Adaptive control system and method based on moment MCB control panel

Abstract

The invention discloses a self-adaptive control system and method based on a moment MCB control panel, the self-adaptive control system based on the moment MCB control panel comprises the moment MCB control panel, a detection device, a main control unit and a motion control unit, the detection device is configured to respond to a request needing self-adaption, and the main control unit is configured to control the motion control unit. A self-adaptive instruction is sent to the moment MCB control panel; after the moment MCB control panel receives the self-adaptive instruction, a first control signal is sent to the detection equipment based on the self-adaptive instruction; the detection equipment is further configured to receive a first control signal sent by the moment MCB control panel and send the first control signal to the main control unit; the main control unit is configured to send a first target control instruction to the motion control unit based on the first control signal so as to control the conveying track to move to enable a bearing object borne by the conveying track to reach a preset position.

Description

Technical Field

[0001] The present invention relates to the field of control technology, and in particular to an adaptive control system and method based on a matrix MCB control board. Background Art

[0002] With the continuous advancement of industrial automation and intelligent manufacturing, the market demand for efficient and flexible visual inspection equipment continues to increase. Machine vision is a crucial technology in the field of manufacturing quality control, especially in the electronics manufacturing, automotive manufacturing, medical equipment manufacturing, communications industry, consumer electronics, aerospace, and industrial automation fields, with a large number of application cases.

[0003] In visual inspection, the application of motion control systems usually involves coordinated control of mechanical motion to support the functions of the visual system. How to use motion control systems in the field of visual inspection to achieve automated inspection and thus achieve precise control of the motion of motion actuators, and realize functions such as precision positioning, automatic focus and alignment, motion smoothness, multi-axis collaborative control, real-time feedback and adjustment is an urgent problem to be solved.

[0004] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of the present invention is to provide an adaptive control system and method based on a matrix MCB control board, aiming to solve the above-mentioned problems in the prior art.

[0006] To achieve the above object, the present invention provides an adaptive control system based on a Matrix MCB control board, comprising a Matrix MCB control board, a detection device, a main control unit, and a motion control unit; wherein,

[0007] The detection device is configured to send an adaptation instruction to the Matrix MCB control board in response to a request for adaptation;

[0008] After receiving the adaptive instruction, the Juzi MCB control board sends a first control signal to the detection device based on the adaptive instruction;

[0009] The detection device is also configured to receive a first control signal sent by the Juzi MCB control board, and send the first control signal to the main control unit;

[0010] The main control unit is configured to send a first target control instruction to the motion control unit based on the first control signal, so as to control the conveying track to move the carried object on the conveying track to a preset position;

[0011] The motion control unit is configured to receive the first target control instruction sent by the main control unit to control the operation of the corresponding motor to control the movement of the conveying track;

[0012] The MCB control board is further configured to obtain the in-place information of the carrier, calculate the deviation value according to the in-place information, and send an adjustment control signal to the detection device according to the deviation value until the calculated deviation value tends to be stable, and the deviation value is the difference between the current position of the carrier on the conveying track and the preset position;

[0013] The detection device also includes receiving the adjustment control signal and sending the adjustment control signal to a main control unit; the main control unit is also configured to send a second target control instruction or a third target control instruction to the motion control unit based on the adjustment control signal to control the movement of the conveying track and adjust the position of the carrier.

[0014] Preferably, in the adaptive control system based on the Matrix MCB control board, the Matrix MCB control board is further configured as follows:

[0015] When the carried object does not reach the preset position, an adjustment control signal for moving the carried object in a forward direction is sent to the detection device, so that the detection device sends a second target control instruction for moving the conveying track in a forward direction through the main control unit;

[0016] When the carried object exceeds the preset position, an adjustment control signal for causing the carried object to move in the reverse direction is sent to the detection device, so that the detection device sends a third target control instruction for causing the conveying track to move in the reverse direction through the main control unit.

