleaning and final confirmation, the identifiers of each retained segment are concatenated according to their physical connection order to generate the final structural segment sequence "ID-01, ID-02, ID-03, ID-04". This sequence accurately reflects the actual physical structure of the cardboard along the unfolding direction and serves as the sole basis for subsequent folding control.

[0029] Specifically, such as Figure 2 , 4 As shown, the sequence determination module is specifically as follows: The segment sequence submodule inputs the structural segment sequence into the topology sorting algorithm to calculate the forming order, collects the multi-structural segment numbers, connection relationship identifiers and adjacent constraint markers, expands nodes and records edges according to the pointing relationship between segments, and generates segment association information; The system receives the generated structural segment sequence "ID-01, ID-02, ID-03, ID-04", which contains four cardboard segments arranged according to their spatial physical location. The input matrix for the topology sorting algorithm is initialized, and a 4x4 adjacency matrix is ​​created to store the connection states between segments, with all initial values ​​set to 0. The system begins collecting multi-segment numbering and corresponding boundary coordinate data. Based on the data from the previous embodiment, the termination coordinate of ID-01 is 300.30 mm, and the starting coordinate of ID-02 is 300.30 mm. The "connection determination tolerance" is set to 0.05 mm; this parameter is set based on the repeatability accuracy of the laser sensor. The connection relationship identifiers are calculated as follows: The difference between the ending coordinate of ID-01 (300.30 mm) and the starting coordinate of ID-02 (300.30 mm) is calculated, resulting in a difference of 0.00 mm; the difference between the ending coordinate of ID-02 (600.50 mm) and the starting coordinate of ID-03 (600.50 mm) is calculated, resulting in a difference of 0.00 mm; the difference between the ending coordinate of ID-03 (900.80 mm) and the starting coordinate of ID-04 (900.80 mm) is calculated, also resulting in a difference of 0.00 mm. Based on these difference results, it is determined that the absolute value of each difference is less than the connection determination tolerance value of 0.05 mm, confirming the validity of the physical connection relationship. The adjacency matrix is ​​updated by setting the elements pointing from ID-01 to ID-02 to 1, the elements pointing from ID-02 to ID-03 to 1, and the elements pointing from ID-03 to ID-04 to 1, forming a unidirectional connected graph structure. Subsequently, adjacency constraint markers are defined, and the non-zero elements in the adjacency matrix are traversed to identify ID-02 as the direct successor node of ID-01, ID-03 as the direct successor node of ID-02, and ID-04 as the direct successor node of ID-03. For each connection point, the "folding degree of freedom" parameter is recorded, with a default value of 180 degrees, indicating that adjacent segments can be folded in reverse along the connection line. Based on the above matrix relationship, the nodes are expanded, with ID-01 defined as the root node (starting end) and ID-04 as the leaf node (ending end), and the connection edge attribute is recorded as "rigid connection". Finally, the segment number, adjacency matrix data, connection point coordinates, and folding degree of freedom are integrated to generate segment association information containing the complete topology. This information clarifies the linear dependency relationship from ID-01 to ID-04, providing a basic data model for subsequent folding logic calculations.

[0030] The sequential relationship submodule, based on the segment association information, records the folding trigger order identifier for multi-fold line segments in the carton forming process, compares the execution order of the bottom plate folding segment, the side wall folding segment, and the flange folding segment, and performs a sequential comparison based on the constraint pointing between segments to obtain the segment sequential constraint sequence. The system calls upon segment association information and, based on standard folding logic from the carton forming process library, quantitatively allocates the folding trigger order for multi-fold line segments. First, it identifies the role of each segment in the forming process. Based on segment length and position attributes (ID-01 is 300.30 mm long, ID-04 is 299.70 mm long and located at both ends, ID-02 and ID-03 are located in the middle), ID-01 and ID-04 are marked as "flange folding segments," and ID-02 and ID-03 are marked as "bottom and side wall folding segments." A "folding priority coefficient" is set as the basis for order comparison. This coefficient ranges from 1 to 10, with smaller values ​​indicating earlier folding actions. The setting rules are as follows: the priority coefficient for the outer flange segment is set to 1, and the priority coefficient for the middle structural segment is set to 5. The following steps are performed: For each segment, a priority comparison and assignment operation is performed: ID-01, as the first segment, is assigned a priority coefficient of 1; ID-04, as the last segment, is assigned a priority coefficient of 1; ID-02 and ID-03, as intermediate support segments, are assigned a priority coefficient of 5. Then, a sequence of constraints is constructed based on the inter-segment constraint directions. The comparison process is as follows: ID-01 (coefficient 1) is compared with ID-02 (coefficient 5). Since 1 is less than 5, ID-01's folding action is determined to precede ID-02; ID-04 (coefficient 1) is compared with ID-03 (coefficient 5). Since 1 is less than 5, ID-04's folding action is determined to precede ID-03. Simultaneously, for ID-01 and ID-04 with the same priority, based on the "feeding direction priority" principle, ID-01, located at the front of the feeding direction, is determined to perform fine-tuning before ID-04. The correction priority of ID-01 is set to 1.0, and the correction priority of ID-04 is set to 1.1. Finally, the folding trigger sequence identifier is recorded, and the segment folding attribute table shown in Table 2 below is generated.

