Linear track control device and linear track system

Abstract

The linear track control device (10A) has a drive control section (C1) that controls a current for generating a driving force between a rear carrier (21P) that moves along a conveyance path on which a stator (51) is arranged and the stator, and controls a current for generating a driving force between a preceding carrier (21Q) that is arranged in a traveling direction of the rear carrier and the stator, and an instruction generation section (11A) that generates a time-series instruction that specifies at least one of a position, a speed, and an acceleration of the rear carrier in time series and outputs the time-series instruction to the drive control section. The instruction generation section has a correction coefficient decision section (19A) that decides a correction coefficient for correcting any of the position, the speed, and the acceleration of the rear carrier, based on a relative value of positions on the conveyance path of the rear carrier and the preceding carrier, that is, a gap, and a time-series instruction generation section (18) that generates the time-series instruction based on a target value of the position or the speed of the rear carrier and the correction coefficient.

Description

Technical Field

[0001] This invention relates to a linear track control device and a linear track system for controlling linear tracks. Background Technology

[0002] A linear track system moves multiple movers (hereinafter referred to as carriers) equipped with permanent magnets along a transport path of a long stator equipped with electromagnetic coils, thereby realizing a logistics system for objects placed on the carriers. In this linear track system, a higher-level control system can independently command the position, velocity, acceleration, etc., of each carrier. Each carrier is independently controlled by the higher-level control system, but due to the setting of the given commands, accidental interference between carriers (such as contact, collision, etc.) may occur.

[0003] The linear track system described in Patent Document 1 determines a limiting speed to prevent interference between the carriers along the transport path based on the relative distance (gap) between the preceding and following carriers and the moving speed of the preceding carrier.

[0004] Patent Document 1: Japanese Patent Application Publication No. 2009-187239 Summary of the Invention

[0005] However, the technology in the aforementioned Patent Document 1 has the following problem: after decelerating to avoid interference between the carriers, the acceleration process for restoring the action before deceleration becomes complicated.

[0006] The present invention was made in view of the above circumstances, and its object is to obtain a linear track control device that can avoid interference between carriers and easily restore the motion before deceleration.

[0007] To address the aforementioned issues and achieve the objective, the linear track control device of the present invention includes a drive control unit that controls the current used to generate a driving force between a first carrier moving along a transport path where a stator is arranged and the stator, and controls the current used to generate a driving force between a second carrier arranged in the travel direction of the first carrier and the stator. Furthermore, the linear track control device of the present invention includes a command generation unit that generates and outputs a timing command, which specifies at least one of the position, velocity, and acceleration of the first carrier according to a time sequence, to the drive control unit. The command generation unit includes: a target setting unit that sets a target value representing the position or velocity of the first carrier; a correction coefficient determining unit that determines a correction coefficient for correcting any one of the position, velocity, and acceleration of the first carrier based on the relative value (gap) of position information representing the position of the first carrier on the transport path and position information representing the position of the second carrier on the transport path; and a timing command generation unit that generates a timing command based on the target value and the correction coefficient.

[0008] The effects of the invention

[0009] The linear track control device involved in this invention has the following effect: it can avoid interference between the carriers and easily restore the operation before deceleration. Attached Figure Description

[0010] Figure 1 This is a diagram showing the structure of a linear track system having the linear track control device according to Embodiment 1.

[0011] Figure 2 This is a diagram showing the internal structure of the linear track control device and conveying unit involved in Embodiment 1.

[0012] Figure 3 This is a flowchart showing the processing sequence of the instructions generated by the instruction generation unit of the linear track control device according to Embodiment 1.

[0013] Figure 4 This is a diagram used to explain the correction coefficients determined by the linear track control device according to Embodiment 1.

[0014] Figure 5 This is a diagram representing an example of a proportional timing instruction.

[0015] Figure 6 This is a diagram illustrating an example of timing commands generated by the linear track control device according to Embodiment 1.

[0016] Figure 7 This is a diagram illustrating an example of a corrected timing command for the linear track control device according to Embodiment 1 when it resumes its original operation after the rear carrier decelerates.

[0017] Figure 8 This is a diagram showing the structure of a linear track system having the linear track control device according to Embodiment 2.

[0018] Figure 9 This is a diagram showing the internal structure of the linear track control device and conveying unit involved in Embodiment 2.

[0019] Figure 10 This is a diagram used to explain the correction coefficients determined by the linear track control device according to Embodiment 2.

[0020] Figure 11 This is a diagram illustrating an example of timing commands generated by the linear track control device according to Embodiment 2.

[0021] Figure 12This is a diagram showing the structure of a linear track system having the linear track control device according to Embodiment 3.

[0022] Figure 13 This is a diagram showing the internal structure of the linear track control device and conveying unit involved in Embodiment 3.

[0023] Figure 14 This is a diagram used to explain the correction coefficients determined by the linear track control device according to Embodiment 3.

[0024] Figure 15 This is a diagram illustrating an example of timing commands generated by the linear track control device according to Embodiment 3.

[0025] Figure 16 This is a diagram showing the structure of a linear track system having the linear track control device according to Embodiment 4.

[0026] Figure 17 This is a diagram showing the internal structure of the linear track control device and conveying unit involved in Embodiment 4.

[0027] Figure 18 This is a diagram used to explain the correction period corresponding to the learning data obtained by the linear track control device according to Embodiment 4.

[0028] Figure 19 This is a flowchart illustrating the processing sequence of the learning process performed by the linear track control device according to Embodiment 4.

[0029] Figure 20 This is a flowchart illustrating the processing sequence of the inference process performed by the linear track control device according to Embodiment 4.

[0030] Figure 21 This is a diagram illustrating an example of the hardware structure of the linear track control device according to Embodiment 1. Detailed Implementation

[0031] The linear track control device and linear track system according to the embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0032] Implementation method 1.

[0033] Figure 1 This is a diagram showing the structure of a linear track system having the linear track control device according to Embodiment 1. Figure 1 The linear track system 1A shown, for example, has a linear track control device 10A, a conveying unit 50, a conveying path 55, and multiple carriers (movers) 21.

[0034] The linear track control device 10A controls the current used to drive each carrier 21 according to user settings. The conveying unit 50 drives the carrier 21 using the current from the linear track control device 10A. The conveying path 55 is formed by combining multiple conveying units 50. Each carrier 21 moves along the conveying unit 50 on the conveying path 55. Furthermore, in... Figure 1 The image shows a case where the shape of the conveying path 55 is a loop composed of a combination of straight lines and curves, but the shape of the conveying path 55 is not limited to a loop.

[0035] The linear track control device 10A includes an instruction generation unit 11A and drive control units C1 to Cm (where m is a natural number). The instruction generation unit 11A generates a timing instruction (corrected timing instruction) 30 that corrects at least one of the timing instructions specifying the position, velocity, and acceleration of each carrier 21 according to a time sequence. The drive control units C1 to Cm control the current used to generate driving force for each carrier 21 based on the timing instruction 30 and position information indicating the position of each carrier 21 on the transport path 55. Furthermore, in the following description, the drive control units C1 to Cm are sometimes referred to as drive control units C unless there is no need to distinguish between them.

[0036] The drive control units C1 to Cm control the conveying unit 50 by controlling the current supplied to it. For example, one drive control unit C controls the current supplied to one conveying unit 50.

[0037] Each conveying unit 50 is capable of driving multiple carriers 21. Furthermore, in the following description, the carrier preceding the others on the conveying path 55 is sometimes referred to as the preceding carrier 21Q, and the carrier following the preceding carrier 21Q on the conveying path 55 is sometimes referred to as the following carrier 21P. If we observe the following carrier 21P, the preceding carrier 21Q is the carrier 21 preceding the following carrier 21P on the conveying path 55 in the direction of travel along the conveying path 55. If we observe the preceding carrier 21Q, the following carrier 21P is the carrier 21 following the preceding carrier 21Q on the direction of travel along the conveying path 55. The following carrier 21P is the first carrier, and the preceding carrier 21Q is the second carrier. There are no other carriers 21 between the preceding carrier 21Q and the following carrier 21P.

[0038] An example of the hardware structure of the instruction generation unit 11A is a programmable logic controller (PLC), and an example of the hardware structure of the drive control unit C1 is a linear motor driver.

[0039] Next, refer to Figure 2 An example of the structure and operation of the linear track control device 10A is given. Figure 2 This is a diagram showing the internal structure of the linear track control device and conveying unit involved in Embodiment 1. Figure 2 The section explains the drive control unit C1's control of the rear carrier 21P. Therefore, in... Figure 2 The drive control units C2 to Cm are omitted from the diagram. As described above, the linear track system 1A also performs drive control on a carrier 21, which is different from the rear carrier 21P, and has multiple drive control units C (not shown) and a transport unit 50. When the drive control unit C1 performs drive control on the rear carrier 21P, the preceding carrier 21Q can be driven by any of the drive control units C1 to Cm. Figure 2 In the illustration, the conveying unit 50 with a stator 51 that generates driving force is omitted between the carrier 21Q and the preceding carrier.

[0040] The linear track control device 10A includes a command generation unit 11A, a drive control unit C, and a subtractor 16. The command generation unit 11A includes a target setting unit 17, a correction coefficient determination unit 19A, and a timing command generation unit 18.

[0041] The target setting unit 17 determines the target position 34, which is a target value relative to the position of one of the plurality of carriers 21, namely the rear carrier 21P, before correction. The target setting unit 17 sends the target position 34 to the timing command generation unit 18.

[0042] The correction coefficient determination unit 19A determines a correction coefficient k for correcting any one of the position, velocity, and acceleration of the rear carrier 21P based on the relative position (relative value) or gap g of the rear carrier 21P and the preceding carrier 21Q, which is different from the rear carrier 21P. The correction coefficient determination unit 19A sends the correction coefficient k to the timing command generation unit 18.

[0043] The timing command generation unit 18 determines a timing command 30 that specifies at least one of the position, velocity, and acceleration of the rear carrier 21P according to a time sequence, based on the target position 34 and the correction coefficient k. For example, the timing command generation unit 18 calculates the velocity command for the rear carrier 21P according to a time sequence based on the target position 34, corrects the velocity command using the correction coefficient k, and thereby generates the timing command 30. In this case, the correction coefficient k is a coefficient used to correct the velocity command. Alternatively, the timing command 30 may be, for example, a command that specifies the velocity of the rear carrier 21P according to a time sequence. Furthermore, when the timing command generation unit 18 corrects the velocity command using the correction coefficient k, it can also generate a timing command 30 that specifies the position or acceleration of the rear carrier 21P according to a time sequence.

[0044] Alternatively, the timing command generation unit 18 may calculate a position command for the rearward carrier 21P based on the target position 34, and correct the position command using a correction coefficient k, thereby generating a timing command 30. In this case, the correction coefficient k is a coefficient used to correct the position command. Furthermore, the timing command 30 is, for example, a position command that specifies the position of the rearward carrier 21P according to the time sequence. Moreover, when the timing command generation unit 18 corrects the position command using the correction coefficient k, it may also generate a timing command 30 that specifies the velocity or acceleration of the rearward carrier 21P according to the time sequence. The timing command generation unit 18 inputs the timing command 30 to the drive control unit C1.

[0045] Here, the relationship between the rear carrier 21P and the preceding carrier 21Q will be explained. The rear carrier 21P is one of the carriers 21 whose position, velocity, and acceleration are the objects of correction to avoid interference between the carriers 21. The preceding carrier 21Q is one of the carriers 21 located in front of the rear carrier 21P in the direction of travel. In Embodiment 1, the correction of the position, velocity, and acceleration of one of these two carriers 21, namely the rear carrier 21P, will be explained. Furthermore, in a linear track system 1A with three or more carriers 21, it is assumed that other carriers 21 are located in front of the preceding carrier 21Q in its direction of travel. In this case, if viewed from the other carriers 21, the preceding carrier 21Q becomes the rear carrier 21P.

[0046] Furthermore, when there are two carriers 21, and the transport path 55 is circular, the rear carrier 21P is located ahead of the preceding carrier 21Q in the direction of travel. In this case, one of the two carriers 21 is both the preceding carrier 21Q and the rear carrier 21P relative to the other. Conversely, the other of the two carriers 21 is both the preceding carrier 21Q and the rear carrier 21P relative to the first.

[0047] Furthermore, regarding the structure of the branching transport path 55, there is a case where multiple carriers 21 are considered as preceding carriers 21Q relative to one following carrier 21P. In this case, the linear track system 1A can further reduce the possibility of interference between carriers 21 in the overall linear track system 1A by applying the control described in Embodiment 1 to multiple carriers 21 in parallel or to each other. That is, the linear track system 1A controls the following carriers 21P separately for each preceding carrier 21Q, thereby further reducing the possibility of interference between carriers 21.

[0048] The relative position of the rear carrier 21P with respect to the preceding carrier 21Q, i.e., the gap g, is the difference between the position of the rear carrier 21P and the position of the preceding carrier 21Q (the difference in travel distance on the transport path 55). If the gap g continues to decrease, the carriers 21 will soon interfere with each other (contact or collide). Each carrier 21 has a physical size, and the center of its position coordinates is, for example, measured near the center of the carrier 21 as 0 (the unit is, for example, millimeters). Therefore, in the description of Embodiment 1, it should be noted that interference between the carriers 21 may occur before the value of the gap g becomes 0.