[0017] Preferably, in the adaptive control system based on the Matrix MCB control board, the Matrix MCB control board is further configured as follows:

[0018] According to the current adjustment control signal and the corresponding deviation value, based on the trained reinforcement learning model, the next adjustment control signal is output.

[0019] Preferably, in the adaptive control system based on the Matrix MCB control board, the Matrix MCB control board is further configured as follows:

[0020] According to different adjustment control signals and corresponding deviation values, the reinforcement learning model is used for training until the deviation value converges to obtain the optimal strategy. The state of the reinforcement learning model includes the current arrival information and the current adjustment control signal, and the action space includes forward and backward actions.

[0021] Preferably, in the adaptive control system based on the matrix MCB control board, the reinforcement learning model is used to perform multiple trainings according to different first control signals and corresponding deviation values ​​until the deviation value converges to obtain the optimal strategy.

[0022] In the learning process of each step of the reinforcement learning model training, the Kalman filter is used to predict the state and error covariance of the next moment, and the new observation data is used to correct the error and update the state estimate. The state estimate updated by the Kalman filter is used to adjust the model parameters.

[0023] Preferably, in the adaptive control system based on the Motion MCB control board, the detection device is further configured to send a sensor calibration instruction to the Motion MCB control board in response to a request for sensor calibration;

[0024] The Juzi MCB control board is also configured to receive a sensor calibration instruction sent by the detection device and send a second control signal to the detection device;

[0025] The detection device is also configured to receive a second control signal sent by the Juzi MCB control board, and send the second control signal to the main control unit;

[0026] The main control unit is configured to receive a second control signal sent by the main control unit, and send a fourth target control instruction to the motion control unit based on the second control signal, so as to control the movement of the conveying track so that the calibration substrate carried by the conveying track is controlled to move back and forth on the conveying track;

[0027] The Matrix MCB control board is also configured to obtain the number of motor movement pulses when the calibration substrate triggers sensors at different positions; and calculate the distance between each sensor based on the number of motor movement pulses and the size of the calibration substrate, and update the conversion relationship between the number of motor movement pulses and the distance, and send the conversion relationship to the detection device.

[0028] Preferably, in the adaptive control system based on the matrix MCB control board, the length of the calibration substrate is a, and a first sensor and a second sensor are sequentially arranged along the conveying direction of the conveying track;

[0029] The Matrix MCB control board is also configured as:

[0030] According to the length of the calibration substrate and the number of pulses of the motor from the time when the first sensor is triggered to the time when the trigger stops, a first relational expression for converting the number of pulses to the distance is obtained;

[0031] According to the first pulse number of the motor when the standard calibration substrate moves from the first sensor to the second sensor and a first relationship, the relationship between the distance from the first sensor to the second sensor and the first pulse number is obtained.

[0032] In order to achieve the above object, the present invention further provides an adaptive control method based on a Matrix Control Board (MCB) control panel, which is applied to the Matrix Control Board (MCB). The adaptive control method based on the Matrix Control Board (MCB) control panel comprises:

[0033] In response to receiving an adaptive instruction sent by the detection device, based on the adaptive instruction, a first control signal is sent to the detection device, so that the detection device sends a first target control instruction to the motion control unit through the main control unit based on the first control signal, so as to control the conveying track to move the carried object carried by the conveying track to a preset position;

[0034] Obtaining the location information of the carried object, and calculating the deviation value according to the location information;

[0035] An adjustment control signal is sent to the detection device according to the deviation value, so that the detection device sends a second target control instruction or a third target control instruction to the motion control unit through the main control unit based on the adjustment control signal to control the movement of the conveying track and adjust the position of the carrier until the calculated deviation value tends to be stable, and the deviation value is the difference between the current position of the carrier on the conveying track and the preset position.