[0031] Table 2. Segment Folding Attributes and Priorities As shown in Table 2, the table details the physical role definition and quantified priority parameters for each segment. Based on the priority values ​​sorted from smallest to largest, the sequential constraint logic was established, with ID-01 acting first, ID-04 second, and ID-02 and ID-03 remaining relatively stationary or cooperating synchronously last, generating the final segment sequential constraint sequence.

[0032] The sequence arrangement submodule, based on the sequence of constraints of the segments, calls the multi-segment folding identifier and constraint release conditions to perform sequential arrangement operation, outputs the segment number stream according to the constraint resolution order, performs integrity verification on the arrangement result, and generates the finalized sequence data; The system receives the sequence of segment constraints, calls the multi-segment folding identifiers containing "Order-01" to "Order-03", and begins sequential sorting operations. A task queue is created, initially containing ID-01, ID-04, ID-02, and ID-03. The constraint release condition is set as follows: the correction priority value of the current segment must be less than the values ​​of all other remaining segments in the queue, or the predecessor dependency node of the current segment has been processed. An iterative loop is started. In the first scan, the correction priority of ID-01 (1.0) is read and compared with other items in the queue (1.1, 5.0). 1.0 is determined to be the minimum value, satisfying the constraint release condition. ID-01 is removed from the task queue, pushed into the output stream, and its status is marked as "folded". In the second scan, the remaining segments in the queue are ID-04, ID-02, and ID-03. Read the correction priority 1.1 of ID-04, compare it with the remaining items (5.0), and determine that 1.1 is the current minimum value. Furthermore, the potential interference object ID-01 in its physical location has been marked as "folded," satisfying the spatial constraint release condition. Remove ID-04 from the queue and push it into the output stream. In the third scan, ID-02 and ID-03 remain, both with a priority of 5.0 and physically consecutive. Identify them as a related whole, determine them as a synchronous action group, and simultaneously push ID-02 and ID-03 into the output stream. At this point, the output stream order is: ID-01→ID-04→{ID-02, ID-03}. Subsequently, perform an integrity check on the arrangement result. Calculate the sum of the lengths of each segment in the output stream: add the lengths of ID-01 (300.30 mm), ID-04 (299.70 mm), ID-02 (300.20 mm), and ID-03 (300.30 mm). The calculation process is as follows: 300.30 + 299.70 + 300.20 + 300.30 = 1200.50 mm. Comparing this result with the original cardboard's total length of 1200.50 mm, the difference is 0.00 mm, indicating that all sections are included and no data is lost. This result shows that the forming sequence data is physically closed-loop and complete. The generated forming sequence data "ID-01 / ID-04 / [ID-02&ID-03]" accurately reflects the working sequence of the carton on the folding machine, i.e., first folding the leading long side, then folding the tail wide side, and finally completing the enclosure with the middle box body.

[0033] Specifically, such as Figure 2 , 5 As shown, the state matching module specifically consists of: The sequential receiving submodule, based on the forming sequence data as the starting input, collects the segment number, folding identifier and sequence index corresponding to each forming sequence item, calculates the difference between adjacent indices to determine whether there is a sequence breakpoint, and generates a forming sequence index set. The system receives the generated molding sequence data "ID-01, ID-04, [ID-02 & ID-03]", which is transmitted as a data stream and contains the arrangement information of individual segments and combined segments. The sequence parsing program is immediately started, performing structured acquisition for each molding sequence item to extract the segment number, fold identifier, and sequence index. For the first item in the sequence, ID-01, it is identified as an independent segment, its position in the data stream is read, its sequence index is assigned 1, and its fold identifier is marked as "one-way fold". For the second item, ID-04, its sequence index is assigned 2, and its fold identifier is marked as "one-way fold". For the third item, [ID-02 & ID-03], it is identified as a synchronous combined item, treated as a whole, its sequence index is assigned 3, and its fold identifier is marked as "synchronous compression". After completing the basic data acquisition, a sequence continuity calculation is performed to detect whether there is packet loss or logical discontinuity during data transmission. The "sequential breakpoint determination threshold" is set to 1. This value is based on the logic of the natural growth of integer sequences, meaning that ideally, the difference between adjacent indices should be strictly equal to 1. The differences between adjacent indices are calculated sequentially: first, the difference between the second index 2 and the first index 1 is calculated as 2 − 1 = 1; then, the difference between the third index 3 and the second index 2 is calculated as 3 − 2 = 1. The calculated difference results are then compared with the breakpoint determination threshold 1. In this embodiment, all adjacent differences are strictly equal to 1, indicating that the sequence index exhibits continuous natural growth, and there are no abnormal situations such as a difference of 2 (meaning a missing item) or a difference of 0 (meaning duplicate processing). After confirming the sequence is correct, the data items that have passed the above verification are encapsulated, establishing a strong mapping relationship between the index and the physical segment, generating a formed sequential index set containing three core index nodes. This set defines the absolute sequence of steps in the process execution: the first step is to process ID-01, the second step is to process ID-04, and the third step is to process ID-02 and ID-03 simultaneously, providing precise timing input for subsequent state machine calls.

