A Design Method for a Non-Blocking Controller of an Automatic Manufacturing System

Abstract

The present disclosure relates to a design method for a non-blocking controller of an automatic manufacturing system, including: constructing an automaton model and an observer corresponding to the automatic manufacturing system; finding an illegal state set in the observer; iteratively prohibiting controllable events and iteratively updating the state information of the observer, and introducing static information, and finally obtaining a non-blocking controller that removes the illegal state set. The design method for the non-blocking controller of the automatic manufacturing system proposed by the present disclosure ensures the non-blocking property of the controlled system and improves the permitted behavior of the controlled system.

Description

Technical Field

[0001] The present disclosure relates to the technical field of automatic manufacturing system control, and particularly to a non-blocking controller design method for an automatic manufacturing system. Background Art

[0002] An automatic manufacturing system refers to a system that processes raw materials into parts or assembles parts into products with less direct or indirect manual intervention, and realizes the automation of the management process and the process process during the processing.

[0003] However, in an automatic manufacturing system, traditional controller design methods may make control decisions that cause the controlled system to deadlock in a certain state and cannot execute other events. In this case, the behavior of the automatic manufacturing system is severely restricted and cannot complete the production task according to the preset requirements. Therefore, it is necessary to improve one or more problems existing in the above-mentioned related technical solutions to improve the behavior permissibility of the manufacturing system and enable the system to reach the preset legal termination state.

[0004] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and thus may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the embodiments of the present disclosure is to provide a non-blocking controller design method for an automatic manufacturing system, which ensures the non-blocking property of the controlled system and improves the permitted behavior of the controlled system.

[0006] The embodiments of the present disclosure provide a non-blocking controller design method for an automatic manufacturing system, and the method includes the following steps:

[0007] Construct an automaton model and an observer corresponding to the automatic manufacturing system;

[0008] Find the set of illegal states in the observer;

[0009] Finally, obtain a non-blocking controller that removes the set of illegal states by iteratively prohibiting controllable events, iteratively updating the state information of the observer, and introducing static information.

[0010] In an exemplary embodiment of the present disclosure, the step of constructing an automaton model and an observer corresponding to the automatic manufacturing system includes:

[0011] Construct an automaton model corresponding to the automatic manufacturing system;

[0012] Analyze the non-blocking property of the automatic manufacturing system. If the automatic manufacturing system does not meet the non-blocking property, construct an observer of the automatic manufacturing system.

[0013] In an exemplary embodiment of the present disclosure, constructing the automaton model corresponding to the automatic manufacturing system includes: G = (X, E, δ, x0, X m ), where X represents the set of states of the automatic manufacturing system; E represents the set of all events; δ represents the transition function, δ: X × E → X; x0 represents the initial state, x0 ∈ X; X m is the set of legal termination states.

[0014] In an exemplary embodiment of the present disclosure, the event set E is divided into an observable event set E o and an unobservable event set E uo , that is, E = E o ∪ E uo .

[0015] In an exemplary embodiment of the present disclosure, the steps of constructing the observer include:

[0016] Constructing the observer of the automatic manufacturing system includes G ob = (Y, E o , δ ob , y0), where Y represents the set of states of the observer, E o represents the set of observable events; δ ob represents the transition function of the observer, y ∈ Y; y0 represents the initial state of the observer, y0 = UR(x0); UR(x) represents the set of all states that the automatic manufacturing system can reach after passing through a certain sequence of unobservable events σ when in state x, R e (x) represents the state set that the automatic manufacturing system reaches after passing through the observable event e when in state x, e ∈ E o , R e (x) = {x'|δ(x, e) = x'}; for an event set E uo , it is defined that is all finite-length strings composed of events in E uo , including the empty string ε.

[0017] In an exemplary embodiment of the present disclosure, in the step of finding the set of illegal states in the observer, the illegal state is defined as if the current state y of the observer contains a certain current state x of the automatic manufacturing system and violates the control requirement, then y is an illegal state, and all the illegal states in the automatic manufacturing system constitute the set of illegal states; where Y represents the set of states of the observer, y ∈ Y; X represents the set of states of the automatic manufacturing system, x ∈ X.