[0036] Preferably, in the adaptive control method based on the matrix MCB control board, sending the adjustment control signal to the detection device according to the deviation value includes:

[0037] When the carried object does not reach the preset position, an adjustment control signal for moving the carried object in a forward direction is sent to the detection device, so that the detection device sends a second target control instruction for moving the conveying track in a forward direction through the main control unit;

[0038] When the carried object exceeds the preset position, an adjustment control signal for causing the carried object to move in the reverse direction is sent to the detection device, so that the detection device sends a third target control instruction for causing the conveying track to move in the reverse direction through the main control unit.

[0039] Preferably, in the adaptive control method based on the matrix MCB control board, in the step of sending the adjustment control signal to the detection device according to the deviation value, the method for obtaining the adjustment control signal includes:

[0040] According to different adjustment control signals and corresponding deviation values, the reinforcement learning model is used for training until the deviation value converges to obtain the optimal strategy, wherein the state of the reinforcement learning model includes the current arrival information and the current adjustment control signal, and the action space includes forward and backward.

[0041] The present invention has at least the following beneficial effects:

[0042] The present invention provides an adaptive control system based on a Matrix MCB control board, wherein the detection device is configured to send an adaptive instruction to the Matrix MCB control board in response to a request for adaptive control; after receiving the adaptive instruction, the Matrix MCB control board sends a first control signal to the detection device based on the adaptive instruction; the detection device is further configured to receive the first control signal sent by the Matrix MCB control board, and send the first control signal to a main control unit; the main control unit is configured to send a first target control instruction to the motion control unit based on the first control signal, so as to control the movement of the conveying track so that the object carried by the conveying track moves to a preset position; the motion control unit is configured to receive the first control signal sent by the main control unit The first target control instruction sent controls the corresponding motor to operate to control the movement of the conveying track; the MCB control board is also configured to obtain the in-place information of the carrier, calculate the deviation value according to the in-place information, and send an adjustment control signal to the detection device according to the deviation value until the calculated deviation value tends to be stable, and the deviation value is the difference between the current position of the carrier on the conveying track and the preset position; the detection device also includes receiving the adjustment control signal and sending the adjustment control signal to the main control unit; the main control unit is also configured to send a second target control instruction or a third target control instruction to the motion control unit based on the adjustment control signal to control the movement of the conveying track and adjust the position of the carrier, so that self-adaptation can be achieved;

[0043] Furthermore, the adaptive control system based on the Motion MCB control board provided by the present invention combines reinforcement learning, continuously gives reward feedback to maximize the cumulative reward, and combines Kalman filtering to correct errors and update state estimates by observing actual reward data. That is, through this learning process, the appropriate stop state for different PCB substrates is obtained. In this way, the present invention realizes precise control of the movement of the transmission mechanism of the machine vision inspection equipment through the adaptive transmission system of the Motion MCB control board combined with machine learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A schematic diagram of an adaptive control system based on a Motion MCB control board provided by the present invention;

[0045] Figure 2A schematic diagram of an adaptive control method based on a Motion MCB control board provided by the present invention;

[0046] Figure 3 Schematic diagram of the position relationship between the sensor and the track of the present invention.

[0047] Reference numerals of the present invention:

[0048] 100-MCB control board, 200-detection equipment, 300-main control unit, 400-motion control unit.

[0049] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0050] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict.

[0051] In the embodiments of the present invention, the term "and / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0052] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0053] In the embodiments of the present invention, the term "plurality" refers to two or more than two, and other quantifiers are similar.

[0054] In the present invention, unless otherwise specified, the directional words used, such as "up, down, top, bottom", usually refer to the directions shown in the drawings, or to the components themselves in the vertical, perpendicular or gravity directions; similarly, for ease of understanding and description, "inside and outside" refer to the inside and outside relative to the outline of each component itself, but the above-mentioned directional words are not used to limit the present invention.

[0055] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. However, it can be understood by those skilled in the art that in the embodiments of the present invention, many technical details are proposed in order to enable the reader to better understand the present invention. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present invention can also be implemented. The division of the following embodiments is for the convenience of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined and referenced with each other under the premise of no contradiction.