[0034] The state determination submodule, based on the forming sequence index set, calls the pre-set folding state condition table and pressing state condition table in the finite state machine, performs a state condition comparison for the segment number corresponding to each sequence index, determines the state affiliation of the current sequence item, and generates an action state mapping sequence. The forming sequence index set is invoked, and the pre-set folding state condition table and pressing state condition table in the finite state machine (FSM) are loaded. "Forming resistance torque" is defined as the core physical parameter for state attribution determination; this parameter reflects the mechanical torque characteristics required by the cardboard at different process stages. The "folding state torque threshold" is set to 10.0 N·m, and the "pressing state torque threshold" is set to 30.0 N·m. The thresholds are set based on the following: the experimentally measured average resistance torque of corrugated cardboard folding freely along the crease line is 3.5 N·m; considering equipment friction and safety margin, 10.0 N·m is set as the upper limit of the folding action; while the box forming pressing needs to overcome the board springback and glue adhesion force, with an average requirement of 45.0 N·m, so 30.0 N·m is set as the lower limit of the pressing action. State condition comparison is performed for the segment number corresponding to each sequence index. First, ID-01 corresponding to index 1 is processed, and its action is simulated by calling the simulation engine, obtaining its forming resistance torque as 2.5 N·m. The comparison operation is performed: 2.5 < 10.0 and 2.5 < 30.0, indicating that the value falls within the folding state range, and the state of ID-01 is defined as "dynamic folding state". Next, ID-04 corresponding to index 2 is processed, and its molding resistance torque is obtained as 2.8 N·m. The operation is performed: 2.8 < 10.0, also indicating "dynamic folding state". Finally, the combination item [ID-02 & ID-03] corresponding to index 3 is processed, and its synchronous molding resistance torque is obtained as 48.5 N·m. The comparison is performed: 48.5 > 10.0 and 48.5 > 30.0, indicating that the value is significantly higher than the pressing threshold and falls within the pressing state range, and its state is defined as "static pressing state". After completing the comparison of all items, the state labels are mapped one-to-one with the sequence indices to generate an action state mapping sequence. The sequence clearly depicts the changes in the physical properties of the process: the first two steps are low-torque rapid folding, and the third step is high-torque strong pressing. This result indicates that the kinetic characteristics of different process segments have been successfully identified.

[0035] The stage chain generation submodule, based on the action state mapping sequence, performs continuity judgment on the state identifiers of adjacent sequential items, detects state identifier change points and records stage switching positions, and reassembles multi-state segments according to the sequential index to generate an action stage chain. The system receives an action state mapping sequence containing three ordered nodes: Node 1 (ID-01, dynamic folded state), Node 2 (ID-04, dynamic folded state), and Node 3 (ID-02 / 03, static compression state). It initiates continuity judgment logic to merge discrete steps with the same process attributes into a continuous operation phase. A "state switching identifier" is set and initialized to 0. The system iterates through the state identifiers of adjacent sequential items: first, it compares Node 1 and Node 2; both are in the "dynamic folded state," and the logical operation result is "isomorphic," indicating no phase switching occurs. Node 1 and Node 2 are merged into the same action phase, temporarily named Phase-A. Next, it compares Node 2 and Node 3; Node 2 is in the "dynamic folded state," and Node 3 is in the "static compression state." Logical operation reveals mutual exclusion, indicating a state abrupt change, identified as a "state identifier change point," and recorded as the phase switching boundary. A segmentation instruction is triggered, ending the input of Phase-A, starting a new phase Phase-B, and assigning Node 3 to Phase-B. The reorganized multi-state segments are encapsulated in order of sequence index to generate an action stage chain. To visually represent the stage division results and their physical meaning, an action stage chain parameter table is constructed as shown in Table 3 below.