[0018] In an exemplary embodiment of the present disclosure, the steps of obtaining a non-blocking controller that removes the illegal state set by iteratively prohibiting controllable events and iteratively updating the state information of the observer, and introducing static information, include:

[0019] The controller prohibits the current controllable event to remove the current illegal state in the observer.

[0020] Update the state information of the observer and introduce static information to obtain an updated observer.

[0021] Judge the illegal state situation in the updated observer to obtain a non-blocking controller that removes the illegal state set.

[0022] In an exemplary embodiment of the present disclosure, the steps of the controller prohibiting the current controllable event to remove the current illegal state in the observer include: the controller observes an event sequence, and then generates a control decision S, and removes the illegal state of the observer through the control decision S.

[0023] In an exemplary embodiment of the present disclosure, the static information includes a static event q, and the static event q is an observable but uncontrollable event.

[0024] In an exemplary embodiment of the present disclosure, the steps of judging whether the updated observer is the final non-blocking controller include:

[0025] When the observer does not contain any illegal states, the non-blocking controller is obtained.

[0026] When the observer still contains illegal states, iteratively perform the steps of removing the current illegal state and updating the observer until a non-blocking controller that removes the illegal state set is obtained.

[0027] The technical solution provided by the present disclosure may include the following beneficial effects:

[0028] In the embodiment of the present disclosure, a design method for a non-blocking controller of an automatic manufacturing system is proposed. By iteratively prohibiting controllable events and iteratively updating the state information of the observer, and introducing static information, all illegal state sets in the observer can be removed. By introducing observable static information, the observer can be updated to a new state, including all deadlock states that the automatic manufacturing system may be in at this time. When the observer state is updated, the controller makes a new control decision, enabling the previously prohibited events to be reactivated, and the subsequent generated states are also added to the observer. In this way, through continuous iteration, all illegal states in the observer are removed, thereby overcoming the problem that the behavior of the controlled system is severely restricted by the existing controller, ensuring the non-blocking property of the controlled system, and improving the permitted behavior of the controlled system. Brief Description of the Drawings

[0029] The accompanying drawings herein are incorporated into and constitute a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0030] Figure 1 Showing a schematic structural diagram of an automatic manufacturing system in an exemplary embodiment of the present disclosure;

[0031] Figure 2 Showing a schematic diagram of an automaton model corresponding to the automatic manufacturing system in an exemplary embodiment of the present disclosure;

[0032] Figure 3 Showing a schematic diagram of the steps of a non-blocking controller design method for an automatic manufacturing system in an exemplary embodiment of the present disclosure;

[0033] Figure 4 Showing a schematic structural diagram of an observer in an exemplary embodiment of the present disclosure;

[0034] Figure 5 Showing a schematic structural diagram of the observer after removing the current illegal state in an exemplary embodiment of the present disclosure;

[0035] Figure 6 Showing a schematic structural diagram of the updated observer in an exemplary embodiment of the present disclosure. Detailed Embodiments

[0036] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more complete and comprehensive, and will fully convey the concept of the example embodiments to those skilled in the art. The features, structures, or characteristics described can be combined in any suitable manner in one or more embodiments.

[0037] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0038] In this example implementation, a method for designing a non-blocking controller for an automatic manufacturing system is provided. The method may include the following steps:

[0039] Step S101: constructing an automaton model and an observer corresponding to the automatic manufacturing system;

[0040] Step S102: Find the illegal state set in the observer;

[0041] Step S103: by iteratively prohibiting controllable events and iteratively updating the state information of the observer, and introducing static information, a non-blocking controller is finally obtained that removes the illegal state set.