[0056] Figure 1 The schematic diagram of the adaptive control system based on the MCB control board provided by the present invention is shown. Figure 1 As shown, the present invention provides an adaptive control system based on a MCB control board, which includes a MCB control board 100, a detection device 200, a main control unit 300, and a motion control unit 400. The adaptive control system based on the MCB control board can achieve precise control of the motion of the transmission mechanism of the detection device 200 through adaptation.

[0057] The main control unit may be, but is not limited to, an RX71 M MCU chip and a high-performance MCX514 servo control chip.

[0058] Figure 2 FIG. 1 is a schematic diagram showing an adaptive control method based on the matrix MCB control board 100 provided by the present invention. Figure 2 As shown, the inspection device 200 is configured to send an adaptation instruction to the Matrix MCB control board 100 in response to a request for adaptation. The inspection device 200 may be, but is not limited to, an AOI visual inspection device 200 .

[0059] After receiving the adaptive instruction, the Juzi MCB control board 100 sends a first control signal to the detection device 200 based on the adaptive instruction; the detection device 200 is also configured to receive the first control signal sent by the Juzi MCB control board 100, and send the first control signal to the main control unit 300; the main control unit 300 is configured to send a first target control instruction to the motion control unit 400 based on the first control signal to control the movement of the conveying track so that the carrier carried by the conveying track moves to a preset position; the motion control unit 400 is configured to receive the first target control instruction sent by the main control unit 300 to control the operation of the corresponding motor to control the movement of the conveying track. It should be noted that the carrier can be, but is not limited to, a PCB substrate.

[0060] The Juzi MCB control board 100 is also configured to obtain the in-place information of the carrier, calculate the deviation value according to the in-place information, and send an adjustment control signal to the detection device 200 according to the deviation value until the calculated deviation value tends to be stable, and the deviation value is the difference between the current position of the carrier on the conveying track and the preset position. More specifically, the Juzi MCB control board 100 is also configured to: when the carrier does not reach the preset position, send an adjustment control signal for moving the carrier forward to the detection device 200, so that the detection device 200 issues a second target control instruction for moving the conveying track forward through the main control unit 300; when the carrier exceeds the preset position, send an adjustment control signal for moving the carrier in the reverse direction to the detection device 200, so that the detection device 200 issues a third target control instruction for moving the conveying track in the reverse direction through the main control unit 300. The in-place information may include exceeding the preset position or not reaching the preset position.

[0061] The detection device 200 also includes receiving the adjustment control signal and sending the adjustment control signal to the main control unit 300; the main control unit 300 is also configured to send a second target control instruction or a third target control instruction to the motion control unit 400 based on the adjustment control signal to control the movement of the conveying track and adjust the position of the carrier.

[0062] like Figure 3 As shown, the preset position is the stop plate position, and the carrier is the PCB substrate as an example. After the motor of the conveying track receives the pulse signal, the PCB substrate may exceed or fail to reach the stop plate position. At this time, the distance exceeded or not reached is recorded as the error value. At this time, the Matrix MCB control board 100 will send an adjustment control signal to control the PCB substrate to move forward or backward to reach the stop plate position based on the error value. When the PCB substrate still does not reach the stop plate position after the motor receives the corresponding pulse signal, the new error value is recorded. After multiple times of convergence of the adjustment control signal given by the Matrix MCB control board 100 -> the PCB substrate continues to move backward or forward if it is not in place and records the new error value -> the Matrix MCB control board 100 continues to converge the adjustment control signal given according to the new error value. Through the adaptive process, the error value gradually converges, and the PCB substrate will also tend to converge when it reaches the stop plate position. For the convenience of description below, the preset position is the stop plate position, and the carrier is the PCB substrate as an example for explanation.