[0036] Table 3. Parameter Table of Action Phase Chain As shown in Table 3, this table details the process stage information after chain-like reorganization. The final generated action stage chain clarifies the control logic of the production line: first, Phase-A is executed, controlling the servo motor to continuously complete the folding of ID-01 and ID-04 in a high-speed, low-torque mode; then, switching to Phase-B, controlling the hydraulic or high-power motor to complete the pressing and shaping of ID-02 and ID-03 in a low-speed, high-torque mode. This result demonstrates the successful transformation of discrete process steps into physically meaningful continuous operation stages, providing a segmentation basis for subsequent batch processing of controller parameters.

[0037] Specifically, such as Figure 2 , 6 As shown, the action constraint module is specifically as follows: The position detection submodule, based on the action stage chain, detects the position signal of the actuator, collects the stroke value and stage identifier value corresponding to the current position, performs a consistency judgment on the stage identifier, records the coordinate sequence corresponding to the current position in multiple action stages, and generates the actuator position sequence. The generated "action stage chain" is retrieved as the logical reference. A sampling period of 2 milliseconds is set to collect position signal feedback from each axial actuator in real time. Within the current time slice, the current position corresponding to the stroke value fed back by the linear slide encoder is read as 285.60 mm, and the current stage identifier value corresponding to the clock is read as "Phase-A" (corresponding to the dynamic folding stage in the previous embodiment). A consistency judgment is performed on the stage identifier to verify whether the physical position is within the valid spatial domain defined by the logical stage. The "stage space tolerance threshold" is set to 2.00 mm, which is based on the maximum allowable value of the servo dynamic following error. First, the theoretical starting index physical coordinates corresponding to Phase-A (i.e., the starting broken line position of ID-01) of 300.30 mm are extracted. Combined with the preset "action lead" of 15.00 mm (this value is calculated by multiplying the conveyor belt speed of 1.5 m / s by the response time of 0.01 seconds), the theoretical cutting position of Phase-A is calculated to be 285.30 mm. Numerical comparison is performed: the current travel value of 285.60 mm is calculated by subtracting the theoretical cutting position of 285.30 mm, resulting in a deviation value of 0.30 mm. Then, the absolute value of this deviation, 0.30 mm, is compared with the stage spatial tolerance threshold of 2.00 mm. Since 0.30 mm is less than 2.00 mm, the current physical position is confirmed to be valid and belongs to Phase-A, with no logical misalignment. After confirmation, the current travel value of 285.60 mm, the timestamp, the phase identifier Phase-A, and the deviation value of 0.30 mm are packaged into a coordinate data packet. This acquisition and judgment process is continuously executed within the multi-action phase, recording the coordinate state at each moment at a frequency of 500 Hz, forming a continuous set of trajectory points. Finally, the discrete point set is smoothed to remove noise data, generating an actuator position sequence. This sequence accurately records the spatial displacement history of the actuator from approaching the polygonal point to completing the folding action, providing measured data support for subsequent precise boundary calculations.

[0038] The boundary calculation submodule obtains the broken line position and carton height based on the actuator position sequence, compares the current position with the broken line position and calculates the action start position condition value, compares the current position with the carton height and calculates the action end position condition, and generates the action position boundary parameter group; The actuator position sequence is invoked, and core structural parameters are retrieved from the database: zigzag line position data (ID-01 corresponds to 300.20 mm, ID-02 corresponds to 600.50 mm, etc.) and the carton height value of 400.00 mm. First, the current position is compared with the zigzag line position, and the action initiation position condition value is calculated. Taking the ID-01 segment as an example, the "start-up lag compensation coefficient" is set to 0.8. This coefficient is used to compensate for the action delay of pneumatic components or robotic arms. The calculation process is as follows: Selecting the zigzag line position of 300.30 mm as the baseline, combined with the process requirement of "folding preparation distance" of 20.00 mm, the basic starting point is calculated as 300.30 − 20.00 = 280.30 mm. Based on this, a compensation coefficient is introduced for correction calculation, adjusting the preparation distance to 20.00 mm. 0.8 = 16.00 mm, and the final calculated action initiation position condition value is 300.30 − 16.00 = 284.30 mm. This means that when the actuator reaches the 284.30 mm coordinate, the folding command must be triggered. Subsequently, the current position is compared with the carton height, and the action termination position condition is calculated. A "forming overshoot" parameter is introduced and set to 1% of the carton height, i.e., 400.00 × 0.01 = 4.00 mm, to ensure the folding is in place and to overcome the board springback. The calculation logic for the action termination is based on the kinematic model of the folding arm. When the folding arm moves from the horizontal position to the vertical position, the projection distance of its end effector in the vertical direction should be equal to the sum of the carton height and the board thickness. However, in the linear transmission direction, the termination condition is transformed into the horizontal coordinate limit when the folding is completed. The "folding interference protection threshold" is set to 0.5 times the carton height, i.e., 200.00 mm. The calculation process is as follows: Taking the fold line position of 300.30 mm as the origin, extending along the fold direction, the maximum allowable stroke termination point is 300.30 + 200.00 = 500.30 mm. This value is compared with the limit stroke of the actual mechanism, and the smaller value is taken as the final motion termination position condition value. The above calculation is repeated for all sections to generate a set of motion position boundary parameters containing precise start coordinates (e.g., 284.30 mm) and termination coordinates (e.g., 500.30 mm), thus clarifying the effective working range of each actuator.