[0042] A non-blocking controller design method for an automatic manufacturing system is proposed in this embodiment. The non-blocking property of the controlled system can be guaranteed, and the static information of the automatic manufacturing system is taken into consideration in the design method of the controller. Specifically, the time for the system to execute each event is upper bounded. When the automatic manufacturing system does not have any output information for a long time, the controller can infer that the automatic manufacturing system is in a deadlock state, and then update the state estimate. In this case, the controller will make a new control strategy to reactivate some previously prohibited events.

[0043] The following is a more detailed description of each step of the above method in this exemplary embodiment.

[0044] In this example embodiment, referring to Figure 1 As shown, the example is a small manufacturing system, which includes three workstations, a handling robot and two storage stations. The robot can move between the workstations and the storage station on the rails to carry the processed parts. At the beginning, the robot leaves workstation 1 and travels on rail 1 or rail 2. When on rail 1, the robot takes type A parts from storage station 1 and sends them to workstation 2 or workstation 3 for processing; after the processing is completed, the robot sends the type A parts to storage station 1. When on rail 2, the robot takes type B parts from storage station 2 and sends them to workstation 3 for processing; after the processing is completed, the robot sends the type B parts to storage station 2. At this point, the robot completes all tasks and returns to workstation 1. The entire workflow can be repeated here.

[0045] In this small manufacturing system, there are two control requirements: first, type A parts can only be processed at workstation 2; second, the robot can always return to workstation 1 after completing a task. First, we use the automaton model to model this small manufacturing system. Figure 2 As shown, Figure 2Twelve states of the automaton model are shown, namely states 0 to 11. The robot starts at workstation 1. Event a represents the detection of the robot leaving workstation 1, and events u1 and u2 represent the robot running on rail 1 or rail 2 respectively.

[0046] When on rail 1, the robot picks up a part of type A from storage station 1. Its action can be detected by a sensor and is modeled as event b. After picking up a part of type A, if the robot sends this part of type A to processing station 2, its action is represented by event c. If the robot sends this part of type A to processing station 3, its action is represented by event d. After the processing of this type A part is completed, the robot sends it to storage station 2 (event b) and returns to workstation 1 (event a).

[0047] On rail 2, the robot picks up a part of type B from storage station 2 (event b) and sends it to processing station 3 for processing (event d). Similarly, after the processing is completed, the robot sends the part to storage station 2 (event b) and returns to workstation 1 (event a).

[0048] In this small manufacturing system, the set of observable events that can be detected by sensors is E o = {a, b, c, d}, and the set of controllable events is E c = {a, c, d}. The robot ultimately needs to return to workstation 1. Therefore, state 0 is a legal termination state, that is, X m = {0}.

[0049] In this problem, there are two control requirements: First, parts of type A can only be processed at workstation 2, so state 7 is an illegal state, and the controller should ensure that the controlled system cannot reach state 7; Second, the robot can always return to workstation 1, that is, the system can ultimately return to state 0, which requires that the controlled system is non-blocking.

[0050] For the above problem, in the general control method, in order to prevent the system from reaching state 7, when the controller observes the observable event sequence w = ab, it is necessary to make the control strategy S(ab) = {a, b, c, u1, u2}, that is, the controllable event d is prohibited. At this time, the system may be locked in state 4 or state 5. When the system is in state 4, it can reach the legal termination state 0; but if the system deadlocks in state 5, it cannot reach the legal termination state. Therefore, the controller needs to make a more restrictive control decision, that is, to prohibit event a at the starting position, S(ε) = {b, c, d, u1, u2}. In this case, the controlled system meets the control requirements. However, under the action of the controller, the controlled system can only stay in state 0 and cannot execute any other events, and the behavior of this system is severely restricted.

[0051] Therefore, in this embodiment, a non-blocking controller design method for an automatic manufacturing system is proposed. This method is based on the static characteristics of the system, where the static characteristics of the system refer to the situation that the system has no output information for a long time. For a system, the upper limit time for the system to execute each event can be known in advance during the system modeling phase, that is, the longest time for the system to stay in each state is known in advance. For a system that does not contain a loop formed by unobservable events, if it has no output information for a long time, the controller can infer that the system is currently in a certain deadlock state. Therefore, the controller can update the corresponding state estimation, including the possible deadlock states that the system may be in currently. Since the control strategy of the controller is closely related to the state estimation, when the state estimation is updated, the controller can make a new control strategy, and the previously prohibited system behaviors can be reactivated. Therefore, the controlled system will have higher permitted behaviors.