[0049] Furthermore, unlike the description in Embodiment 1, it is possible to configure the carriers 21 such that when the position of each carrier 21 is set at the physical end of the carrier (or, further outward than the end), the carriers 21 will not interfere with each other until the value of the gap g becomes 0 or less than 0. In the case described above, the correction coefficient determination unit 19A also considers the offset of the gap g to determine the correction coefficient k.

[0050] The drive control unit C1 includes a motion control unit 12 and a current control unit 13. The transport unit 50 includes a stator 51 and a position detector 52. The stator 51 receives current controlled by the drive control unit C1 to generate driving force for the rear carrier 21P.

[0051] Position detector 52 detects the position of carrier 21, such as rear carrier 21P. That is, position detector 52 detects the relative position of carrier 21 and stator 51. Furthermore, advance position information 40Q is also detected by any position detector 52, similar to rear position information 40P. Position detector 52 sends the rear position information 40P, representing the position of rear carrier 21P on the transport path 55, to motion control unit 12 and subtractor 16.

[0052] The motion control unit 12 generates a driving force command 31, which specifies, according to a time sequence, the driving force generated by the rear carrier 21P to make the rear position information 40P follow any of the position, velocity, and acceleration indicated by the time sequence command 30. That is, the motion control unit 12 generates the time-series driving force command 31 in a manner that matches any of the position, velocity, and acceleration corresponding to the time sequence command 30 with any of the position, velocity, and acceleration of the rear carrier 21P corresponding to the rear position information 40P. The motion control unit 12 sends the driving force command 31 to the current control unit 13.

[0053] The current control unit 13 controls the current used to generate a driving force for the rear carrier 21P based on the driving force command 31. The motion control unit 12 and the current control unit 13 use techniques such as feedback control to drive the carrier 21 so that physical quantities such as the position of the carrier 21 follow the target value. That is, the drive control unit C1 controls the current supplied to the stator 51 to generate a driving force between the rear carrier 21P and the stator 51 through the motion control unit 12 and the current control unit 13 to move the rear carrier 21P in the traveling direction.

[0054] In addition, the motion control unit 12 and the current control unit 13 may use control methods different from feedback control, such as feedforward control, or they may combine feedforward control and feedback control.

[0055] Subtractor 16 calculates the gap g based on the preceding position information 40Q and the following position information 40P, which represent the position of the preceding carrier 21Q on the transport path 55. Specifically, subtractor 16 subtracts the coordinates on the transport path 55 represented by the following position information 40P from the coordinates on the transport path 55 represented by the preceding position information 40Q, thereby calculating the gap g along the transport path 55. Subtractor 16 sends the gap g as the result of the subtraction to the correction coefficient determination unit 19A.

[0056] The rear carrier 21P receives driving force from the conveying unit 50 to carry out production activities (such as conveying items, gripping workpieces, assembling parts, packaging products, processing materials, etc.) implemented using the linear track system 1A.

[0057] An example of the hardware structure of stator 51 is an electromagnet, an example of the hardware structure of carrier 21 is a permanent magnet, and an example of the hardware structure of position detector 52 is a linear encoder.

[0058] Alternatively, instead of setting the target position 34, the target setting unit 17 can output the target value, i.e., the target speed, relative to the rear carrier 21P before correction, to the timing command generation unit 18. In this case, the timing command generation unit 18 generates a timing command that corrects the speed command corresponding to the target speed using a correction coefficient k and sends it to the motion control unit 12. The motion control unit 12 calculates the speed of the rear carrier 21P by differentiating the rear position information 40P and performs feedback control based on the speed of the rear carrier 21P. That is, the motion control unit 12 generates a driving force command 31 that specifies the driving force generated for the rear carrier 21P according to the time sequence, in a manner that the speed of the rear carrier 21P follows the speed command represented by the timing command 30.

[0059] The correction factor k can be used to correct any of the position, velocity, and acceleration commands. The following explanation will primarily focus on the case where the correction factor k is used to correct the velocity command.

[0060] Next, refer to Figure 3 The operation of the target setting unit 17, the correction coefficient determination unit 19A, and the timing instruction generation unit 18 will be explained in detail. Figure 3 This is a flowchart illustrating the processing sequence of the instructions generated by the command generation unit of the linear track control device according to Embodiment 1. Figure 3 The operation flow of the target setting unit 17, the correction coefficient determination unit 19A, and the timing instruction generation unit 18 will be explained.

[0061] First, the target setting unit 17 determines the target position 34 for the rear carrier 21P that is the object to be controlled (step S11). In Embodiment 1, the target setting unit 17 determines a value representing the final position of the rear carrier 21P as the target position 34. As described above, the method of controlling an object to move from an initial position to a final position by specifying the final position is called positioning control or PTP (Point To Point) control. The target setting unit 17 sends the set target position 34 to the timing command generation unit 18.

[0062] Subtractor 16 calculates the gap g, which is the position difference between the rear carrier 21P and the preceding carrier 21Q, and sends it to the correction coefficient determination unit 19A. Thus, the correction coefficient determination unit 19A obtains the gap g (step S12). Furthermore, the linear orbital system 1A sometimes needs to determine the preceding carrier 21Q from among multiple carriers 21 before processing in step S12.

[0063] The preceding carrier 21Q is a carrier located in the direction of travel of the following carrier 21P. The linear track system 1A can determine the direction of travel of the following carrier 21P based on the sign of its velocity. The sign of the velocity is obtained through timing commands or the time derivative of timing commands. Among all carriers ahead of the following carrier 21P in its direction of travel, the linear track system 1A selects the carrier 21 closest to the following carrier 21P as the preceding carrier 21Q. That is, on the transport path 55, the linear track system 1A selects the carrier 21 closest to the following carrier 21P from among the carriers 21 ahead of the following carrier 21P in its direction of travel as the preceding carrier 21Q.

[0064] Furthermore, during the operation of the linear track system 1A, the lead carrier 21Q may be removed for some reason. For example, if the carrier 21 is removed from the transport path 55 by other external devices, the lead carrier 21Q may sometimes separate from the path of the rear carrier 21P at a branch of the transport path 55. In these cases, the linear track system 1A determines a new lead carrier 21Q each time. Furthermore, if there is no lead carrier 21Q corresponding to the rear carrier 21P, the linear track system 1A cannot calculate the gap g in step S12, so a large value can be virtually substituted into the gap g. Specifically, the linear track system 1A sets the gap g to a value larger than the setting value (the setting value R described later) used in the calculation of the correction coefficient k.

[0065] In the description of Embodiment 1, the position information of each carrier 21 used for calculating the gap g is the actual position of the carrier 21 detected by the position detector 52. In synchronous motors and the like, the commanded position and the actual position are made approximately consistent through feedback control, so the linear track system 1A can use the commanded position instead of the position detected by the position detector 52. That is, the linear track control device 10A can generate the rear position information 40P and the preceding position information 40Q based on the timing command 30.

[0066] The correction coefficient determination unit 19A determines the correction coefficient k based on the gap g (step S13). Here, the correction coefficient k will be explained. Figure 4 This is a diagram used to explain the correction coefficients determined by the linear track control device according to Embodiment 1.

[0067] Figure 4 The horizontal axis of the graph shown is the gap g, and the vertical axis is the correction coefficient k. Figure 4The correction function F1 of the correction coefficient determination unit 19A is shown in the figure. The correction coefficient determination unit 19A of Embodiment 1 determines the correction coefficient k based on the gap g at each time by the correction function F1 shown in the following equation (1).

[0068] Formula 1

[0069]

[0070] The correction function in the second row of the correction function F1 expressed by equation (1) is set to a range where the gap g is larger than the set value Z and smaller than the set value R. Here, the set value Z and the set value R are positive constants set by the designer of the linear track system 1A. The correction coefficient determination unit 19A determines a correction coefficient k that is proportional to the difference between the gap g and the set value Z. Alternatively, the correction coefficient determination unit 19A may also set a correction coefficient k that is proportional to the value obtained by performing a specific calculation on the difference between the gap g and the set value Z.

[0071] In Implementation 1, the correction coefficient k that is proportional to the difference between the gap g and the set value Z, and the correction coefficient k that is proportional to the value obtained by performing a specific calculation on the difference between the gap g and the set value Z, are called the correction coefficient k that is proportional to the gap g.

[0072] The setting value Z is a minimum gap g that is desired to prevent interference between the carriers 21. Additionally, the setting value R is a gap g that is determined to be sufficient to prevent interference between the carriers 21 even without timing command correction. That is, if the gap g of the rear carrier 21P is lower than the setting value R, it decelerates and stops upon reaching the setting value Z.

[0073] First, the design method for the setpoint Z will be explained. The setpoint Z is determined based on the center coordinates of the carrier 21, taking into account the ends of the carrier 21 or components attached to the carrier 21, the workpiece held by the carrier 21, etc. Furthermore, the setpoint Z can be determined by taking into account the overshoot, vibration magnitude, and prevention of foreign object clamping generated in the machine by the control of the drive control unit C1.

[0074] Next, the design method for the setpoint R will be explained. The setpoint R is the size of the gap g at which the rear carrier 21P begins to decelerate relative to the preceding carrier 21Q to prevent interference between the carriers 21. To avoid interference between the carriers 21, a sufficiently large setpoint R is set considering the maximum acceleration and thrust of the rear carrier 21P. However, if the setpoint R is too large, it will cause the action of maintaining an excessively large gap g to avoid interference, thus reducing the efficiency of the linear track system 1A, which needs to be noted.

[0075] In addition, the setting value Z and setting value R can take into account the selection method of the origin of each carrier 21 position, the physical size of the carrier 21 from the origin, etc., and set different values ​​for each carrier 21 in accordance with the direction of travel of the rear carrier 21P.

[0076] As shown in Equation (1), the correction coefficient determination unit 19A of Embodiment 1 determines the correction coefficient k in such a way that the correction coefficient k takes the value from 0 to 1. However, Embodiment 1 is not limited to the determination of the correction coefficient k as described above. For example, when the gap g is less than or equal to the set value Z, the correction coefficient determination unit 19A may also extend the straight part (k = (g - Z) / (R - Z)) to determine the correction coefficient k in such a way that the correction coefficient k becomes negative.

[0077] The correction coefficient determination unit 19A determines the correction coefficient k by making it negative. This allows for the automatic generation of a timing command to separate from the preceding carrier 21Q even when the gap g is lower than the set value Z due to factors such as feedback control delays and vibrations. After the rear carrier 21P decelerates and stops to avoid interference, the unit ensures that the gap g is automatically maintained at or above the set value Z. Therefore, when the linear track system 1A performs the operation of maintaining a specific distance between the carriers 21, the correction coefficient k becomes an appropriate value.

[0078] The correction coefficient determination unit 19A sends the determined correction coefficient k to the timing command generation unit 18. Based on the target position 34 and the correction coefficient k, the timing command generation unit 18 determines at least one timing command 30 that specifies the position, velocity, and acceleration of the rear carrier 21P according to a time sequence (step S14). That is, the timing command generation unit 18 corrects any one of the position command, velocity command, and acceleration command corresponding to the target position 34 using the correction coefficient k, thereby generating the timing command 30. The timing command generation unit 18 inputs the timing command 30 to the drive control unit C1.

[0079] The linear track control device 10A determines whether the drive control process for the rearward carrier 21P has ended (step S15). If the drive control process for the rearward carrier 21P has not ended (step S15, No), the target setting unit 17 determines whether the target position 34 of the rearward carrier 21P has changed (step S16).

[0080] If the target position 34 of the rear carrier 21P remains unchanged (step S16, No), the linear track control device 10A returns to the process of step S12 and executes the processes of steps S12 to S15. On the other hand, if the target position 34 of the rear carrier 21P has changed (step S16, Yes), the linear track control device 10A returns to the process of step S11 and executes the processes of steps S11 to S15.

[0081] When the drive control process of the rearward carrier 21P is completed (step S15, Yes), the linear track control device 10A terminates the drive control of the rearward carrier 21P.

[0082] Here, for the purpose of explaining the process of step S14, refer to... Figure 5 and Figure 6 The operation of the timing instruction generation unit 18 will be explained. First, using... Figure 5 The timing instructions for the comparative example will be explained, and then an example of the timing instructions involved in Implementation 1 will be explained.

[0083] Figure 5 This is a diagram representing an example of a proportional timing instruction. Figure 5 and what follows Figure 6 , 7 The horizontal axis of graphs 11, 15, and 18 represents time, and the scale of the horizontal axis is consistent across all graphs. Figure 5 The first paragraph indicates that the final position (target position) of the rear carrier 21P as the target is represented by position P.

[0084] Figure 5 The waveform W1 shown in the first segment is the waveform of the timing instruction generated according to the time sequence to move the rear carrier 21P from the initial position, i.e., position 0, to position P (the waveform of the instruction that specifies the position according to the time sequence). Figure 5 The timing instructions shown in the comparative example are the timing instructions before correction in Implementation 1.