[0039] The mutual exclusion determination submodule, based on the action position boundary parameter group, obtains the adjacent broken line positions in the structural segment sequence and performs mutual exclusion determination, marks the boundary intervals where actions cannot be triggered simultaneously, analyzes the dataset including the constraint relationship between the start condition and the termination condition, and generates action constraint data. Based on the action position boundary parameter set, the positions of adjacent broken lines in the structural segment sequence are extracted and mutual exclusion judgment is performed. Focus is placed on the pair of adjacent segments ID-01 (broken line position 300.30 mm) and ID-02 (broken line position 600.50 mm). First, the boundary parameters calculated in the previous steps are obtained: the action operation range of ID-01 is [284.30, 500.30] (unit: mm), and the action operation range of ID-02 is similarly calculated to be [584.50, 800.50] (assuming that the start logic of ID-02 is the same as that of ID-01, with a baseline of 600.50 mm). The "mechanism physical interference radius" is defined as 120.00 mm, which represents half of the maximum lateral coverage width of the folding robotic arm when it is fully extended. The process begins by marking the boundary intervals where actions cannot be triggered simultaneously, and then performing the following analysis: The physical distance between the termination position (500.30 mm) of the ID-01 action interval and the starting position (584.50 mm) of the ID-02 action interval is calculated as 584.50 − 500.30 = 84.20 mm. Subsequently, the minimum safe distance between the two adjacent mechanisms when they are extremely close is calculated, which is twice the physical interference radius: 120.00 × 2 = 240.0 mm. The mutual exclusion logic is as follows: if the physical distance between the segments is less than the minimum safe distance, then a mutual exclusion constraint is determined to exist. In this example, the actual distance of 84.20 mm is significantly less than the safe distance of 240.00 mm, indicating that if ID-01 has not yet completed its exit action (i.e., reached the 500.30 mm position), while the ID-02 mechanism has entered its start-up preparation area (approaching 584.50 mm), the two will collide in space. Based on this determination, the dataset including the constraint relationship between start and termination conditions is analyzed to generate mandatory temporal mutual exclusion logic. The "Action Termination Complete Bit" of ID-01 is defined as the unlocking condition, and the "Action Start Allow Bit" of ID-02 is defined as the controlled object. The "safety delay displacement" is calculated, which is the safe distance of 240.00 mm minus the actual distance of 84.20 mm, resulting in 155.80 mm. This means that the triggering of ID-02 must be logically delayed or physically delayed based on the original position. Finally, the above calculation results are integrated to generate the action constraint data table shown in Table 4 below.

[0040] Table 4 Motion Constraint Data and Mutual Exclusion Parameters As shown in Table 4, this table quantifies the spatial conflict risk between adjacent sections. The physical distance between ID-01 and ID-02 is insufficient to accommodate the simultaneous operation of the two mechanisms. Therefore, a "serial locking" constraint strategy is generated, requiring that the next action can only enter the interference zone after the previous action has completely finished. This result directly corrects the underlying triggering logic of the control, ensuring the physical safety of the production process.