[0052] Therefore, referring to Figure 3 as shown, in step S101 of this embodiment, the system is first modeled as a finite-state automaton model, that is, G=(X, E, δ, x0, X m ), where X represents the set of states of the automatic manufacturing system; E represents the set of all events; δ represents the transition function, δ: X×E→X; x0 represents the initial state, x0∈X; X m is the set of termination states. Here, the set of all events E can be divided into the set of observable events E o and the set of unobservable events E uo , that is, E = E o ∪E uo . The observable events of the automatic manufacturing system refer to the corresponding behaviors of the automatic manufacturing system that can be observed by sensors. The mapping function P: E * →E o * can be defined as: for any e∈E, σ∈E * , P(σe)=P(σ)P(e). If e∈E o , then P(e)=e; conversely, if e∈E uo , then P(e)=ε. This indicates that if the event e∈E o , under the action of the mapping P, it can be observed by external observers, otherwise external observers can only observe that the automatic manufacturing system has occurred an empty event string ε through the mapping P.

[0053] Generally, for a set of events E, define E * as all finite-length strings composed of events in E, including the empty string ε. For any state x∈X, the formula δ(x, ε)=x always holds. The transition function δ can be extended from the form X×E→X to X×E *→X, i.e., (1) δ(x, ε) = x; (2) δ(x, eσ) = δ(δ(x, e), σ), where e ∈ E and σ ∈ E * . Similarly, for an event set E uo , define as all finite-length strings composed of events in E uo , including the empty string ε, and it can be defined in the same way.

[0054] Referring to Figure 4 as shown, construct the observer of the automatic manufacturing system as G ob = (Y, E o , δ ob , y0), where Y represents the state set of the observer, E o represents the observable event set; δ ob represents the transition function of the observer, y ∈ Y; y0 represents the initial state of the observer, y0 = UR(x0); UR(x) represents all the state sets that the automatic manufacturing system can reach after passing through an unobservable event sequence σ when in state x, R e (x) represents the state set that the automatic manufacturing system can reach after passing through an observable event e when in state x, e ∈ E o , R e (x) = {x' | δ(x, e) = x'}.

[0055] Next, in step S102, find the set of illegal states in the observer. An illegal state in the observer refers to a system state that contains a violation of the control requirements. Specifically, if the current state y of the observer contains a current state x of the automatic manufacturing system that violates the control requirements, then y is an illegal state, and all the illegal states in the system constitute the set of illegal states; where Y represents the state set of the observer, y ∈ Y; X represents the state set of the automatic manufacturing system, x ∈ X.

[0056] In this embodiment, there are two control requirements: First, type A parts can only be processed at workstation 2; Second, after the robot completes a task, it can always return to workstation 1. States 7 and 10 violate the control requirements (because their corresponding type A parts are processed at workstation 3), so the observer states y5 and y6 are illegal states.

[0057] In step S103, the steps of obtaining the non-blocking controller that removes the set of illegal states by iteratively prohibiting controllable events, iteratively updating the state information of the observer, and introducing static information include the following sub-steps:

[0058] Step S1031: The controller removes the current illegal state in the observer by prohibiting the current controllable event.

[0059] Step S1032: Update the state information of the observer and introduce the static information to obtain the updated observer.

[0060] Step S1033: Determine the illegal state condition of the updated observer, so as to obtain a non-blocking controller that removes the set of illegal states.

[0061] Specifically, in Step S1031, based on Step S102, the controller removes the illegal state in the observer by prohibiting the controllable event. It should be noted that when removing the illegal state, the legal behavior of the system should be retained as much as possible. In this embodiment, if the states y5 and y6 in the observer are to be removed, the controller can prohibit the controllable event d at state y2, that is, when the controller observes the time series ab, make the control decision S(ab) = {a, b, c, u1, u2}, and the observer after removing the illegal state is as Figure 5 shown.