[0063] Since the error value is a state of up and down fluctuations, in order to suppress the fluctuations of the error value, reinforcement learning can be combined. More specifically, according to the current adjustment control signal and the corresponding deviation value, based on the trained reinforcement learning model, the next adjustment control signal is output. The forward / backward instruction is controlled by the adjustment control signal sent by the Matrix MCB control board 100, and positive and negative reward feedback is performed to obtain the learning data of the reinforcement learning and update the parameters of the reinforcement learning model. The current adjustment control signal is input, and this pulse command value may have errors. The fluctuation can be suppressed by reinforcement learning, and the next adjustment control signal after learning correction is output, so that the fluctuations of the error between the more violent ones can be suppressed at least partially, and by maximizing the cumulative reward, the error value of the upper and lower fluctuations eventually tends to be dynamically stable, so that the distance of the PCB substrate to the stop board position will also tend to converge, and finally obtain the appropriate stop board state of the current PCB substrate. That is, the adaptive process of the present invention can obtain the appropriate stop board state for different PCB substrates.

[0064] More specifically, according to different adjustment control signals and corresponding deviation values, the reinforcement learning model is used for training until the deviation value converges to obtain the optimal strategy, where the state of the reinforcement learning model includes the current arrival information and the current adjustment control signal, and the action space includes forward and backward actions. The optimal strategy is the optimal action of forward or backward.

[0065] Furthermore, the Kalman filter model can be combined with reinforcement learning to suppress the fluctuation of error values. In each step of the reinforcement learning model training, the Kalman filter is used to predict the state and error covariance at the next moment, and the new observation data is used to correct the error and update the state estimate. The state estimate updated by the Kalman filter is used to adjust the model parameters.

[0066] It should be noted that there is an error between the pulse signal received by the motor and the expected value, so the Kalman filter model is needed to reduce the error, which can further suppress the error value. After giving the initial first control signal instruction, the MCB control board 100 compares the pulse signal received by the motor with the expected value to obtain the initial error. In each reinforcement learning process, the Kalman filter is used to predict the state at the next moment and the error between the pulse signal received by the motor and the expected value, and the new observation data (such as the actual reward) is used to correct the error and update the state estimate, and the next pulse signal is given.

[0067] The state estimate updated by the Kalman filter can effectively adjust the parameters of the model, so that each policy update not only depends on experience replay or gradient information, but also performs a more stable state estimate update through the state and error filtering process.

[0068] As machine learning progresses, the Kalman filter continuously adjusts model parameters and error estimates. When the system error gradually converges, the model state gradually stabilizes. At this point, reinforcement learning and the Kalman filter will continue to provide feedback and adaptive learning with smaller adjustments to cope with possible environmental changes and noise interference.

[0069] In addition, the adaptive control system based on the Matrix MCB control board can also perform self-calibration of the sensor spacing. Figure 2 As shown, the detection device 200 is also configured to send a sensor calibration instruction to the Juzi MCB control board 100 in response to a request for sensor calibration. The Juzi MCB control board 100 is also configured to receive the sensor calibration instruction sent by the detection device 200 and send a second control signal to the detection device 200. The detection device 200 is also configured to receive the second control signal sent by the Juzi MCB control board 100 and send the second control signal to the main control unit 300. The main control unit 300 is configured to receive the second control signal sent by the main control unit 300, and send a fourth target control instruction to the motion control unit 400 based on the second control signal to control the movement of the conveying track so that the calibration substrate carried by the conveying track is controlled to move back and forth on the conveying track. In this embodiment, the corresponding motor can be controlled to drive the conveying track to carry the calibration substrate to move back and forth in the conveying track to collect the motor movement pulse information when the calibration substrate triggers sensors at different positions.

[0070] The Matrix MCB control board 100 is also configured to obtain the number of motor movement pulses when the calibration substrate triggers sensors at different positions; and calculate the distance between each sensor based on the number of motor movement pulses and the size of the calibration substrate, and update the conversion relationship between the number of motor movement pulses and the distance, and send the conversion relationship to the detection device 200 to achieve sensor calibration.