[0041] Specifically, such as Figure 2 , 7 As shown, the instruction generation module specifically consists of: The position acquisition submodule acquires the current position of the actuator, collects the stroke value and timestamp identifier corresponding to the current position, performs a consistency judgment on the stroke value interval, records the coordinate status of the current position within the current sampling period, and generates the current position status quantity of the actuator. The system connects in real-time to the feedback port of the servo drive via an industrial Ethernet bus, and continuously captures data on the physical state of the actuator with a standard sampling period of 2 milliseconds. At the current sampling time slice Tk, the pulse feedback signal from the high-resolution absolute encoder of the servo motor is read and converted into a linear physical travel value using internal electronic gear ratio parameters. The acquired current travel value is 285.60 mm, and a microsecond-level timestamp generated by the clock is recorded, with a value of 12005000 microseconds. To ensure the physical authenticity of the acquired data and eliminate instantaneous noise that may be caused by electromagnetic interference, a strict interval consistency judgment is performed on the travel value. The historical position data of 285.30 mm stored in the previous sampling period Tk−1 is retrieved. First, the position increment is calculated by subtracting the historical value of 285.30 mm from the current value of 285.60 mm, resulting in a displacement difference of 0.30 mm. Subsequently, the "maximum displacement threshold in a single cycle" is introduced as a judgment criterion. The threshold is set based on the product of the actuator motor's maximum physical speed of 2000.00 mm / s and the sampling period of 0.002 seconds, multiplied by a safety factor of 1.2. The calculation process is 2000.00 × 0.002 × 1.2 = 4.80 mm. The actual displacement difference of 0.30 mm is compared with the threshold of 4.80 mm. It is determined that 0.30 mm is much smaller than 4.80 mm, indicating that the displacement change is within the reasonable dynamic response range of the machine, and no abnormal abrupt changes have occurred in the data. After confirming the validity of the displacement, the instantaneous velocity is further calculated by dividing the displacement difference of 0.30 mm by the sampling time of 0.002 seconds, resulting in a current instantaneous velocity of 150.00 mm / s. The verified current travel value of 285.60 mm, the timestamp of 12005000 microseconds, and the calculated instantaneous velocity of 150.00 mm / s are structured and encapsulated to record the coordinate state of the current position within the current sampling period, ultimately generating the current position state of the actuator. This state variable not only includes static position coordinates, but also implies dynamic velocity trends, providing complete input data for subsequent precise control.

[0042] The constraint determination submodule, based on the current position state of the actuator, combines the action start boundary and action end boundary in the action constraint data, compares the current position value with multiple boundary values, determines whether it has entered the start interval or the end interval, and generates an action command trigger flag. The system receives the generated current position status value of the actuator (stroke value 285.60 mm) and retrieves the boundary parameters matching the current process step from the motion constraint dataset. The motion start boundary value corresponding to segment ID-01 is locked at 284.30 mm, and the motion termination boundary value is 500.30 mm. Combining the motion constraint data, a "boundary trigger dead zone" parameter is first introduced, set to 0.50 mm. This parameter is used to eliminate critical misjudgments caused by mechanical transmission backlash and signal fluctuations. Multiple comparison logic is executed between the current position value and multiple boundary values ​​to determine whether the system has entered the start or termination range. First, a start determination is performed: the difference between the current position 285.60 mm and the motion start boundary 284.30 mm is calculated, resulting in 1.30 mm. This difference is compared with the boundary trigger dead zone of 0.50 mm. If 1.30 mm is greater than 0.50 mm, it indicates that the mechanism has significantly crossed the start line. Simultaneously, the position trend is checked, confirming that the current position of 285.50 mm is less than the action termination boundary of 500.30 mm, indicating that the mechanism is within the valid operating range. To accurately distinguish the trigger moment, the position value of 285.30 mm from the previous cycle is compared, revealing that the previous value was also greater than the start boundary. Therefore, the current state is determined to be "action hold" rather than "action start". The "action completion" index is further calculated by dividing the difference between the current position and the start boundary (1.30 mm) by the total operating range length (500.30 − 284.30 = 216.00 mm), resulting in the formula 1.30 / 216.00 ≈ 0.006. This value indicates that the action has just begun, at 0.6% of the total folding stroke. Based on the above determination, an action command trigger flag is generated, the status flag is set to "Active-Hold", and the completion parameter 0.006 and the current range index ID-01 are written into the flag. This flag clearly indicates that the control should maintain the action output of ID-01 and adjust the control strategy according to the progress.

[0043] The sequence generation submodule calls the corresponding stage state value of the action stage chain according to the action instruction trigger flag, makes a matching judgment between the stage state and the execution order of the trigger flag, arranges the instruction codes corresponding to each stage, and generates a complete instruction sequence. Based on the action instruction trigger flag (state: Active-Hold, object: ID-01, progress: 0.006), the corresponding active phase state value "Phase-A" in the action phase chain is invoked. Phase-A defines two parallel or serial operation objects, ID-01 and ID-04, within the immediate process phase. Matching the phase state with the trigger flag execution order aims to transform abstract logical states into concrete hardware control code. First, the task list of Phase-A is scanned, identifying ID-01 as active, while ID-04 is inactive because its physical location has not yet reached the trigger boundary (referring to previous data, ID-04 is further back). The corresponding instruction codes for each phase are arranged according to the matching results. For ID-01, the "soft-start acceleration strategy" is matched based on its progress value of 0.006, and the action code in the preset instruction library is invoked. The instruction encoding rules are defined as follows: the high 8 bits represent the action type, the middle 8 bits represent the target axis number, and the low 8 bits represent the speed level. Extract the "folding action" code 0x10, the ID-01 axis number code 0x01, and the "acceleration level 1" code 0x02, and combine them to generate the instruction code 0x100102. For ID-04, generate the "axis lock maintenance" instruction, coded as 0x000200 (0x00 represents maintenance, 0x02 represents the axis number, and 0x00 represents zero speed). Sort these two instructions according to the priority logic defined in Phase-A to generate the forming instruction parameter table shown in Table 5 below.