[0062] In Step S1032, update the state information of the observer of the system and add the static information to obtain the updated observer. Since in Step S1031, the controller prohibited some controllable events, it may cause the system to deadlock. And when the system has no output for a long time, it can be inferred that the system has reached a certain deadlock state. Therefore, a static event q can be introduced to simulate the static information of the system. It should be noted that the static event q is an observable but uncontrollable event. By observing the static event, the observer of the system is updated to a new state, including all possible deadlock states of the system at this time. After the observer state is updated, the controller makes a new control decision, and the previously prohibited events are reactivated, and the newly generated subsequent states should also be added to the observer.

[0063] In this embodiment, the controller prohibited the controllable event d at state y2, and at this time the system state 5 becomes a deadlock state (since only event d occurs at state 5). Therefore, at state y2, by observing the static event q, the observer is updated to a new state y7 = {5}. At state y7, the controller reactivates event d, and the newly generated subsequent states should also be updated to the observer. The new observer structure is as Figure 6 shown.

[0064] In Step S1033, determine the illegal state condition of the updated observer, so as to obtain a non-blocking controller that removes the set of illegal states. This step includes two cases:

[0065] When the observer does not contain any illegal states, a non-blocking controller is obtained;

[0066] When the observer still contains illegal states, iteratively perform the steps of removing the current illegal states and updating the observer until a non-blocking controller that removes the set of illegal states is obtained.

[0067] That is, the controller reactivates an event that was previously prohibited, and the newly generated system behavior may still violate the control requirements. Therefore, it is necessary to determine whether the updated observer still contains illegal states. If the updated observer does not contain any illegal states at this time, it is the non-blocking controller to be finally obtained; otherwise, it is necessary to iteratively repeat step S1031 and step S1032 to continue removing the illegal states in the observer and updating the observer until the observer does not contain illegal states, thereby obtaining the final non-blocking controller.

[0068] In this embodiment, the control requirements of the system are as follows: First, type A parts can only be processed at workstation 2; Second, the robot can always return to workstation 1 (i.e., the system is non-blocking). Figure 6 The updated observer is shown, which does not contain any illegal states that violate the control requirements. Therefore, it can be used as the final non-blocking controller.

[0069] When the controller S=(Y, E o , δ s , y0) is obtained, the corresponding control strategy can be obtained according to the controller. For a certain state y in the observer, define E(y) as the set of controllable events that are defined under the system state x contained in y, that is In addition, define D(y) as the set of controllable events prohibited by the controller at state y, that is D(y)=E(y)\{e∈E c |δ s (y, e)=y'}. When the controller observes a certain observable sequence w, assuming that its corresponding state in the controller S is y, then the control strategy S(w)=E\D(y). For example, in this example, when the observable sequence w = ab is observed, its corresponding control strategy is S(w)={a, b, c, u1, u2, q}, that is, the controllable event d is prohibited by the controller.

[0070] In summary, the non-blocking controller design method for the automatic manufacturing system proposed in the embodiments of the present disclosure constructs an automaton model and an observer of the system, and makes corresponding control strategies through the controller to prohibit the system from entering illegal states and ensure that the controlled system can finally reach the preset legal termination state to complete the production task. When the controller prohibits a controllable event, it needs to be updated again, adding static information to obtain a new observer state. In the newly generated observer state, the controller reactivates the events that were previously prohibited. Finally, it is necessary to iteratively update the observer state to find all illegal states until the newly generated observer does not contain any illegal states, that is, a non-blocking controller is obtained. In the traditional controller design method, when the controller prohibits a controllable event and the system is in a deadlock state, static information and the state observer will not be considered at this time, and the system will remain in the deadlock state forever. In this embodiment, if the traditional method is used to design the controller, the controlled system can only stay at state 0 and cannot execute any other events, and the behavior of the system is severely restricted. Therefore, it can be known that the design method provided in this embodiment greatly improves the permitted behavior of the controlled system.