[0085] Figure 5 The waveform W2 shown in the second segment is the waveform of the speed of the timing command generated according to the time sequence so that the rear carrier 21P moves from position 0 to position P (the waveform of the command that specifies the speed according to the time sequence).

[0086] Figure 5 The waveform W3 shown in the third segment is the acceleration waveform of the timing command generated according to the time sequence to move the rear carrier 21P from position 0 to position P (the waveform of the command that specifies the acceleration according to the time sequence).

[0087] Figure 5 The waveform W1 shown in the first segment is the value of the position of the moving target of the carrier 21 represented by the position command generated according to the time sequence. The waveform is displayed according to the time sequence with the horizontal axis as time, and the position value of each position command is called the position of the timing command. Figure 5The waveform W2 shown in the second paragraph is a waveform that transforms the position value represented by the position of the timing instruction into a velocity value by differentiating the position of the timing instruction in time. The waveform is displayed in time sequence with the horizontal axis as the time value of the carrier 21. These velocity values ​​are called the velocity of the timing instruction. Figure 5 The waveform W3 shown in the third section is a waveform that transforms the velocity value represented by the timing command's velocity into an acceleration value by differentiating the velocity of the timing command over time. The waveform is displayed as a time sequence with the horizontal axis representing time, and these acceleration values ​​are called the acceleration of the timing command. That is, waveform W2 is obtained by differentiating waveform W1, and waveform W3 is obtained by differentiating waveform W2. Furthermore, waveform W2 is obtained by integrating waveform W3, and waveform W1 is obtained by integrating waveform W2.

[0088] Furthermore, the position, velocity, and acceleration of the timing instructions are defined as the values ​​of the velocity of the carrier 21 represented by the velocity instructions generated according to the time sequence. The velocity values ​​of each velocity instruction in the waveform W2, displayed according to the time sequence with the horizontal axis as time, are called the velocity of the timing instructions. The position of the carrier 21, obtained by integrating the velocity of the timing instructions, is called the position of the timing instructions. The acceleration value of the carrier 21, obtained by differentiating the velocity of the timing instructions, is called the acceleration of the timing instructions. Additionally, the position, velocity, and acceleration of the timing instructions are defined as the values ​​of the acceleration of the carrier 21 represented by the acceleration instructions generated according to the time sequence. The acceleration values ​​of each acceleration instruction in the waveform W3, displayed according to the time sequence with the horizontal axis as time, are called the acceleration of the timing instructions. The velocity value of the carrier 21, obtained by integrating the acceleration of the timing instructions, is called the velocity of the timing instructions. The position of the carrier 21, obtained by integrating the velocity of the timing instructions, is called the position of the timing instructions.

[0089] In positioning systems using linear orbits, the maximum thrust and velocity of the carrier are limited. Therefore, the simplest timing command to achieve a short-duration positioning action is... Figure 5 The timing instructions are shown. That is, in a positioning system, such as... Figure 5 As shown in paragraph 3, timing instructions are determined by taking the maximum, 0, and minimum values ​​of acceleration, which makes it easy to determine the timing instructions. Specifically, as... Figure 5As shown in the third paragraph, timing instructions are generated in the following manner: acceleration 0 from time 0 to time Ta, acceleration Aa (where 0 < Aa) from time Ta to time Tb, acceleration 0 from time Tb to time Tc, acceleration Ad (where Ad < 0) from time Tc to time Td, and acceleration 0 after time Td.

[0090] as a result, Figure 5 The speed of the second sequence of instructions monotonically increases from time Ta to time Tb, maintains the maximum speed Vmax from time Tb to time Tc, monotonically decreases from time Tc to time Td, and becomes 0 after time Td. Finally, as... Figure 5 As shown in the first paragraph, the position of the timing instruction is the initial position 0 until time Ta, and changes from position 0 to position P from time Ta to time Td. After time Td, it stays at position P.

[0091] Figure 6 This diagram illustrates an example of timing commands generated by the linear track control device according to Embodiment 1. The timing command generation unit 18 uses... Figure 4 The timing instructions are generated using the correction function F1 shown.

[0092] When the timing instruction generation unit 18 generates a waveform relative to, for example, the speed at which timing instructions are generated using a correction factor k, it generates a waveform that is... Figure 5 The second paragraph shows the timing instructions before correction. When the speed of the original timing instruction is set to 100%, the timing instructions are changed proportionally by the correction factor k. The timing instructions generated under these conditions are called the corrected timing instructions. This timing instruction correction method is known in industrial equipment, primarily servo motors and robots, as a function used to change speed during positioning, or "override function (or speed override function)".

[0093] In the speed override function, when it is particularly desirable to suppress the speed of the mechanical equipment compared to normal operation, a value from 0 to 1 is selected as the correction factor (also known as the override factor) k. Specifically, when the correction factor k is 0, the speed of the timing command becomes 0, and as a result, the carrier 21 stops. In the speed override function, if the correction factor k is then changed to a value greater than 0, timing commands are generated again until the final target position 34 is reached.

[0094] Figure 6The waveform W11 shown in the first segment illustrates the time sequence of the position difference g between the rear carrier 21P and the preceding carrier 21Q. The timing instruction generation unit 18 generates timing instructions 30 such that the gap g represented by the waveform W11 is not less than a set value Z.

[0095] exist Figure 6 The second segment, represented by the dashed line, shows waveform W12b, which is intended to make... Figure 5 The waveform W1 represents the position of the timing command before correction, generated by the rear carrier 21P moving from position 0 to position P, indicated by the solid line. Figure 6 The image shows the state where the advance carrier 21Q is stopped at position Pf, between position 0 and position P. Figure 6 The second segment, waveform W13, represented by a dotted line, is the waveform indicating the position of the timing instruction of the preceding carrier 21Q. Figure 6 The second segment, represented by the solid line waveform W14a, shows the waveform of the position of the corrected timing instruction of the rear carrier 21P generated by the timing instruction generation unit 18 according to Embodiment 1.

[0096] exist Figure 6 The third segment, represented by the dashed line, shows waveform W15b, which is... Figure 5 The waveform W2 shows the speed of the timing command before correction, represented by the solid line. Figure 6 The third segment, represented by the solid line waveform W16a, shows the waveform of the speed of the corrected timing command of the rear carrier 21P generated by the timing command generation unit 18 according to Embodiment 1.

[0097] exist Figure 6 The fourth segment, represented by the dashed line, shows waveform W17b, which is... Figure 5 The waveform W3 shows the acceleration waveform of the timing command before correction, represented by a solid line. Figure 6 The fourth segment, represented by the solid line waveform W18a, shows the waveform of the acceleration of the corrected timing command of the rear carrier 21P generated by the timing command generation unit 18 according to Embodiment 1.

[0098] As described above, the timing instruction generation unit 18, in order to prevent interference between the carriers 21, will... Figure 6 The waveforms W12b, W15b, and W17b were corrected to waveforms W14a, W16a, and W18a. Figure 6 The fifth segment, represented by the solid line W19, shows the waveform of the timing data for the correction coefficient k.

[0099] exist Figure 6 The second paragraph, and Figure 5 The first segment will similarly become the final location of the target, represented by position P. Figure 6 In the following situation, assuming if Figure 6 The second segment, represented by the dashed line, is the timing instruction (waveform W12b) before correction. In actual control, the rear carrier 21P will reach position Pf before reaching position P, and the preceding carrier 21Q and the rear carrier 21P will interfere with each other.

[0100] Furthermore, the timing instruction generation unit 18 only needs to generate at least one of the timing instructions 30 for waveforms W14a, W16a, and W18a based on the correction coefficient k. For example, the timing instruction generation unit 18 generates the timing instruction 30 for waveform W16a. The timing instruction generation unit 18 sends the generated timing instruction 30 to the motion control unit 12.

[0101] Here, for Figure 6 The change process of the waveform shown is explained. During the period from time 0 to time Ta, the position of the rear carrier 21P is position 0, the position of the preceding carrier 21Q is position Pf, and the gap g between the rear carrier 21P and the preceding carrier 21Q becomes a constant value (Pf).

[0102] After time Ta, the gap g gradually decreases as the position of the rear carrier 21P changes. After time Tc1, the size of the gap g is lower than the set value R for switching the formula determining the correction coefficient k. Therefore, after time Tc1, the correction coefficient k gradually decreases and soon approaches 0. That is, in Figure 6 The fifth segment, represented by a solid line, shows the correction coefficient k as "1" up to time Tc1. Then, after the gap g reaches the set value R at time Tc1, the correction coefficient k gradually decreases and soon approaches 0.

[0103] exist Figure 6 The speed of the corrected timing instruction, represented by the solid line in the third segment, becomes waveform W16a, obtained by multiplying the speed of the uncorrected timing instruction (waveform W15b), represented by the dashed line, by the correction factor k. This speed decreases as the correction factor k decreases after time Tc1. Then, similarly to the correction factor k, the speed of the corrected timing instruction asymptotically approaches 0.

[0104] about Figure 6 The positions and accelerations of the timing commands shown in paragraphs 2 and 4, due to the effect of multiplying the velocity by the correction factor k, will also show differences before and after correction after time Tc1. Figure 6 The second segment, represented by a solid line, shows the position of the corrected timing command (waveform W14a) asymptotically approaching the position PX0 separated by the set value Z before reaching the final target position (position P), compared to the position Pf where the preceding carrier 21Q stops. Figure 6The fourth segment, represented by the solid line, shows the acceleration of the corrected timing instruction (waveform W18a), which decelerates from the previous time Tc1, compared to the acceleration of the uncorrected timing instruction (waveform W17b), represented by the dashed line, at the time Tc when deceleration begins. The acceleration gradually approaches 0 as the speed of the corrected timing instruction decreases.

[0105] Next, based on Figure 7 The action of the preceding carrier 21Q when it changes from a stopped state to an operating state, the gap g increases, and the following carrier 21P, which decelerates once, resumes its action before deceleration, will be explained.

[0106] Figure 7 This diagram illustrates an example of a corrected timing command for the linear track control device according to Embodiment 1 when it resumes its original operation after the rear vehicle has decelerated. (Regarding...) Figure 7 The time gap g, position, velocity, acceleration, and correction coefficient k before time Te1, and... Figure 6 Same, explanation omitted.

[0107] Figure 7 The waveforms W11x, W12bx, W13x, W14ax, W15bx, W16ax, W17bx, W18ax, and W19x shown are respectively Figure 6 The waveforms shown are those following W11, W12b, W13, W14a, W15b, W16a, W17b, W18a, and W19.

[0108] exist Figure 6 In the process, waveforms W12b, W15b, and W17b were corrected to waveforms W14a, W16a, and W18a. Figure 7 In the process, waveforms W12bx, W15bx, and W17bx are corrected to waveforms W14ax, W16ax, and W18ax.

[0109] exist Figure 7 The second segment, represented by the dashed line, shows the waveform W13x, which at time Te1, indicates that the preceding carrier 21Q, which had previously stopped at position Pf, begins to move in the direction of separation from the rear carrier 21P.

[0110] Furthermore, the movement of the advance carrier 21Q is accompanied by a change in the target position 34 via the target setting unit 17 corresponding to the advance carrier 21Q, which is different from the operation of the target setting unit 17 corresponding to the rear carrier 21P.

[0111] The result of the movement of the advance carrier 21Q is that, Figure 7The first segment, represented by the solid line, shows the gap g of waveform W11x, which begins to increase after time Te1 and becomes larger than the set value R after time Tf1. Therefore, in Figure 7 The correction coefficient k of the waveform W19x, represented by the solid line in the fifth segment, begins to increase after time Te1 and reaches "1" at time Tf1.

[0112] exist Figure 7 The second, third, and fourth segments, represented by solid lines, show the position, velocity, and acceleration of the corrected timing commands, which begin to increase after time Te1 due to the operation of the timing command generation unit 18. That is, waveforms W14ax, W16ax, and W18ax begin to increase after time Te1. Furthermore, after time Tf1 when the correction coefficient k reaches "1", the acceleration of the timing commands of the rear carrier 21P (waveform W18ax) reaches the set acceleration Aa.

[0113] Then, at time Tg1 Figure 7 The third segment, represented by the solid line, shows the speed of the timing command (waveform W16ax), reaching its maximum speed Vmax, and the acceleration becomes 0 from time Tg1 to time Th1. Then, during the period from time Th1 to time Ti1, the timing command generation unit 18 decelerates the rear carrier 21P using a set acceleration Ad for stopping it at the target position, i.e., position P. Furthermore, at time Ti1, the position of the timing command of the rear carrier 21P (waveform W14ax) reaches position P and stops.

[0114] As described above, the linear track control device 10A according to Embodiment 1 uses the correction coefficient determination unit 19A to correct the timing command based on the correction coefficient k determined by the gap. This allows for operation to avoid interference between the carriers 21 without complex processing. Furthermore, after avoiding interference between the carriers 21, the linear track control device 10A uses the correction coefficient determination unit 19A to correct the timing command based on the correction coefficient k determined by the gap. This allows for rapid restoration to the operation before interference avoidance without performing additional acceleration processing on the rear carrier 21P.