[0071] Taking the length of the calibration substrate as a as an example, a first sensor and a second sensor are arranged in sequence along the conveying direction of the conveying track; the matrix MCB control board 100 is also configured to: obtain a first relationship between the number of pulses and the distance according to the length of the calibration substrate and the number of pulses of the motor from the time the first sensor is triggered to the time the trigger stops; obtain the relationship between the distance between the first sensor and the second sensor and the first number of pulses according to the first number of pulses of the motor when the standard calibration substrate moves from the first sensor to the second sensor and the first relationship.

[0072] More specifically, the sensor calibration instruction can be issued by triggering a corresponding button on the GUI page of the MCB control board 100, and the sensor position calibration can be performed as follows.

[0073] When doing specific operations, such as Figure 3 As shown, the calibration substrate is placed at the entrance of the conveying track of the detection device 200. After the detection device 200 sends a second control signal (board entry instruction), the main control unit 300 controls the motor to rotate according to the board entry instruction to drive the track to enter the board. At this time, the sensor at the entrance can sense that the calibration substrate has entered the track.

[0074] You can also trigger the reset command of the sensor spacing calibration on the GUI page of the MCB control board 100, record the X-direction length "a" of the calibration substrate used, and then trigger the corresponding button to trigger the sensor calibration command to control the motor to carry the calibration substrate on the track. The calibration substrate will travel back and forth in the track of the detection device 200. In this process, the sensors on the track will be triggered in sequence by the calibration substrate that walks back and forth (import → deceleration → stop plate → exit → stop plate → deceleration → import). When the rightmost end of the calibration substrate in progress triggers the deceleration sensor (i.e., the moment the sensor lights up), the pulse value at this point is recorded; the calibration substrate continues to move forward, and when its rightmost end triggers the stop plate sensor, the motor forward pulse value X1 (unit: steps) from the deceleration sensor to the stop plate sensor can be obtained; the calibration substrate continues to move forward, and when its rightmost end triggers the exit sensor, the motor forward pulse value X2 (unit: steps) from the stop plate sensor to the exit sensor can be obtained; the calibration substrate turns back, and its leftmost end triggers the deceleration sensor and the import sensor in turn, and the motor forward pulse value X3 (unit: steps) from the deceleration sensor to the import sensor can be obtained. In this way, combined with the X-direction length "a" value (unit: mm) of the calibration substrate, the distance relationship between the sensors and the conversion relationship between pulse and distance can be calculated. Assume that the X-direction length "a" of a calibration substrate is 100 mm. During the period from the moment the deceleration sensor is triggered to the moment it stops triggering, the number of pulses of the motor is 3509 steps. The conversion relationship between pulses and distance is as follows:

[0075] That is, the conversion relationship between pulse and distance can be:

[0076] The distance corresponding to each pulse number of the motor = 100 / 3509 = 0.028498 (unit: mm / steps);

[0077] Or the number of pulses required per millimeter distance = 3509 / 100 = 35.09 (unit: steps / mm);

[0078] The distance relationship between sensors is similar:

[0079] The distance from the deceleration sensor to the stop sensor: X1 (steps) × 0.028498 (mm / steps);

[0080] The distance from the stop sensor to the exit sensor: X2 (steps) × 0.028498 (mm / steps);

[0081] The distance from the inlet sensor to the deceleration sensor is: X3 (steps) × 0.028498 (mm / steps).

[0082] The above calculations can determine the conversion relationship between pulse and distance and the distance relationship between sensors. In this way, when the sensor position of the detection device 200 changes or the flow direction is switched, the sensor spacing needs to be recalibrated.

[0083] The present invention also provides an adaptive control method based on the MCB control board 100, which is applied to the MCB control board 100. Figure 2 A schematic diagram of an adaptive control method based on the matrix MCB control board 100 is shown.