[0044] Table 5. Molding Instruction Parameters and Encoding Mapping Table As shown in Table 5, this table details the specific instructions generated within the current control cycle. Finally, SEQ-01 and SEQ-02 are packaged together to generate a molding instruction sequence. This result demonstrates that the physical signals fed back from the position sensor have been successfully transformed into a digital instruction stream that drives the servo motor to perform specific process actions through layers of logical analysis and constraint determination, achieving a closed-loop logic from physical perception to execution control.

[0045] Please see Figure 8 A method for controlling the forming of packaging cartons, based on the aforementioned packaging carton forming control system, includes the following steps: S1: Collect the length, width, height and fold line positions of the packaging cardboard, calculate the corresponding relationship between the cardboard length and the fold line positions, divide the fold line segments and sort them to generate a structural segment sequence; S2: Input the structural segment sequence into the topological sorting algorithm to determine the sequential relationship of the multi-fold line segments in the carton forming process, arrange the bottom plate, side wall and flange folding segments in order, and generate forming sequence data; S3: Input the molding sequence data into the finite state machine, match the molding sequence with the folding and pressing state conditions in the state machine, analyze the corresponding states of multiple molding segments, and generate the action stage chain. S4: Based on the action stage chain, detect the position of the actuator, combine the broken line position and the carton height to calculate the start and end positions of the action, perform mutual exclusion judgment on adjacent broken line positions, and generate action constraint data; S5: Obtain the current position of the actuator, compare the action constraint data to determine whether the action start and action termination conditions are met, and generate a complete instruction sequence by combining the corresponding state of the action stage chain.

[0046] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A packaging carton forming control system, characterized by, The system comprises: a structure analysis module, which collects the length, width, height of the packaging paperboard and the position of the folding line, performs corresponding calculation on the length of the paperboard and the position of the folding line, divides the folding line section and sorts it, generates a structure section sequence and transmits it to a sequence determination module; the sequence determination module inputs the structure section sequence into a topological sorting algorithm, determines the sequence of the multiple folding line sections in the forming of the carton, arranges the folding sections of the bottom plate, side wall and flange in sequence, generates forming sequence data and transmits it to a state matching module; the state matching module inputs the forming sequence data into a finite state machine, matches the forming sequence with the folding and pressing state conditions in the state machine, analyzes the corresponding state of the multiple forming sections, generates an action stage chain and transmits it to an action constraint module; the action constraint module detects the position of the execution mechanism based on the action stage chain, calculates the action start and end positions in combination with the folding line position and the height of the carton, performs mutual exclusion judgment on the adjacent folding line positions, generates action constraint data and transmits it to an instruction generation module; the instruction generation module obtains the current position of the execution mechanism, compares the action constraint data to determine whether the action start and action end conditions are met, and generates a forming instruction sequence in combination with the corresponding state of the action stage chain.

2. A packaging carton forming control system according to claim 1, wherein, The structure section sequence includes the folding line section, folding line position and paperboard length, the forming sequence data includes the folding sequence of the bottom plate, the folding sequence of the side wall, the folding sequence of the flange, and the section correlation sequence, the action stage chain includes the folding stage state, the pressing stage state, the holding stage state and the release stage state, the action constraint data includes the action start position value, the action end position value, the section mutual exclusion identifier and the position effective interval, and the forming instruction sequence includes the start instruction code, the end instruction code and the stage switching instruction.

3. A packaging carton forming control system according to claim 1, wherein, The structure analysis module specifically comprises: a size collection submodule, which collects the length, width and height of the packaging paperboard and the position of the folding line, performs consistency detection on the collected multiple values, performs boundary verification on the folding line position based on the paperboard length reference value, records the calibrated coordinates of the multiple folding lines in the unfolding direction, and generates a paperboard unfolding size set; a folding line section submodule, which performs position calculation based on the paperboard length value and the calibrated coordinates of the multiple folding lines according to the paperboard unfolding size set, performs section division judgment on the relative position of the folding line in the unfolding direction, analyzes the continuous section identifier based on the distance between adjacent folding lines and the boundary position, and generates a folding line section distribution table; a section generation submodule, which calls the folding line section distribution table, performs sequence arrangement on the multiple folding line sections, sorts the sections based on the starting coordinates in the unfolding direction of the paperboard, eliminates overlapping sections and retains the section identifier, and generates a structure section sequence.