[0115] Furthermore, the linear track system 1A includes a position detector 52 along the transport path 55 to measure the position of the carrier 21. In Embodiment 1, an example using a linear encoder in the position detector 52 was described, but the position of the carrier 21 can also be estimated without using a linear encoder. For example, the linear track control device 10A can replace the linear encoder and estimate the position of the carrier 21 without sensors based on the current generated by the drive control unit C.

[0116] Furthermore, in the linear track system 1A, multiple conveying units 50 are combined to ensure the flexibility of the conveying path 55. In this case, in the linear track system 1A, the carrier 21 moves by switching between multiple conveying units 50, thereby achieving a long conveying path 55. Additionally, in the linear track system 1A, by making the shape of the conveying units 50 curved instead of straight, it is also possible to achieve a conveying path 55 that can be drawn in an arc shape.

[0117] Furthermore, in the linear track system 1A, the position information of the carrier 21 can be managed commonly across the conveying path 55 spanning each conveying unit 50. Therefore, the linear track system 1A can control each carrier 21 using integrated coordinates, regardless of the number of conveying units 50 or the position of the carrier 21. Additionally, the linear track system 1A can control each carrier 21 even when the rear carrier 21P and the preceding carrier 21Q are positioned on the same conveying unit 50. Furthermore, the linear track system 1A can also control each carrier 21 even when the rear carrier 21P and the preceding carrier 21Q are positioned on different conveying units 50.

[0118] Furthermore, in the linear track system 1A, one transport unit 50 is equipped with Figure 2 The linear track control device 10A controls the current flowing in each electromagnet individually, thereby enabling multiple carriers 21 to move on a single transport unit 50.

[0119] In addition, the case where each conveying unit 50 has one drive control unit C in the description of Embodiment 1 has been described, but one drive control unit C can also control the current for multiple conveying units 50.

[0120] Furthermore, the linear track system 1A can control the carrier 21 by setting a target value for the speed or a target value for the position, depending on the user's purpose. The linear track system 1A can obtain the speed by integrating the acceleration of the timing command, and it can also obtain the position by integrating the speed, based on the general properties of differentiation and integration. Therefore, setting a target value for position is essentially the same as indirectly setting a target value for speed. Thus, the linear track system 1A can also set the target value for either position or speed.

[0121] Furthermore, in the description of Embodiment 1, an example was given in which the elongated stator 51 was constructed using an electromagnet and the carrier 21 was constructed using a permanent magnet. However, the linear track system 1A of Embodiment 1 is not limited to this structure. For example, in the linear track system 1A, the stator 51 may also be constructed using a permanent magnet and the carrier 21 may be constructed using an electromagnet.

[0122] In addition, Figure 1 The illustration shows an example of combining multiple conveying units 50 to form a circular conveying path 55, but the shape of the conveying path 55 is not limited to a circular shape. In the linear track system 1A, it is not necessary for the conveying path 55 to circle around the two ends, and the conveying unit 50 may be a single unit instead of multiple units, allowing the conveying path 55 to branch or merge.

[0123] Furthermore, in Embodiment 1, for the sake of simplicity, an example of generating timing commands using three values—upper limit, 0, and lower limit—of acceleration was described. However, the method of command generation is not limited to this. For example, the linear track control device 10A may impose restrictions on acceleration changes to suppress jerk (acceleration), and may use linear or nonlinear filters to suppress mechanical vibrations. As described above, various methods for generating timing commands can be applied to the linear track system 1A.

[0124] As described above, according to Embodiment 1, the linear track system 1A uses a correction coefficient k to correct the position of the rear carrier 21P. Therefore, without complicated processing, collisions between the carriers 21 can be avoided, and the carriers 21 can easily and quickly return to the position before deceleration.

[0125] Implementation method 2.

[0126] Next, use Figures 8 to 11 Embodiment 2 will be described. In Embodiment 2, the linear track control device uses a correction function that makes the acceleration during deceleration constant to control the carrier 21.

[0127] Figure 8 This is a diagram showing the structure of a linear track system having the linear track control device according to Embodiment 2. Regarding... Figure 8 The realization and Figure 1 In the linear orbital system 1A of Embodiment 1 shown, structural elements with the same function are labeled with the same reference numerals, and repeated descriptions are omitted.

[0128] Figure 8 The linear track system 1B shown includes, for example, a linear track control device 10B, a conveying unit 50, a conveying path 55, and multiple carriers 21. That is, compared to the linear track system 1A, the linear track control device 10B replaces the linear track control device 10A. Compared to the linear track control device 10A, the linear track control device 10B replaces the command generation unit 11A.

[0129] Figure 9This is a diagram showing the internal structure of the linear track control device and conveying unit involved in Embodiment 2. Regarding... Figure 9 The realization and Figure 2 In the linear track control device 10A of Embodiment 1 shown, structural elements with the same function are labeled with the same reference numerals, and repeated descriptions are omitted.

[0130] Compared to the instruction generation unit 11A, the instruction generation unit 11B has a correction coefficient determination unit 19B instead of the correction coefficient determination unit 19A. Compared to the correction coefficient determination unit 19A, the correction coefficient determination unit 19B determines a correction coefficient k2 instead of a correction coefficient k. The correction coefficient k2 is determined based on the fact that the acceleration during deceleration becomes a constant correction function.

[0131] In implementation method 1, such as Figure 6 As shown in waveform W18a, the acceleration after time Tc1 increases in the negative direction immediately following time Tc1, and gradually approaches 0 as the velocity of the rear carrier 21P decreases. Given the limitations on the set accelerations Aa and Ad, it is further preferable to ensure acceleration during deceleration to a certain extent so that deceleration can be achieved in a short time, enabling efficient control. Therefore, the linear track control device 10B of Embodiment 2 uses a correction coefficient k2 that ensures acceleration during deceleration to a certain extent to drive and control the rear carrier 21P.

[0132] The correction coefficient k2 used to correct the position of the rear carrier 21P, like the correction coefficient k, can be a correction coefficient used to correct any of the position command, velocity command, and acceleration command. In the following description, we will mainly explain the case where the correction coefficient k2 is used to correct the velocity command. The correction coefficient determination unit 19B sends the correction coefficient k2 to the timing command generation unit 18.

[0133] The processing order of the instruction generation unit 11B of the linear track control device 10B is the same as that of the instruction generation unit 11A of the linear track control device 10A, so its description is omitted.

[0134] Figure 10 This is a diagram used to explain the correction coefficients determined by the linear track control device according to Embodiment 2. Figure 10 The horizontal axis of the graph shown is the gap g, and the vertical axis is the correction coefficient k2. Figure 10 The correction function F2 of the correction coefficient determination unit 19B is shown in the figure. The correction coefficient determination unit 19B of Embodiment 2 determines the correction coefficient k2 based on the gap g at each time by the correction function F2 shown in the following equation (2).

[0135] Formula 2

[0136]

[0137] The correction function F2, expressed by equation (2), is a correction function proportional to the square root of the calculated value using the gap g. More specifically, the correction function F2 is a correction function proportional to the square root of the calculated value using the gap g. Here, the setpoint Z1 and setpoint R1 are constants set by the designer of the linear track system 1B. The correction coefficient determination unit 19A of Embodiment 1 determines a correction coefficient k proportional to the difference between the gap g and the setpoint Z. In contrast, unlike Embodiment 1, the correction coefficient determination unit 19B of Embodiment 2 determines a correction coefficient k2 proportional to the square root of the difference between the gap g and the setpoint Z1.

[0138] Furthermore, the correction coefficient determination unit 19B can also determine the correction coefficient k2 based on a correction function that is proportional to the square root (power root) other than the square root of the difference between the gap g and the set value Z1. Alternatively, the correction coefficient determination unit 19B can also determine the correction coefficient k2 based on a correction function that is proportional to the square root of the value obtained by performing specific calculations on the difference between the gap g and the set value Z1.

[0139] In Implementation Method 2, the square root of the difference between the gap g and the set value Z1, and the square root of the value obtained by performing a specific calculation on the difference between the gap g and the set value Z1, are called the square root of the calculated value using the gap g. The correction coefficient k2, expressed by the relationship in the second paragraph of Equation (2), is proportional to the square root of the calculated value using the gap g. The case where the correction coefficient k2 is proportional to the square root of the calculated value obtained using the gap g will be explained below.

[0140] The design method for setting value Z1 is the same as that for setting value Z in Implementation Method 1. The design method for setting value R1 is determined based on setting value Z1, the maximum speed Vmax set for the rear carrier 21P, and the maximum acceleration Amax (here, 0 < Amax). For example, it is determined that R1 = Vmax. 2 / (2×Amax)+Z1, thus, when the preceding carrier 21P decelerates at the maximum acceleration Amax relative to the stopped preceding carrier 21Q, it is possible to determine the minimum set value R1 at which the preceding and following carriers 21P can stop when the gap g is less than the set value Z1.

[0141] Linear track system 1B can be applied to a linear track system in which the lead vehicle 21Q and the follower vehicle 21P move in directions that approach each other. In this case, R1 = Vmax is further determined based on the maximum velocity Vmaxf and maximum acceleration Amaxf set for the lead vehicle 21Q (here, 0 < Amaxf). 2 / (2×Amax)+Vmaxf 2 / (2×Amaxf)+Z1, thus, under the condition that the maximum acceleration deceleration is applied to both parties, the minimum set value R1 that can be determined before the gap g is less than the set value Z1 can be stopped.

[0142] Next, refer to Figure 11 The operation of the timing instruction generation unit 18 will be explained. Figure 11 This is a diagram illustrating an example of the timing commands generated by the linear track control device according to Embodiment 2. Furthermore, regarding... Figure 11 The waveform shown is related to... Figure 6 For identical or similar waveforms, their descriptions are omitted. Figure 11 The waveforms W21, W22b, W23, W24a, W25b, W26a, W27b, W28a, and W29 shown are respectively... Figure 6 The waveforms W11, W12b, W13, W14a, W15b, W16a, W17b, W18a, and W19 shown correspond to each other. In Figure 11 The second paragraph, and Figure 5 The first segment will also become the final position of the target, represented by position P.

[0143] Figure 11 The waveform W21 shown in the first segment illustrates the time sequence of the position difference g between the rear carrier 21P and the preceding carrier 21Q. The timing instruction generation unit 18 generates timing instruction 30 such that the gap g represented by waveform W21 is not less than the set value Z1.

[0144] exist Figure 11 The second segment, represented by the dashed line, shows waveform W22b, which is... Figure 5 The waveform W1, represented by a solid line, shows the position of the timing command generated before correction to move the rear carrier 21P from position 0 to position P. Figure 11 The image shows the state where the advance carrier 21Q is stopped at position Pf, between position 0 and position P. Figure 11 The second segment, waveform W23, represented by a dashed line, is the waveform indicating the position of the timing instruction of the preceding carrier 21Q. Figure 11The second segment, represented by the solid line waveform W24a, shows the position of the corrected timing instruction of the rear carrier 21P generated by the timing instruction generation unit 18 according to Embodiment 2.

[0145] exist Figure 11 The third segment, represented by the dashed line, shows waveform W25b, which is... Figure 5 The waveform W2 shows the speed of the timing command before correction, represented by the solid line. Figure 11 The third segment, represented by the solid line waveform W26a, shows the speed of the corrected timing instruction of the rear carrier 21P generated by the timing instruction generation unit 18 according to Embodiment 2.

[0146] exist Figure 11 The fourth segment, represented by the dashed line, shows waveform W27b, which is... Figure 5 The waveform W3 shows the acceleration waveform of the timing command before correction, represented by a solid line. Figure 11 The fourth segment, represented by the solid line waveform W28a, shows the acceleration of the corrected timing instruction of the rear carrier 21P generated by the timing instruction generation unit 18 according to Embodiment 2.

[0147] As mentioned above, Figure 11 The waveforms W22b, W25b, and W27b were corrected to waveforms W24a, W26a, and W28a respectively to prevent interference between carriers 21. Figure 11 The fifth segment, represented by the solid line waveform W29, shows the timing data for the correction coefficient k2.

[0148] exist Figure 11 In the following situation, assuming if Figure 11 The timing instructions before correction, represented by the dashed line in the second segment, are used in actual control. In this case, the rear carrier 21P will reach position Pf before reaching position P, and the preceding carrier 21Q and the rear carrier 21P will interfere with each other.

[0149] Furthermore, the timing instruction generation unit 18 may generate at least one timing instruction 30 for waveforms W24a, W26a, and W28a based on the correction coefficient k2. For example, the timing instruction generation unit 18 generates a timing instruction 30 for waveform W26a. The timing instruction generation unit 18 then sends the generated timing instruction 30 to the motion control unit 12.

[0150] Here, for Figure 11 The change process of the waveform shown is explained. During the period from time 0 to time Ta, the position of the rear carrier 21P is position 0, the position of the preceding carrier 21Q is position Pf, and the gap g between the rear carrier 21P and the preceding carrier 21Q becomes a constant value (Pf).