[0084] like Figure 2 As shown, in step S2, in response to receiving the adaptive instruction sent by the detection device 200, based on the adaptive instruction, a first control signal is sent to the detection device 200, so that the detection device 200 sends a first target control instruction to the motion control unit 400 through the main control unit 300 based on the first control signal, so as to control the conveying track to move so that the carried object carried by the conveying track moves to a preset position;

[0085] In step S7, the position information of the carried object is obtained, and the deviation value is calculated according to the position information;

[0086] In step S8, an adjustment control signal is sent to the detection device 200 according to the deviation value, so that the detection device 200 sends a second target control instruction or a third target control instruction to the motion control unit 400 through the main control unit 300 based on the adjustment control signal to control the movement of the conveying track and adjust the position of the carrier until the calculated deviation value tends to be stable, and the deviation value is the difference between the current position of the carrier on the conveying track and the preset position. According to different adjustment control signals and corresponding deviation values, the reinforcement learning model is used for training until the deviation value converges to obtain the optimal strategy, wherein the state of the reinforcement learning model includes the current in-place information and the current adjustment control signal, and the action space includes forward and backward.

[0087] Specifically, when the carrier does not reach the preset position, an adjustment control signal for moving the carrier in the forward direction is sent to the detection device 200, so that the detection device 200 issues a second target control instruction for moving the conveying track in the forward direction through the main control unit 300; when the carrier exceeds the preset position, an adjustment control signal for moving the carrier in the reverse direction is sent to the detection device 200, so that the detection device 200 issues a third target control instruction for moving the conveying track in the reverse direction through the main control unit 300.

[0088] Obviously, the embodiments described above are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, ordinary technicians in this field can make other different forms of changes or modifications without creative work, which should fall within the scope of protection of the present invention.

Claims

1. An adaptive control system based on the MCB control board, characterized in that: It includes MCB control board, detection equipment, main control unit, and motion control unit; among which, The detection device is configured to send an adaptation instruction to the Matrix MCB control board in response to a request for adaptation; After receiving the adaptive instruction, the Juzi MCB control board sends a first control signal to the detection device based on the adaptive instruction; The detection device is also configured to receive a first control signal sent by the Juzi MCB control board, and send the first control signal to the main control unit; The main control unit is configured to send a first target control instruction to the motion control unit based on the first control signal, so as to control the conveying track to move the carried object on the conveying track to a preset position; The motion control unit is configured to receive the first target control instruction sent by the main control unit to control the operation of the corresponding motor to control the movement of the conveying track; The MCB control board is further configured to obtain the in-place information of the carrier, calculate the deviation value according to the in-place information, and send an adjustment control signal to the detection device according to the deviation value until the calculated deviation value tends to be stable, and the deviation value is the difference between the current position of the carrier on the conveying track and the preset position; The detection device also includes receiving the adjustment control signal and sending the adjustment control signal to a main control unit; the main control unit is also configured to send a second target control instruction or a third target control instruction to the motion control unit based on the adjustment control signal to control the movement of the conveying track and adjust the position of the carrier.

2. The adaptive control system based on the matrix MCB control board according to claim 1, characterized in that: The Matrix MCB control board is also configured as: When the carried object does not reach the preset position, an adjustment control signal for moving the carried object in a forward direction is sent to the detection device, so that the detection device sends a second target control instruction for moving the conveying track in a forward direction through the main control unit; When the carried object exceeds the preset position, an adjustment control signal for causing the carried object to move in the reverse direction is sent to the detection device, so that the detection device sends a third target control instruction for causing the conveying track to move in the reverse direction through the main control unit.

3. The adaptive control system based on the matrix MCB control board as claimed in claim 2, characterized in that: The Matrix MCB control board is also configured as: According to the current adjustment control signal and the corresponding deviation value, based on the trained reinforcement learning model, the next adjustment control signal is output.

4. The adaptive control system based on the matrix MCB control board as claimed in claim 3 is characterized in that: The Matrix MCB control board is also configured as: According to different adjustment control signals and corresponding deviation values, the reinforcement learning model is used for training until the deviation value converges to obtain the optimal strategy. The state of the reinforcement learning model includes the current arrival information and the current adjustment control signal, and the action space includes forward and backward actions.