4. A packaging carton forming control system according to claim 3, wherein, The paperboard length reference value is determined by collecting multiple sets of paperboard length data, sorting all the collected paperboard length values, and analyzing the median value of the sorted paperboard length data.

5. A packaging carton forming control system according to claim 1 wherein, The sequence determination module specifically comprises: The segment sequence sub-module inputs the structural segment sequence to a topological sorting algorithm to perform shaping sequence calculation, collects multiple structural segment numbers, connection relationship identifiers, and adjacent constraint markers, performs node expansion and edge recording according to the pointing relationship between the segments, and generates segment association information; The sequence arrangement sub-module performs sequence arrangement operation according to the segment sequence constraint sequence, outputs segment numbers in the constraint resolution sequence, and performs integrity verification on the arrangement result to generate shaping sequence data. The state matching module specifically comprises:

6. A packaging carton forming control system according to claim 1 wherein, The sequence receiving sub-module collects segment numbers, folding identifiers, and sequence indexes corresponding to each shaping sequence item based on the shaping sequence data as the starting input, calculates the adjacent index difference to determine whether there is a sequence breakpoint, and generates a shaping sequence index set; The state judgment sub-module executes state condition comparison for the segment numbers corresponding to each sequence index according to the folding state condition table and the pressing state condition table pre-stored in the finite state machine, judges the state attribution of the current sequence item, and generates an action state mapping sequence; The phase chain generation sub-module executes continuity judgment for the state identifiers of adjacent sequence items based on the action state mapping sequence, detects state identifier change points and records phase switching positions, reorganizes multiple state paragraphs according to sequence indexes, and generates an action phase chain. The finite state machine is composed of a state set, an input condition set, a state transition relationship, an initial state, and a judgment state set.

7. A packaging carton forming control system according to claim 6, wherein, The action constraint module specifically comprises:

8. A packaging carton forming control system according to claim 1 wherein, The position detection sub-module plans an expected execution mechanism position sequence according to the action phase chain, collects the stroke value and phase identifier value corresponding to the current position, performs consistency judgment on the phase identifier, records the coordinate sequence corresponding to the current position in the multiple action phases, and generates an execution mechanism position sequence; The boundary calculation sub-module obtains the fold line position and the carton height according to the execution mechanism position sequence, performs comparison and calculation of the action start position condition value for the current position and the fold line position, performs comparison and calculation of the action termination position condition for the current position and the carton height, and generates an action position boundary parameter group; The mutual exclusion judgment sub-module obtains adjacent fold line positions in the structural segment sequence and performs mutual exclusion judgment based on the action position boundary parameter group, marks the boundary interval of actions that cannot be triggered at the same time, analyzes a data set including the start condition and termination condition constraint relationship, and generates action constraint data. The instruction generation module specifically comprises:

9. A packaging carton forming control system according to claim 1 wherein, The position acquisition sub-module acquires the current position of the execution mechanism, collects the stroke value and timestamp identifier corresponding to the current position, performs interval consistency judgment on the stroke value, records the coordinate state corresponding to the current position in the current sampling period, and generates an execution mechanism current position state quantity. ​ The constraint judgment sub-module, based on the current position state quantity of the execution mechanism, combines the action start boundary and the action termination boundary in the action constraint data, performs comparison on the current position value and the multi-boundary value, judges whether to enter the start interval or the termination interval, and generates an action instruction trigger flag quantity; The sequence generation sub-module, according to the action instruction trigger flag quantity, calls the corresponding stage state value of the action stage chain, performs sequential matching judgment on the stage state and the trigger flag, arranges the instruction code corresponding to each stage, and generates a forming instruction sequence.

10. A method of controlling the forming of a packaging carton, characterized by, The packaging carton forming control system according to any one of claims 1-9 is executed, Comprising the following steps: S1: Collecting the length, width and height of the packaging paperboard and the fold line position, calculating the length of the paperboard and the fold line position, dividing the fold line section and sorting, and generating a structure section sequence; S2: Inputting the structure section sequence into a topological sorting algorithm, judging the sequence relationship of the multi-fold line section in the carton forming, arranging the bottom plate, side wall and flanging folding section in sequence, and generating forming sequence data; S3: Inputting the forming sequence data into a finite state machine, matching the forming sequence and the folding and pressing state condition in the state machine, analyzing the corresponding state of the multi-forming section, and generating an action stage chain; S4: Detecting the current position of the execution mechanism based on the action stage chain, calculating the action start and termination positions in combination with the fold line position and the carton height, performing mutual exclusion judgment on the adjacent fold line positions, and generating action constraint data; S5: Acquiring the current position of the execution mechanism, comparing the action constraint data to judge whether the action start and termination conditions are met, and generating a forming instruction sequence in combination with the corresponding state of the action stage chain.

Images