[0151] After time Ta, the gap g gradually decreases as the position of the rear carrier 21P changes. After time Tc2, the size of the gap g falls below the set value R1 used to switch the formula for the correction coefficient k2. Therefore, after time Tc2, the correction coefficient k2 gradually decreases, and soon becomes 0 at time Td2. That is, at... Figure 11 The fifth segment, represented by a solid line, shows the correction coefficient k2 as "1" up to time Tc2. Then, after the gap g reaches the set value R1 at time Tc2, the correction coefficient k2 gradually decreases and soon becomes 0.

[0152] exist Figure 11 The third segment, represented by the solid line, shows the speed of the corrected timing instruction, which becomes waveform W26a obtained by multiplying the speed of the uncorrected timing instruction (waveform W25b), represented by the dashed line, by the correction factor k2. This waveform decreases as the correction factor k2 decreases after time Tc2. Then, similarly to the correction factor k, the speed of the corrected timing instruction becomes 0.

[0153] about Figure 11 The positions and accelerations of the timing commands shown in paragraphs 2 and 4, due to the effect of multiplying the velocity by the correction factor k2, will show differences before and after correction after time Tc2. Figure 11 The second segment, represented by a solid line, shows the position of the corrected timing command (waveform W24a) before reaching the final position P of the target, and stops at position PX1, which is separated from the position Pf where the preceding carrier 21Q stops by the set value Z1. Figure 11 The fourth segment, represented by the solid line (waveform W28a), shows the acceleration of the corrected timing instruction (waveform W27b) starting to decelerate at time Tc, compared to the acceleration of the uncorrected timing instruction (waveform W27b) represented by the dashed line. The acceleration starts to decelerate from the previous time Tc2, and becomes 0 at time Td2 as the speed of the corrected timing instruction decreases.

[0154] In implementation method 1, such as Figure 6 As shown in waveform W18a, the acceleration after time Tc1 increases in the negative direction immediately after time Tc1, and as the velocity of the rear carrier 21P approaches 0, the acceleration gradually approaches 0.

[0155] On the other hand, in implementation method 2, the difference from implementation method 1 is that, as by... Figure 11As shown in waveform W28a, the acceleration during deceleration becomes approximately constant. As a result, in Embodiment 2, the stopping to avoid interference is not a gradual stop as in Embodiment 1, but a rapid stop. That is, compared to the linear track control device 10A of Embodiment 1, the linear track control device 10B of Embodiment 2 can shorten the time spent from the start of deceleration of the rear carrier 21P to a stop. This is the effect achieved by the correction coefficient determination unit 19B of Embodiment 2 using the square root to determine the correction coefficient k2. As described above, in Embodiment 2, the linear track control device 10B performs deceleration in a manner where the acceleration of the rear carrier 21P is constant in order to avoid interference with the stopping preceding carrier 21Q. Therefore, compared to the case performed by the linear track control device 10A, interference can be avoided in a shorter time, and efficient control can be achieved.

[0156] As described above, the linear track control device 10B according to Embodiment 2 corrects the timing command using the correction coefficient k2 determined by the correction coefficient determination unit 19B, thereby enabling the acceleration of the following vehicle 21P to decelerate relative to the stopped preceding vehicle 21Q to be constant. Therefore, the linear track control device 10B achieves more efficient control based on the effects of Embodiment 1.

[0157] In addition, the linear track control device 10B can set the acceleration of the rear carrier 21P to a constant when it decelerates, so that the speed of the rear carrier 21P can change smoothly without interference, and the control with less vibration and noise can be achieved.

[0158] Implementation method 3.

[0159] Next, use Figures 12 to 15 Embodiment 3 will be described. In Embodiment 3, the linear track control device can control the carrier 21 using a correction function that sets the acceleration during deceleration to a constant value and suppresses changes in acceleration when the vehicle stops.

[0160] Figure 12 This is a diagram showing the structure of a linear track system having the linear track control device according to Embodiment 3. Regarding... Figure 12 The realization and Figure 8 In the linear orbital system 1B of Embodiment 2 shown, structural elements with the same function are labeled with the same reference numerals, and repeated descriptions are omitted.

[0161] The linear track system 1C uses a correction function obtained by combining the correction function F1 used in Embodiment 1 and the correction function F2 used in Embodiment 2. When the linear track system 1C decelerates the rear carrier 21P, it first uses the correction function F2 to set the acceleration during deceleration to a constant value, and then uses the correction function F1 to suppress changes in acceleration when the rear carrier 21P stops.

[0162] The linear track system 1C, for example, includes a linear track control device 10C, a conveying unit 50, a conveying path 55, and multiple carriers 21. That is, compared to the linear track system 1B, the linear track system 1C replaces the linear track control device 10B with the linear track control device 10C. Compared to the linear track control device 10B, the linear track control device 10C replaces the command generation unit 11B with the command generation unit 11C.

[0163] Figure 13 This is a diagram showing the internal structure of the linear track control device and conveying unit involved in Embodiment 3. Regarding... Figure 13 The realization and Figure 9 In the linear track control device 10B of Embodiment 2 shown, structural elements with the same function are labeled with the same reference numerals, and repeated descriptions are omitted.

[0164] Compared to the instruction generation unit 11B, the instruction generation unit 11C replaces the correction coefficient determination unit 19B. Compared to the correction coefficient determination unit 19B, the correction coefficient determination unit 19C determines the correction coefficient k3 instead of determining the correction coefficient k2. The correction coefficient k3 is determined based on a correction function that ensures constant acceleration in the initial stage of deceleration and suppresses changes in acceleration in the later stage of deceleration.

[0165] In implementation method 2, such as Figure 11 As shown in waveform W29, immediately preceding time Td2, a roughly constant instantaneous rise in acceleration is observed, characterized by a peak relative to the negative direction (deceleration direction). This produces the phenomenon that, given a setpoint Z1 < g, if the gap g approaches the setpoint Z1, then... Figure 10 The slope of the correction function F2 shown diverges infinitely, causing a temporary increase in the absolute value of acceleration in discrete numerical calculations. This rapid change in acceleration, as described above, causes mechanical vibration, noise, and degradation. Therefore, the linear track control device 10C in Embodiment 3 suppresses the acceleration of the rear carrier 21P just before it comes to a stop.

[0166] The correction coefficient k3 used to correct the position of the rear carrier 21P, like correction coefficients k and k2, can be used to correct any of the position command, velocity command, and acceleration command. In the following description, the case where correction coefficient k3 is used to correct the velocity command will be primarily explained. The correction coefficient determination unit 19C sends the correction coefficient k3 to the timing command generation unit 18.

[0167] The processing order of the command generation unit 11C of the linear track control device 10C is the same as that of the command generation unit 11A of the linear track control device 10A, so its description is omitted.

[0168] Figure 14 This is a diagram used to explain the correction coefficients determined by the linear track control device according to Embodiment 3. Figure 14 The horizontal axis of the graph shown is the gap g, and the vertical axis is the correction coefficient k3. Figure 14 The correction function F3 of the correction coefficient determination unit 19C is shown in the figure. The correction coefficient determination unit 19C of Embodiment 3 determines the correction coefficient k3 based on the gap g at each time by the correction function F3 shown in the following equation (3).

[0169]

Formula 3

[0170]

[0171] The correction function in the third row of the correction function F3 expressed by equation (3) is the first correction function, and the correction function in the second row is the second correction function. The first correction function is set to the range where the gap g is greater than the set value D2 and less than the set value R2, which is the first gap range. In addition, the second correction function is set to the range where the gap g is greater than the set value Z2 and less than or equal to the set value D2, which is the second gap range.

[0172] The correction coefficient determination unit 19C determines intermediate variables A2, B2, and S2 based on the following equations (4), (5), and (6), thereby determining the correction coefficient k3 whose slope of the correction function F3 is consistent before and after the set value D2 of the gap g and changes smoothly with respect to the gap g.

[0173] Formula 4

[0174]

[0175]

Formula 5

[0176]

[0177]

Formula 6

[0178]

[0179] Here, the setpoints Z2, R2, and D2 are constants set by the designer of the linear orbital system 1C. The design method for setpoint Z2 is the same as that for setpoint Z in Embodiment 1. The design method for setpoint R2 is the same as that for setpoint R1 in Embodiment 2.

[0180] The setpoint D2 is a constant used to determine the ratio between the interval for determining the correction coefficient by the correction function F1 and the interval for determining the correction coefficient by the correction function F2, and is determined between the setpoint Z2 and the setpoint R2. For example, when the setpoint D2 is set to the setpoint R2, the correction coefficient k3 is the same as the correction coefficient k in Embodiment 1. When the setpoint D2 is set to the setpoint Z2, the correction coefficient k3 is the same as the correction coefficient k2 in Embodiment 2.

[0181] If the setpoint D2 is decreased, the linear track control device 10C can increase the acceleration used to avoid interference, enabling efficient deceleration. On the other hand, if the setpoint D2 is increased, the linear track control device 10C can strongly suppress the increase in acceleration near the stop, achieving control with low vibration and noise.

[0182] When considering the efficiency of control time in a linear track system 1C, from the viewpoint of fully utilizing acceleration, it is preferable to have as much of the region of constant acceleration as possible. That is, it is preferable to set the range from the setpoint Z2 to the setpoint D2 to a relatively narrow range, and the range from the setpoint D2 to the setpoint R2 to a relatively wide range. For example, if the setpoint D2 is set using distance d, such as D2 = Z2 + d, and d is set slightly smaller, such as 30 mm, then the efficiency of control time in the linear track system 1C becomes good. In addition, if the setpoint D2 is set using the ratio Ra, such as D2 = Z2 + (R2 - Z2) × Ra, and r is set slightly smaller, such as r = 0.05 or less, then the efficiency of control time in the linear track system 1C becomes good.

[0183] In Embodiment 1, the correction coefficient determining unit 19A determines a correction coefficient k that is proportional to the difference between the gap g and the set value Z. On the other hand, in Embodiment 3, the correction coefficient determining unit 19C determines a correction coefficient k3 based on a second correction function that is proportional to the gap g in the interval where the gap g is greater than the set value Z2 and less than or equal to the set value D2. Furthermore, the correction coefficient determining unit 19C determines the correction coefficient k3 based on a first correction function that is proportional to the square root of the calculated value using the gap g in the interval where the gap g is greater than the set value D2 up to the set value R2. That is, the correction coefficient determining unit 19C determines the same correction coefficient as in Embodiment 1 for the interval where the gap g is less than or equal to the set value D2, and the same correction coefficient as in Embodiment 2 for the interval where the gap g is greater than the set value D2. Therefore, the linear track system 1C can prevent vibration and noise and improve the efficiency of control time.

[0184] Next, refer to Figure 15 The operation of the timing instruction generation unit 18 will be explained. Figure 15 This is a diagram illustrating an example of the timing commands generated by the linear track control device according to Embodiment 3. Furthermore, regarding... Figure 15 The waveform shown is related to... Figure 11 For identical or similar waveforms, their descriptions are omitted. Figure 15 The waveforms W31, W32b, W33, W34a, W35b, W36a, W37b, W38a, and W39 shown are respectively... Figure 11 The waveforms W21, W22b, W23, W24a, W25b, W26a, W27b, W28a, and W29 shown correspond to each other. In Figure 15 The second paragraph, and Figure 5 The first segment will also become the final position of the target, represented by position P.

[0185] Figure 15 The waveform W31 shown in the first segment illustrates the time sequence of the position difference g between the rear carrier 21P and the preceding carrier 21Q. The timing instruction generation unit 18 generates timing instruction 30 such that the gap g represented by waveform W31 is not less than the set value Z2.

[0186] exist Figure 15 The second segment, represented by the dashed line, shows waveform W32b, which is... Figure 5 The waveform W1, represented by a solid line, shows the position of the timing command generated before correction to move the rear carrier 21P from position 0 to position P. Figure 15 The image shows the state where the advance carrier 21Q is stopped at position Pf, between position 0 and position P. Figure 15The second segment, waveform W33, represented by a dotted line, is the waveform indicating the position of the timing instruction of the preceding carrier 21Q. Figure 15 The second segment, represented by the solid line waveform W34a, shows the position of the corrected timing instruction of the rear carrier 21P generated by the timing instruction generation unit 18 according to Embodiment 3.

[0187] exist Figure 15 The third segment, represented by the dashed line, shows waveform W35b, which is... Figure 5 The waveform W2 shows the speed of the timing command before correction, represented by the solid line. Figure 15 The third segment, represented by the solid line waveform W36a, shows the speed of the corrected timing instruction of the rear carrier 21P generated by the timing instruction generation unit 18 according to Embodiment 3.

[0188] exist Figure 15 The fourth segment, represented by the dashed line, shows waveform W37b, which is... Figure 5 The waveform W3 shows the acceleration waveform of the timing command before correction, represented by a solid line. Figure 15 The fourth segment, represented by the solid line waveform W38a, shows the acceleration of the corrected timing instruction of the rear carrier 21P generated by the timing instruction generation unit 18 according to Embodiment 3.

[0189] As mentioned above, Figure 15 The waveforms W32b, W35b, and W37b were corrected to waveforms W34a, W36a, and W38a respectively to prevent interference between carriers 21. Figure 15 The fifth segment, represented by the solid line waveform W39, shows the timing data for the correction coefficient k3.