5. The adaptive control system based on the matrix MCB control board as claimed in claim 4, characterized in that: According to different first control signals and corresponding deviation values, the reinforcement learning model is used to perform multiple training until the deviation value converges to obtain the optimal strategy. In the learning process of each step of the reinforcement learning model training, the Kalman filter is used to predict the state and error covariance of the next moment, and the new observation data is used to correct the error and update the state estimate. The state estimate updated by the Kalman filter is used to adjust the model parameters.

6. The adaptive control system based on the matrix MCB control board as claimed in claim 1, characterized in that: The detection device is also configured to send a sensor calibration instruction to the Matrix MCB control board in response to a request for sensor calibration; The Juzi MCB control board is also configured to receive a sensor calibration instruction sent by the detection device and send a second control signal to the detection device; The detection device is also configured to receive a second control signal sent by the Juzi MCB control board, and send the second control signal to the main control unit; The main control unit is configured to receive a second control signal sent by the main control unit, and send a fourth target control instruction to the motion control unit based on the second control signal, so as to control the movement of the conveying track so that the calibration substrate carried by the conveying track is controlled to move back and forth on the conveying track; The Matrix MCB control board is also configured to obtain the number of motor movement pulses when the calibration substrate triggers sensors at different positions; and calculate the distance between each sensor based on the number of motor movement pulses and the size of the calibration substrate, and update the conversion relationship between the number of motor movement pulses and the distance, and send the conversion relationship to the detection device.

7. The adaptive control system based on the matrix MCB control board as claimed in claim 6, characterized in that: The length of the calibration substrate is a, and a first sensor and a second sensor are arranged in sequence along the conveying direction of the conveying track; The Matrix MCB control board is also configured as: According to the length of the calibration substrate and the number of pulses of the motor from the time when the first sensor is triggered to the time when the trigger stops, a first relational expression for converting the number of pulses to the distance is obtained; According to the first pulse number of the motor when the standard calibration substrate moves from the first sensor to the second sensor and a first relationship, the relationship between the distance from the first sensor to the second sensor and the first pulse number is obtained.

8. An adaptive control method based on a Matrix Control Board (MCB) is applied to a Matrix Control Board, characterized in that: include: In response to receiving an adaptive instruction sent by the detection device, based on the adaptive instruction, a first control signal is sent to the detection device, so that the detection device sends a first target control instruction to the motion control unit through the main control unit based on the first control signal, so as to control the conveying track to move the carried object carried by the conveying track to a preset position; Obtaining the location information of the carried object, and calculating the deviation value according to the location information; An adjustment control signal is sent to the detection device according to the deviation value, so that the detection device sends a second target control instruction or a third target control instruction to the motion control unit through the main control unit based on the adjustment control signal to control the movement of the conveying track and adjust the position of the carrier until the calculated deviation value tends to be stable, and the deviation value is the difference between the current position of the carrier on the conveying track and the preset position.

9. The adaptive control method based on the matrix MCB control board as claimed in claim 8, characterized in that: The step of sending an adjustment control signal to a detection device according to the deviation value comprises: When the carried object does not reach the preset position, an adjustment control signal for moving the carried object in a forward direction is sent to the detection device, so that the detection device sends a second target control instruction for moving the conveying track in a forward direction through the main control unit; When the carried object exceeds the preset position, an adjustment control signal for causing the carried object to move in the reverse direction is sent to the detection device, so that the detection device sends a third target control instruction for causing the conveying track to move in the reverse direction through the main control unit.

10. The adaptive control method based on the matrix MCB control board according to claim 8, characterized in that: In the step of sending the adjustment control signal to the detection device according to the deviation value, the method for obtaining the adjustment control signal includes: According to different adjustment control signals and corresponding deviation values, the reinforcement learning model is used for training until the deviation value converges to obtain the optimal strategy, wherein the state of the reinforcement learning model includes the current arrival information and the current adjustment control signal, and the action space includes forward and backward.

Images