[0071] It should be noted that although the steps of the methods in the present disclosure are described in a specific order in the drawings, this does not require or imply that these steps must be executed in that specific order, or that all the steps shown must be executed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc. Also, it is easily understood that these steps may be executed synchronously or asynchronously, for example, in multiple modules / processes / threads.

[0072] It should be noted that although several units of the system for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the present disclosure. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0073] Those skilled in the art will readily conceive of other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure, which follow the general principles of the present disclosure and include known common knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and examples are only regarded as exemplary, and the true scope and spirit of the present disclosure are pointed out by the appended claims.

Claims

1. A non-blocking controller design method for an automatic manufacturing system, characterized in that, It includes the following steps: Construct an automaton model and an observer corresponding to the automatic manufacturing system, including: Construct the automaton model corresponding to the automatic manufacturing system; Analyze the non-blocking property of the automatic manufacturing system. If the automatic manufacturing system does not meet the non-blocking property, then construct the observer of the automatic manufacturing system; Find the set of illegal states in the observer; By iteratively prohibiting controllable events and iteratively updating the state information of the observer, and introducing static information, finally obtain a non-blocking controller that removes the set of illegal states, including: The controller removes the current illegal state in the observer by prohibiting the current controllable event; Update the state information of the observer and introduce static information to obtain an updated observer; Judge the illegal state situation of the updated observer, so as to obtain a non-blocking controller that removes the set of illegal states; Among them, the observer for constructing the automatic manufacturing system includes G ob =(Y, E o , δ ob , y0), where Y represents the state set of the observer, E o represents the set of observable events; δ ob represents the transition function of the observer, y ∈ Y; y0 represents the initial state of the observer, y0 = UR(x0); UR(x) represents all the state sets that the automatic manufacturing system can reach after passing through a certain sequence of unobservable events σ when in state x, R e (x) represents the state set that the automatic manufacturing system reaches after passing through the observable event e when in state x, e ∈ E o , R e (x) = {x'|δ(x, e) = x'}; for an event set E uo , it is defined that is all finite-length strings composed of events in E uo , including the empty string ε.

2. The non-blocking controller design method according to claim 1, characterized in that, Building the automaton model corresponding to the automatic manufacturing system includes: G = (X, E, δ, x0, X m ), where X represents the set of states of the automatic manufacturing system; E represents the set of all events; δ represents the transition function, δ: X × E → X; x0 represents the initial state, x0 ∈ X; X m is the set of termination states.

3. The non-blocking controller design method according to claim 2, characterized in that, The event set E is partitioned into an observable event set E o and an unobservable event set E uo , that is, E = E o ∪E uo .

4. The non-blocking controller design method according to claim 1, wherein In the step of finding the set of illegal states in the observer, the illegal state is defined as follows: if the current state y of the observer contains the current state x of a certain automatic manufacturing system and violates the control requirement, then y is an illegal state, and all the illegal states in the automatic manufacturing system constitute the set of illegal states; where, Y represents the set of states of the observer, y ∈ Y; X represents the set of states of the automatic manufacturing system, x ∈ X.

5. The non-blocking controller design method according to claim 1, characterized in that The step that the controller removes the current illegal state in the observer by prohibiting the current controllable event includes: the controller observes the event sequence, and then generates a control decision S, and removes the illegal state of the observer through the control decision S.

6. The non-blocking controller design method according to claim 1, wherein The static information includes a static event q, and the static event q is an observable but uncontrollable event.

7. The non-blocking controller design method according to claim 1, characterized in that The step of judging whether the updated observer is the final non-blocking controller includes: When the observer does not contain any illegal states, then obtain the non-blocking controller; When the observer still contains illegal states, iteratively perform the steps of removing the current illegal state and updating the observer until a non-blocking controller that removes the set of illegal states is obtained.

Images