[0190] exist Figure 15 In the following situation, assuming if Figure 15 The timing instructions before correction, represented by the dashed line in the second segment, are used in actual control. In this case, the rear carrier 21P will reach position Pf before reaching position P, and the preceding carrier 21Q and the rear carrier 21P will interfere with each other.

[0191] Furthermore, the timing instruction generation unit 18 may generate at least one timing instruction 30 for waveforms W34a, W36a, and W38a based on the correction coefficient k3. For example, the timing instruction generation unit 18 generates a timing instruction 30 for waveform W36a. The timing instruction generation unit 18 then sends the generated timing instruction 30 to the motion control unit 12.

[0192] Here, for Figure 15The change process of the waveform shown is explained. During the period from time 0 to time Ta, the position of the rear carrier 21P is position 0, the position of the preceding carrier 21Q is position Pf, and the gap g between the rear carrier 21P and the preceding carrier 21Q becomes a constant value (Pf).

[0193] After time Ta, the gap g gradually decreases as the position of the rear carrier 21P changes. After time Tc3, the size of the gap g falls below the set value R2 used to switch the formula for the correction coefficient k3. Therefore, after time Tc3, the correction coefficient k3 gradually decreases. Then, the gap g becomes the set value D2 at time Td3. Furthermore, the correction coefficient k3 also decreases after time Td3, asymptotically approaching 0. That is, at... Figure 11 The correction coefficient k3, represented by the solid line in the fifth segment, is "1" up to time Tc3. Then, after time Tc3 when the gap g reaches the set value R2, the correction coefficient k3 decreases approximately linearly. Moreover, after time Td3 when the gap g reaches the set value D2, the slope of the correction coefficient k3 becomes slightly shallower and then gradually approaches 0.

[0194] exist Figure 15 The speed of the corrected timing instruction, represented by the solid line in the third segment, becomes waveform W36a, obtained by multiplying the speed of the uncorrected timing instruction (waveform W35b), represented by the dashed line, by the correction factor k3. The speed decreases approximately linearly as the correction factor k3 decreases after time Tc3. After time Td3, similar to the correction factor k, the speed of the corrected timing instruction gradually approaches 0.

[0195] about Figure 15 The positions and accelerations of the timing commands shown in paragraphs 2 and 4, due to the effect of multiplying the velocity by the correction factor k3, will show differences before and after correction after time Tc3. Figure 15 The second segment, represented by a solid line, shows the position of the corrected timing command (waveform W34a). It stops at position PX2, separated from the target's final position P, before reaching the target's final position P, and compared to the position Pf where the preceding carrier 21Q stops. In Figure 15 The fourth segment, represented by the solid line (waveform W38a), shows the acceleration of the corrected timing instruction (waveform W37b) starting to decelerate at time Tc, compared to the acceleration of the uncorrected timing instruction (waveform W37b) represented by the dashed line. It starts to decelerate from the previous time Tc3 and gradually approaches 0 as the speed of the corrected timing instruction decreases.

[0196] In implementation method 1, Figure 6As shown by the solid line in waveform W18a, the acceleration after time Tc1 changes significantly in the negative direction immediately after time Tc1, and the acceleration gradually approaches 0 as the velocity of the rear carrier 21P approaches 0.

[0197] On the other hand, in implementation method 3, the difference from implementation method 1 is that, as Figure 15 As shown in waveform W38a, the acceleration remains approximately constant from time Tc3 to time Td3 during deceleration. As a result, the linear track control device 10C of Embodiment 3 takes longer to decelerate from the rear carrier 21P to a near stop compared to the linear track control device 10A of Embodiment 1. That is, the linear track control device 10C of Embodiment 3 can shorten the time taken from deceleration from the rear carrier 21P to a stop compared to the linear track control device 10A of Embodiment 1. This is the effect achieved by the correction coefficient determination unit 19C of Embodiment 3 determining the correction coefficient k3 using the square root. As described above, in Embodiment 3, the linear track control device 10C performs deceleration in a manner where the acceleration of the rear carrier 21P remains constant to avoid interference with the stopped preceding carrier 21Q. Therefore, it can avoid interference in a shorter time compared to the linear track control device 10A, achieving efficient control.

[0198] Additionally, in implementation method 2, such as Figure 11 As shown in waveform W28a, an almost constant instantaneous increase in acceleration is observed immediately before time Td2, exhibiting a peak value relative to the negative direction (deceleration direction). This abrupt change in acceleration, as described above, contributes to mechanical vibration, noise, and degradation.

[0199] On the other hand, the linear track control device 10C in Embodiment 3 begins to decelerate after acceleration (via... Figure 15 The waveform W38a (representing the acceleration of the timing command) does not have a negative peak after the gap g is lower than the set value D2, and gradually approaches 0. Therefore, the linear track control device 10C suppresses the acceleration of the rear carrier 21P before it stops. Thus, the linear track control device 10C according to Embodiment 3 can achieve both efficient control through short-term interference avoidance and control with low vibration and noise.

[0200] As described above, the linear track control device 10C according to Embodiment 3 corrects the timing commands using the correction coefficient k3 determined by the correction coefficient determination unit 19C, thereby enabling the acceleration of the decelerating preceding carrier 21Q to remain constant until midway through the deceleration operation. Furthermore, by using the correction coefficient k3 to correct the timing commands, the linear track control device 10C can suppress changes in the acceleration of the following carrier 21P before it comes to a stop. As a result, the linear track control device 10C can efficiently achieve control with less vibration and noise.

[0201] Implementation method 4.

[0202] Next, use Figures 16 to 20 Implementation method 4 will be described. In implementation method 4, the linear track control device learns the parameters (set in the correction function) used to determine the correction coefficients.

[0203] Figure 16 This is a diagram showing the structure of a linear track system having the linear track control device according to Embodiment 4. Regarding... Figure 16 The realization and Figure 1 In the linear orbital system 1A of Embodiment 1 shown, structural elements with the same function are labeled with the same reference numerals, and repeated descriptions are omitted.

[0204] Figure 16 The linear track system 1D shown includes, for example, a linear track control device 10D, a transport unit 50, a transport path 55, multiple carriers 21, and a trained model storage unit 53. That is, compared to the linear track system 1A, the linear track system 1D has a linear track control device 10D instead of the linear track control device 10A. Compared to the linear track control device 10A, the linear track control device 10D has an instruction generation unit 11D instead of the instruction generation unit 11A.

[0205] Furthermore, the linear orbit system 1D has a trained model storage unit 53, which is not present in the linear orbit system 1A. The trained model storage unit 53 stores the trained model 38, which will be described later. In addition, the trained model storage unit 53 can also be configured outside the linear orbit system 1D.

[0206] In embodiments 1 to 3, the performance of the control device varies depending on the set values ​​and the shape of the function related to the method of determining the correction coefficients. This includes the length of time required for deceleration, the breadth of conditions that can avoid interference between the carriers 21, the magnitude of the generated acceleration, and the magnitude of the generated vibration and noise. Furthermore, in embodiments 1 to 3, it is time-consuming for the user to meticulously adjust the set values ​​and the shape of the function applied to each linear track system 1A to 1C. Therefore, in embodiment 4, the linear track system 1D automatically or adaptively determines the set values ​​related to the method of determining the correction coefficients and the values ​​related to the design of the function shape.

[0207] Figure 17 This is a diagram showing the internal structure of the linear track control device and conveying unit involved in Embodiment 4. Regarding... Figure 17 The realization and Figure 1 In the linear track control device 10A of Embodiment 1 shown, structural elements with the same function are labeled with the same reference numerals, and repeated descriptions are omitted.

[0208] Compared to the instruction generation unit 11A, the instruction generation unit 11D replaces the correction coefficient determination unit 19A. Furthermore, the instruction generation unit 11D has a learning unit 60, which is not present in the instruction generation unit 11A.

[0209] The correction coefficient determination unit 19D, compared with the correction coefficient determination unit 19A, determines the correction coefficient k4 instead of the correction coefficient k. Using the learning results obtained by the learning unit 60 (the trained model 38 described later), the correction coefficient determination unit 19D infers the parameters (coefficient parameters 39) used by the correction function (hereinafter referred to as the correction function F4) corresponding to the correction coefficient k4, and generates the correction function F4 using the coefficient parameters 39. Furthermore, the correction coefficient determination unit 19D determines the correction coefficient k4 based on the correction function F4 and the gap g.

[0210] As described above, the correction coefficient determination unit 19D uses the trained model 38 to infer the coefficient parameter 39 and determine the correction coefficient k4. The coefficient parameter 39 is, for example, the set value Z2, the set value D2, and the set value R2.

[0211] The learning unit 60 includes a learning data acquisition unit 61 and a model generation unit 62. The learning data acquisition unit 61 acquires learning data 37, including the gap g during the correction period and coefficient parameters 39 used to determine the correction coefficient k4 during the correction period. The learning data 37 may include at least one of the position, velocity, acceleration, and jerk (accelerometer) of the rear carrier 21P during the correction period. At least one of the position, velocity, acceleration, and jerk of the rear carrier 21P during the correction period is used in the calculation of the return described later.

[0212] The correction period is the period during which the correction coefficient k4 is applied to the drive of the rear carrier 21P. That is, the correction period is used to prevent interference between the carriers 21, or to restore the rear carrier 21P to its pre-deceleration state. During the correction period, the position, velocity, acceleration, or jerk of the rear carrier 21P is corrected. The learning data acquisition unit 61 determines the correction period based on the gap g and the coefficient parameter 39. The learning data acquisition unit 61 sends the learning data 37 during the correction period to the model generation unit 62.

[0213] The model generation unit 62 generates a trained model 38 based on the learning data 37, which is used to infer the coefficient parameters 39 (parameters of the correction function F4 used to determine the correction coefficient k4) according to the gap g. The model generation unit 62 has a reward calculation unit 63 and a function update unit 64. Details about the reward calculation unit 63 and the function update unit 64 will be described later. The model generation unit 62 outputs the generated trained model 38 to the trained model storage unit 53.

[0214] The trained model storage unit 53 stores the trained model 38. The trained model 38 stored in the trained model storage unit 53 is read out by the correction coefficient determination unit 19D.

[0215] The correction coefficient determination unit 19D includes an inference data acquisition unit 65 and an inference unit 66. The inference data acquisition unit 65 acquires inference data 41 including the gap g. The inference unit 66 infers the correction coefficient determination parameter, i.e., the coefficient parameter 39, based on the trained model 38 stored in the trained model storage unit 53 and the inference data 41. The inference unit 66 uses the coefficient parameter 39 to determine the correction function F4, and determines the correction coefficient k4 based on the correction function F4 and the gap g. The inference unit 66 outputs the correction coefficient k4 to the timing instruction generation unit 18.

[0216] <Learning Stage>

[0217] The learning algorithm used in the learning unit 60 can employ known algorithms such as teacher-led learning, teacherless learning, and reinforcement learning. As an example, we will explain the case where reinforcement learning is applied in the learning algorithm used in the model generation unit 62. In reinforcement learning, an agent (acting entity) within an environment observes the current state (parameters of the environment) and decides on the action to be taken. The environment dynamically changes through the agent's actions, and the agent is rewarded accordingly. The agent repeats this action, learning the action strategy that yields the highest reward through a series of actions. The model generation unit 62 includes a reward calculation unit 63 and a function update unit 64 for reinforcement learning.

[0218] The learning data acquisition unit 61 acquires learning data 37, which includes the correction coefficient determination parameters, namely coefficient parameters 39 (action) and gap g (state). During reinforcement learning, the learning data acquisition unit 61 can acquire the time-series data of coefficient parameters 39 and gap g during the correction period when the linear orbit system 1D is running, and use it as a dataset of learning data 37.

[0219] During the first learning iteration, the learning data acquisition unit 61 acquires either the user-set coefficient parameter 39 or the coefficient parameter 39 preset in the linear track control device 10D as initial values. The coefficient parameter 39 used during the first learning iteration is, for example, in... Figure 14 The coefficient parameters (set value Z2, set value D2, and set value R2) of the correction function F3 are explained in the text. In addition, the learning data acquisition unit 61 acquires the coefficient parameters 39 inferred by the inference unit 66 during the second and subsequent learning.

[0220] Here, refer to Figure 18 The method for determining the correction period for the learning data 37 is explained. Figure 18 This is a diagram used to explain the correction period corresponding to the learning data obtained by the linear track control device according to Embodiment 4. Figure 18 The method for determining the correction period for acquiring the data contained in the learning data 37 will be explained below. As previously described, the data contained in the learning data 37 includes at least one of the following: the position of the rear carrier 21P, the velocity of the rear carrier 21P, the acceleration of the rear carrier 21P, and the jerk of the rear carrier 21P, the gap g, and the coefficient parameter 39. The correction period applied by the learning data acquisition unit 61 is the period during which the correction coefficient k4 corresponding to the learning data 37 is applied, and is set by the learning data acquisition unit 61.

[0221] Figure 18 The horizontal axis represents time. Figure 18 The first segment shows the waveform W41 representing the time shift of the gap g, in Figure 18 The second paragraph shows the waveform W42, representing the time shift of the correction coefficient k4 corresponding to the gap g. That is, in Figure 18 The example shown is the time series data for the gap g and the correction coefficient k4.

[0222] exist Figure 18 The first segment, represented by the solid line waveform W41, is an example of timing data of the gap g between carriers 21 obtained when the linear orbital system 1D is operated under specific conditions. Figure 18 The second segment, represented by the solid line waveform W42, is an example of a correction coefficient k4 obtained under the same specific conditions as the first segment. The correction coefficient k4 represented by waveform W42 can be a correction coefficient corresponding to the correction function F4 generated by the inference unit 66 using the inferred coefficient parameter 39, or it can be a correction coefficient corresponding to... Figure 14 The correction coefficients corresponding to the correction function F3 described in the text.

[0223] exist Figure 18 The graph shown in the first paragraph illustrates a reference value H for the size of the gap g, which is preset by the user without implementing deceleration or other measures to avoid interference between the rear carrier 21P and the preceding carrier 21Q. Additionally, in Figure 18 The figure shown in the first paragraph illustrates the reference value F for the size of the gap g that should be maintained to avoid interference between the rear carrier 21P and the preceding carrier 21Q. Here, F < H.

[0224] exist Figure 18 In the example shown in the first paragraph, as indicated by waveform W41, the gap g is greater than the set value H up to time T1, decreases between the set value H and the set value F from time T1 to time T2, oscillates while decaying near the set value F from time T2 to time T3, and increases between the set value H and the set value F from time T3 to time T4. Furthermore, the gap g is greater than the set value H from time T4 to time T5, less than or equal to the set value H and greater than or equal to the set value F from time T5 to time T6, and becomes larger than the set value H after time T6.

[0225] exist Figure 18 In the second paragraph, the correction coefficient k4, represented by waveform W42, is determined by the correction coefficient determination unit 19D based on the correction model. This correction model has a parameter, namely coefficient parameter 39, used to determine the correction coefficient k4. The linear track control device 10D of Embodiment 4 is used as the correction model. Figure 14The same correction function F3 is shown in Embodiment 3. That is, the linear track control device 10D uses a correction function defined by set values ​​Z2, D2, R2, etc.

[0226] exist Figure 18 In the example shown in paragraph 2, as an example of the initial value of coefficient parameter 39 at the start of learning, the correction coefficient k4 is shown when the value of setting value F is applied to setting value Z2, the value of setting value Z2 is applied to setting value D2, and the value of setting value H is applied to setting value R2. Figure 18 As shown, when the gap g exceeds the set value H, the correction coefficient k4 saturates to "1", and when the gap g is below the set value F, the correction coefficient k4 saturates to "0".

[0227] In the description of Embodiment 4, the gap g and correction coefficient k4 under the above-mentioned settings are shown as an example of the result at the start of learning; however, Embodiment 4 is not limited to this method. For example, the relationship between the gap g and the correction coefficient k4 will naturally change as learning progresses.

[0228] The learning data acquisition unit 61 extracts the correction period for determining the learning data 37 based on the gap g. For example, the learning data acquisition unit 61 extracts data for a series of periods where the gap g is continuously lower than a set value H, and obtains the learning data 37 corresponding to a single dataset. For example, the measurement result of the gap g is obtained... Figure 18 In the case shown, the learning data acquisition unit 61 sets the period from time T1 to time T4 as a correction period for generating one set of learning data 37 corresponding to the dataset. Furthermore, the learning data acquisition unit 61 sets the period from time T5 to time T6 as a correction period for generating one set of learning data 37 corresponding to the dataset. In this case, the learning data acquisition unit 61... Figure 18 The results shown indicate that the correction period for the corresponding amounts of the two datasets was set, and 37 training data points corresponding to the two datasets were generated.

[0229] Furthermore, the learning data acquisition unit 61 can also extract the correction period used to determine the learning data 37 using other methods. The learning data acquisition unit 61 can extract learning data 37 corresponding to a dataset from a series of periods where the gap g is continuously lower than a set value H, and where the gap g is lower than a set value F at least once. When using this method, the learning data acquisition unit 61 will... Figure 18 The period from time T1 to time T4 is set as the correction period for generating learning data 37. In this case, the learning data acquisition unit 61... Figure 18The results shown indicate that the correction period for a total of 1 dataset is set, and 37 training data points corresponding to 1 dataset are generated.

[0230] The learning data acquisition unit 61 acquires the coefficient parameter 39 used when obtaining data during each correction period and uses it as the learning data 37 corresponding to the "action" of reinforcement learning described later. Alternatively, the learning data acquisition unit 61 may replace the coefficient parameter 39 and directly acquire the time-series data of the correction coefficient k4 as one of the learning data 37.

[0231] Furthermore, the learning data acquisition unit 61 acquires the temporal data of the intervals g of each period and uses it as one piece of learning data 37 corresponding to the "state" of reinforcement learning described later. Alternatively, the learning data acquisition unit 61 may acquire the learning data as one piece of learning data 37 after transforming the temporal data into the feature quantities required for calculating the reward described later, even if it does not store all the temporal data of the intervals g of each period.

[0232] The model generation unit 62 learns the coefficient parameter 39 based on the learning data 37, which includes the coefficient parameter 39 and the gap g. That is, it generates a trained model 38 that infers the coefficient parameter 39 based on the gap g between the carriers 21.

[0233] As representative methods of reinforcement learning, Q-learning and TD-learning are known. For example, in the case of Q-learning, the general update formula of the action value function Q(s,a) is expressed by the following equation (7).

[0234]

Formula 7

[0235]

[0236] In equation (7), s t a represents the state of the environment at time t. t Indicates the action at time t. Through action a t The state changes to s t+1 r t+1 Let γ represent the reward resulting from a change in its state, α represent the discount rate, and α represent the learning coefficient. Furthermore, γ is in the range of 0 < γ ≤ 1, and α is in the range of 0 < α ≤ 1. In implementation 4, the coefficient parameter 39 becomes action a. t The gap g becomes state s t The learning department has 60 pairs of states s at time t. t The best action in t To learn.

[0237] The update formula, expressed by equation (7), states that if the action value Q of action a with the highest Q value at time t+1 is greater than the action value Q of action a executed at time t, then the action value Q is increased; otherwise, the action value Q is decreased. In other words, the action value function Q(s, a) is updated in a way that makes the action value Q of action a at time t close to the best action value at time t+1. Thus, the best action value in a given environment is propagated sequentially to the action values ​​in previous environments.

[0238] The return calculation unit 63 calculates the return based on the learning data 37. The return calculation unit 63 calculates the return r based on a specific return benchmark (hereinafter referred to as the return increase benchmark and the return decrease benchmark). For example, in the case of the return increase benchmark, the return calculation unit 63 increases the return r (e.g., assigns a return of "1"), and on the other hand, in the case of the return decrease benchmark, decreases the return r (e.g., assigns a return of "-1").

[0239] Here, referring to the above Figure 18 The reward benchmark is explained. The reward increase benchmark is based on whether the gap g has performed the action desired by the user. The reward calculation unit 63 increases the reward r if the gap g has performed the action desired by the user.

[0240] For example, the condition for satisfying the benchmark of increased returns could be a series of periods of training data 37 for a certain dataset (in... Figure 18 The lengths from time T1 to T4 and from time T5 to T6 in the timing data shown are shorter than a specific set value. In other words, after the rear carrier 21P begins its interference avoidance action, the reward calculation unit 63 provides a good reward while quickly reverting to the action before interference avoidance. By configuring it as described above, the linear orbit system 1D can achieve an avoidance action that is close to the original command before correction, thus enabling efficient control.

[0241] The reward reduction benchmark is based on whether the gap g has performed an action that the user does not want. The reward calculation unit 63 reduces the reward r if the gap g performs an action that the user does not want.

[0242] For example, the condition for meeting the return reduction benchmark could be that the minimum value of the gap g in a series of periods of the training data 37 of a certain dataset is lower than the set value F by ( Figure 18 The difference between the gap Gp1 at time Tp1 and the set value F (i.e., the difference component B1) is greater than the preset set value Bx (not shown).

[0243] By determining the return reference as described above, the linear orbital system 1D can minimize the gap g between the rear carrier 21P and the leading carrier 21Q by a amount lower than the set value F, thereby achieving control with a high probability of avoiding interference between the carriers 21.

[0244] Alternatively, the return reduction benchmark can be determined based on information other than the gap g. For example, the condition for satisfying the return reduction benchmark may be that at least one of the driving force, acceleration, and jerk (jump) of the carrier 21 during the corresponding period exceeds a predetermined set value. By setting as described above, the linear track system 1D can suppress at least one of the driving force, acceleration, and jerk (change in acceleration) that is associated with the avoidance of interference, thereby achieving control with less vibration and noise.

[0245] Furthermore, if the reward increase benchmark is set in a manner that shortens the length of a series of periods as described above, it is possible to learn from a trained model 38 that frequently repeats actions where the interval g is slightly lower than the set value H and then exceeds the set value H in a very short time. Therefore, the time from the end of obtaining one training data 37 ( Figure 18 From time T4 to the start time of acquiring the next learning data 37 ( Figure 18 The condition for satisfying the reward reduction benchmark is that the length of time up to time T5 (times T4 to T5) is smaller than a specific set value. That is, the reward calculation unit 63 can reduce the reward r if the period during which the learning data 37 is not obtained is smaller than the specific set value. As described above, the linear orbit system 1D can be configured to add a penalty (set to reduce the reward r) if the gap g slightly vibrates near the boundary of the set value H, thus increasing the reward r.

[0246] In addition, various feedback criteria are considered to improve control efficiency and the probability of avoiding interference by considering factors such as the driving force, gap g, position of carrier 21, velocity of carrier 21, acceleration of carrier 21, and jerk of carrier 21 associated with actions used to avoid interference. For example, the linear track system 1D serves as a feedback criterion by setting an upper limit value for the RMS (Root Mean Square) of the driving force command 31, and using the condition of exceeding the upper limit value as a condition for satisfying the feedback reduction criterion, thereby enabling control with low drive energy requirements. Furthermore, the linear track system 1D sets an upper limit value for the average value of gap g, and using the condition of exceeding the upper limit value as a condition for satisfying the feedback reduction criterion, thereby enabling control with a small average gap g. Additionally, the linear track system 1D sets an upper limit value for the maximum value of the signal obtained by applying a high-pass filter to the frequency of gap g, and using the condition of exceeding the upper limit as a condition for satisfying the feedback reduction criterion, thereby enabling control with low vibration of gap g.

[0247] The aforementioned return increase benchmark and return decrease benchmark can be appropriately combined. Furthermore, the return benchmark described as a return increase benchmark can be changed to a return decrease benchmark by reversing the order of the compared values. Similarly, the return benchmark described as a return decrease benchmark can be changed to a return increase benchmark by reversing the order of the compared values. That is, if the condition for satisfying the return increase benchmark is that condition X1 is met, the return increase benchmark can be changed to a return decrease benchmark by setting the condition for satisfying the return decrease benchmark to that which is not met. Similarly, if the condition for satisfying the return decrease benchmark is that condition X2 is met, the return decrease benchmark can be changed to a return increase benchmark by setting the condition for satisfying the return increase benchmark to that which is not met.

[0248] The function update unit 64 updates the function used to determine the coefficient parameter 39 according to the reward r calculated by the reward calculation unit 63, and outputs it to the trained model storage unit 53. For example, in the case of Q-learning, the function update unit 64 uses the action value function Q(s) expressed by equation (7). t a t ) and is used as a function to calculate the coefficient parameter 39.

[0249] The learning unit 60 repeats the above learning process. The trained model storage unit 53 stores the updated action value function Q(s) through the function update unit 64. t a t That is, the trained model 38 is stored.

[0250] Next, use Figure 19 This section explains the handling of the 60 learning sessions in the Learning Department. Figure 19 This is a flowchart illustrating the processing sequence of the learning process performed by the linear track control device according to Embodiment 4.

[0251] The learning data acquisition unit 61 acquires the coefficient parameter 39 and the gap g as learning data 37 (step S21). The learning data acquisition unit 61 sends the learning data 37 to the model generation unit 62.

[0252] The model generation unit 62 calculates the return r based on the coefficient parameter 39 and the gap g (step S22). Specifically, the return calculation unit 63 obtains the coefficient parameter 39 and the gap g, and determines whether to increase the return r based on a predetermined return benchmark (step S23).

[0253] If the reward calculation unit 63 determines that the reward r should be increased (step S23, Yes), it increases the reward r (step S24). On the other hand, if the reward calculation unit 63 determines that the reward should be decreased (step S23, No), it decreases the reward r (step S25).

[0254] After steps S24 and S25, the function update unit 64 updates the action value function Q(s) represented by equation (7) stored in the trained model storage unit 53 based on the reward r calculated by the reward calculation unit 63. t a t Update (step S26).

[0255] Learning unit 60 repeats the above steps S21 to S26 to generate the action value function Q(s) t a t ) is stored in the trained model storage section 53 as a trained model 38.

[0256] Furthermore, in order to obtain more learning data 37, the inference unit 66 can determine and output new coefficient parameters 39 each time steps S21 to S26 are repeated. Thus, the linear orbit system 1D can operate using the new coefficient parameters 39 to obtain a new gap g. Therefore, the linear orbit system 1D obtains new learning data 37 based on the newly obtained gap g and other information, thereby enabling learning suitable for various conditions.

[0257] Furthermore, in Embodiment 4, the case where the trained model storage unit 53 is located outside the learning unit 60 has been described, but the trained model storage unit 53 may also be located inside the learning unit 60.

[0258] <Effective Use Phase>

[0259] Next, the operation of the correction coefficient determination unit 19D will be explained. The inference data acquisition unit 65 acquires the gap g at each time when the correction coefficient k4 is determined. The inference data acquisition unit 65 sends the inference data 41 containing the gap g to the inference unit 66.

[0260] The inference unit 66 reads the trained model 38 from the trained model storage unit 53. The inference unit 66 uses the trained model 38 to infer the coefficient parameters 39. That is, the inference unit 66 inputs the gap g obtained by the inference data acquisition unit 65 into the trained model 38, thereby enabling it to infer the coefficient parameters 39 suitable for the gap g.

[0261] The inference unit 66 generates a correction function F4 defined by the inferred coefficient parameter 39, and determines the correction coefficient k4 by inputting a gap g to the correction function F4. The inference unit 66 sends the inferred coefficient parameter 39 to the learning data acquisition unit 61, and sends the determined correction coefficient k4 to the timing instruction generation unit 18.

[0262] Furthermore, in Embodiment 4, the case where the inference unit 66 outputs the coefficient parameter 39 using the trained model 38 learned by the model generation unit 62 was described. However, the inference unit 66 may also obtain the trained model 38 from other linear orbital systems and output the coefficient parameter 39 based on the trained model 38.

[0263] Next, use Figure 20 The processing of the inference coefficient parameter 39 by the inference unit 66 will be explained. Figure 20 This is a flowchart illustrating the processing sequence of the inference process performed by the linear track control device according to Embodiment 4.

[0264] The inference data acquisition unit 65 acquires the gap g at each time point and uses it as inference data 41 (step S31). The inference data acquisition unit 65 sends the inference data 41 containing the gap g to the inference unit 66.

[0265] The inference unit 66 reads the trained model 38 from the trained model storage unit 53. The inference unit 66 inputs a gap g to the trained model 38 read from the trained model storage unit 53 (step S32) to obtain coefficient parameters 39. The inference unit 66 sets the obtained coefficient parameters 39 to the correction function F4. That is, the inference unit 66 generates a correction function F4 with the coefficient parameters 39 set (step S33). The correction function F4 generated by the inference unit 66 can be stored either by the inference unit 66 or externally. If the correction function F4 is stored in a storage device external to the inference unit 66, the inference unit 66 outputs the correction function F4 to the external storage device.

[0266] The correction function F4 is a function defined using the output coefficient parameter 39. The correction function F4 corresponds to the correction coefficient k4. The inference unit 66 determines the correction coefficient k4 based on the gap g and the correction function F4. Specifically, the inference unit 66 determines the correction coefficient k4 by inputting the gap g into the correction function F4 (step S34). The inference unit 66 sends the determined correction coefficient k4 to the timing instruction generation unit 18.

[0267] The linear track control device 10D repeats steps S31 to S34 in each cycle of updating the correction coefficient k4, thereby enabling the operation of the linear track system 1D with low vibration and noise based on the learned trained model 38.

[0268] Furthermore, in Embodiment 4, the application of reinforcement learning in the learning algorithm used by the inference unit 66 was described, but it is not limited to this. Regarding the learning algorithm, in addition to reinforcement learning, teacher-assisted learning or semi-teacher-assisted learning can also be applied.

[0269] In addition, the learning algorithm used in the model generation unit 62 can be deep learning, which learns by extracting the feature quantity itself, or it can perform machine learning according to other well-known methods, such as neural networks, genetic programming, functional logic programming, support vector machines, etc.

[0270] Furthermore, the learning unit 60 and the inference unit 66 may be separate devices connected to the linear orbit system 1D via a network. Alternatively, the learning unit 60 and the inference unit 66 may reside on a cloud server.

[0271] Furthermore, the model generation unit 62 can learn the coefficient parameters 39 using learning data 37 obtained from multiple linear track systems. Additionally, the model generation unit 62 can obtain learning data 37 from multiple linear track systems used in the same area, or it can learn the coefficient parameters 39 using learning data 37 collected from multiple linear track systems operating independently in different areas. Furthermore, the linear track control device that collects the learning data 37 can be added to the object midway, or it can be removed from the object. Moreover, the learning unit 60, which has learned the coefficient parameters 39 for a certain linear track system, can be applied to other linear track systems, and the coefficient parameters 39 can be updated by relearning for those other linear track systems.

[0272] As described above, in Embodiment 4, the learning unit 60 of the linear track control device 10D generates a trained model 38 for inferring the coefficient parameters 39 of the correction function F4, and the correction coefficient determination unit 19D uses the trained model 38 to infer the coefficient parameters 39. Therefore, the user of the linear track system 1D can achieve efficient control with low vibration and noise even without setting the coefficient parameters 39 of the correction function F4 in detail.

[0273] Here, the hardware structure of the linear track control devices 10A to 10D will be described. Furthermore, since the linear track control devices 10A to 10D have the same hardware structure, the hardware structure of the linear track control device 10A will be described here.

[0274] Figure 21 This diagram illustrates an example of the hardware structure for implementing the linear track control device according to Embodiment 1. The linear track control device 10A can be implemented using an input device 300, a processor 100, a memory 200, and an output device 400. Examples of the processor 100 include a CPU (also known as a Central Processing Unit, processing device, arithmetic unit, microprocessor, microcomputer, DSP (Digital Signal Processor)) or a system LSI (Large Scale Integration). Examples of the memory 200 include RAM (Random Access Memory) and ROM (Read Only Memory).

[0275] The linear track control device 10A is implemented by the processor 100 reading and executing a computer-executable control program (linear track control program) stored in the memory 200 for performing the actions of the linear track control device 10A. The program for performing the actions of the linear track control device 10A, i.e., the control program, can be said to be the sequence or method by which the computer performs the actions of the linear track control device 10A.

[0276] The control program executed by the linear track control device 10A is a modular structure including the instruction generation unit 11A and the drive control units C1 to Cm. The instruction generation unit 11A and the drive control units C1 to Cm are downloaded to the main storage device, and the instruction generation unit 11A and the drive control units C1 to Cm are generated in the main storage device.

[0277] The input device 300 receives rear position information 40P and preceding position information 40Q from the position detector 52 and sends them to the processor 100.

[0278] The memory 200 stores control programs, etc. Additionally, the memory 200 is used as temporary storage when the processor 100 performs various processes. The output device 400 outputs current to the stator 51.

[0279] The control program can be provided as a computer program product, stored on a computer-readable storage medium in an installable or executable form. Alternatively, the control program can also be provided to the linear track control device 10A via a network such as the Internet. Furthermore, the functions of the linear track control device 10A can be implemented partly through dedicated hardware such as dedicated circuits, and partly through software or firmware.

[0280] Alternatively, the hardware structure of the instruction generation unit 11A can be set as follows: Figure 21The hardware structure shown is shown. Alternatively, the hardware structure of the drive control units C1 to Cm can also be set as follows: Figure 21 The hardware structure shown. Alternatively, a portion of the instruction generation units 11A-11D (e.g., the learning unit 60, the correction coefficient determination unit 19D) may be configured as... Figure 21 The hardware structure shown.

[0281] Furthermore, the hardware structure of the linear track control devices 10A to 10D can also include multiple processors, multiple memories, multiple input devices, and multiple output devices. Additionally, the linear track control devices 10A to 10D can be control devices comprised of a single frame, or they can be divided into multiple frames, each containing a processor, memory, input devices, and output devices, with the frames cooperating to form the linear track control devices 10A to 10D.

[0282] The structure shown in the above embodiments is an example, and it can also be combined with other known technologies, and the embodiments can be combined with each other. Without departing from the spirit of the subject, some parts of the structure can be omitted or changed.

[0283] Explanation of the label

[0284] 1A-1D linear track system, 10A-10D linear track control device, 11A-11D command generation unit, 12 motion control unit, 13 current control unit, 16 subtractor, 17 target setting unit, 18 timing command generation unit, 19A-19D correction coefficient determination unit, 21 carrier, 21P rear carrier, 21Q advance carrier, 30 timing command, 31 driving force command, 34 target position, 37 learning data, 38 trained model, 39 coefficient parameters, 40P rear position information, 40Q advance position Information, 41 Inference data, 50 Transport unit, 51 Stator, 52 Position detector, 53 Trained model storage unit, 55 Transport path, 60 Learning unit, 61 Learning data acquisition unit, 62 Model generation unit, 63 Feedback calculation unit, 64 Function update unit, 65 Inference data acquisition unit, 66 Inference unit, 100 Processor, 200 Memory, 300 Input device, 400 Output device, C, C1 to Cm drive control unit, F1 to F4 correction functions, g, Gp1 gap, k, k2 to k4 correction coefficients.

Claims

1. A linear track control device, characterized in that, have: The drive control unit controls the current used to generate the driving force between the first carrier moving along the transport path where the stator is arranged and the stator, and controls the current used to generate the driving force between the second carrier arranged in the travel direction of the first carrier and the stator. as well as The instruction generation unit generates a timing instruction that specifies at least one of the position, velocity, and acceleration of the first carrier according to a time sequence, and outputs it to the drive control unit. The instruction generation unit has: The target setting unit sets a target value for a target representing the position or velocity of the first carrier; The correction coefficient determining unit determines a correction coefficient for correcting any one of the position, velocity, and acceleration of the first carrier, based on the relative value (i.e., the gap) between position information representing the position of the first carrier on the transport path and position information representing the position of the second carrier on the transport path; and The timing instruction generation unit generates the timing instructions based on the target value and the correction coefficient. The correction coefficient determination unit inputs the value of the gap into a correction function that represents the relationship between the gap and the correction coefficient, thereby determining the correction coefficient.

2. The linear track control device according to claim 1, characterized in that, The correction coefficient is a coefficient used to correct the speed command, which is used to control the speed of the first carrier.

3. The linear track control device according to claim 2, characterized in that, The correction coefficient determination unit determines the correction coefficient, represented by numbers from 0 to 1, based on the correction function.

4. The linear track control device according to claim 2 or 3, characterized in that, The correction coefficient determination unit determines the correction coefficient based on the correction function, which is proportional to the square root of the calculated value obtained using the gap.

5. The linear track control device according to claim 4, characterized in that, The correction coefficient determination unit determines the correction coefficient based on the correction function, which is proportional to the square root of the calculated value obtained using the gap.

6. The linear track control device according to claim 2 or 3, characterized in that, The correction function includes a first correction function that is proportional to the square root of the calculated value using the gap and a second correction function that is proportional to the gap. The first correction function is set within the range of the gap, i.e., the first gap range. The second correction function is set within the range of the gap, i.e., the second gap range. The value of the gap included in the first gap range is greater than the value of the gap included in the second gap range.

7. The linear track control device according to claim 1, characterized in that, The correction coefficient is a coefficient used to correct the position command, which is used to instruct the position of the first carrier.

8. The linear track control device according to claim 1, characterized in that, It also has: A learning data acquisition unit acquires learning data, which includes parameters for determining the correction coefficient, namely, correction coefficient determination parameters, and the gap corresponding to the correction coefficient determination parameters; and The model generation unit generates a trained model based on the learning data, which is used to infer parameters for determining the correction coefficients according to the gaps.

9. The linear track control device according to claim 8, characterized in that, The learning data includes at least one of the following: the gap during the period of position correction of the first carrier, the position of the first carrier, the velocity of the first carrier, the acceleration of the first carrier, and the jerk of the first carrier.

10. The linear track control device according to claim 8 or 9, characterized in that, The correction coefficient determining unit has: The inference data acquisition unit acquires inference data including the gap; and The inference unit infers the parameters for determining the correction coefficients based on the trained model and the inference data, and determines the correction coefficients using the parameters for determining the correction coefficients.

11. A linear orbital system, characterized in that, have: The conveying path is equipped with a stator; The first carrier moves along the transport path; A second carrier, which is positioned in the direction of travel of the first carrier; and A linear track control device that drives the first carrier and the second carrier. The linear track control device has: A drive control unit controls the current used to generate the drive force between the stator and the first carrier, and also controls the current used to generate the drive force between the stator and the second carrier. as well as The instruction generation unit generates a timing instruction that specifies at least one of the position, velocity, and acceleration of the first carrier according to a time sequence, and outputs it to the drive control unit. The instruction generation unit has: The target setting unit sets a target value for a target representing the position or velocity of the first carrier; The correction coefficient determining unit determines a correction coefficient for correcting any one of the position, velocity, and acceleration of the first carrier, based on the relative value (i.e., the gap) between position information representing the position of the first carrier on the transport path and position information representing the position of the second carrier on the transport path; and The timing instruction generation unit generates the timing instructions based on the target value and the correction coefficient. The correction coefficient determination unit inputs the value of the gap into a correction function that represents the relationship between the gap and the correction coefficient, thereby determining the correction coefficient.

Images