Information Processing Apparatus, Control Design Support Method, and Program

Abstract

Provided is an information processing device comprising: a first controller model acquisition unit that acquires a first controller model which contains information indicating a control condition based on a measured value and information indicating a control operation which specifies an operation to be controlled when the control condition is satisfied; and a second controller model output unit that outputs a second controller model which can tolerate a measured value including a measurement error, wherein the second controller model includes information that indicates a control condition based on the information indicating the control condition which is contained in the first controller model and information that indicates a control operation based on the information indicating the control condition and the control operation which are contained in the first controller model.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, a control design support method, and a program.

Background Art

[0002] Controllers have been developed to control a control target based on measurement results from sensors and the like. Therefore, techniques for guaranteeing the safety of a control target using a formal model are known.

[0003] For example, Non-Patent Document 1 discloses a method for calculating abnormal operations of a control target that a controller can withstand with respect to a controller model and a control target model expressed by "Labeled transition systems".

Prior Art Documents

Non-Patent Documents

[0004]

Non-Patent Document 1

Non-Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the conventional technology, when measurement errors are included in measurement results, it is necessary to evaluate the measurement errors as abnormal operations and construct a controller model through complex design work. Therefore, there is a problem that it is difficult to construct a controller model that can withstand measurement errors.

[0006] The disclosed technology aims to provide a controller model that can withstand measurement errors.

Means for Solving the Problem

[0007] The disclosed technology includes a first controller model acquisition unit that acquires a first controller model including information indicating control conditions based on measurement values and information indicating control operations that define the operations of a control target when the control conditions are satisfied, and a second controller model output unit that outputs a second controller model that can withstand measurement values including measurement errors. The second controller model includes information indicating control conditions based on the information indicating control conditions included in the first controller model and information indicating control operations based on the information indicating the control conditions and control operations included in the first controller model, and is an information processing apparatus.

Effects of the Invention

[0008] It is possible to provide a controller model that can withstand measurement errors.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Best Mode for Carrying Out the Invention

[0010] (First Embodiment) Hereinafter, embodiments (this embodiment) of the present invention will be described with reference to the drawings. The embodiments described below are merely examples, and the embodiments to which the present invention is applied are not limited to the following embodiments.

[0011] In the text of this specification, for convenience of description, "^" or " ~ " is attached in front of the character. "^t", " ~ t" is an example thereof.

[0012] FIG. 1 is a functional configuration diagram of an information processing apparatus.

[0013] The information processing apparatus 10 according to this embodiment outputs a second controller model 103 and robustness condition data 104 based on a first controller model 101 and error characteristic information 102.

[0014] The first controller model 101 is a formal model that defines the operation of a controller that controls a control target based on measurement results from a sensor or the like. The first controller model 101 is a model that does not consider measurement errors from a sensor or the like. It is assumed that the control conditions, parameter restrictions, and control operations described in the first controller model 101 satisfy the safety conditions described in the model.

[0015] The error characteristic information 102 is information indicating the characteristics of measurement errors. Specifically, the error characteristic information 102 is information indicating the relationship between the true value and the measured value of the measurement target, and is information indicating the degree of the measurement error. For example, an expression (t ∈ [^t - 3, ^t + 3]) representing the relationship between the actual temperature t of the measurement target and the measured value ^t is an example of the error characteristic information 102.

[0016] In this case, the error characteristic information 102 is information indicating the characteristic that the measurement error of the temperature t is within the range of ±3. Note that the error characteristic indicated by the error characteristic information 102 is not limited to a range of a certain value, and may be a range of a certain ratio, or may be a complex characteristic in which the error range varies depending on the magnitude of the measured value.

[0017] Also, the error characteristic information 102 may be such that the characteristic of the measurement error is unknown, and in that case, the error characteristic information 102 may not be provided.

[0018] The second controller model 103 is a controller model that takes into account the measurement error having the characteristic indicated by the error characteristic information 102 based on the first controller model 101. Note that when the characteristic of the measurement error is unknown, the second controller model 103 is generated in a format in which information indicating the characteristic of the measurement error can be added later.

[0019] The first controller model 101 and the second controller model 103 each include information indicating safety conditions, control conditions, parameter restrictions, and control operations. The safety conditions are the conditions of safety to be guaranteed, and specifically, are the conditions that the actual value to be measured should satisfy by the result of control. The safety conditions of the second controller model 103 are a copy of those described in the first controller model 101 as they are. The control conditions are the conditions under which each control operation is executed. The parameter restrictions are the conditions that the parameters included in the content of the control operation should satisfy. The parameter restrictions may be the conditions for the parameters to satisfy the safety conditions, or may be conditions stricter than the safety conditions within the range that satisfies the safety conditions.

[0020] The robustness condition data 104 is information indicating the robustness of the second controller model 103, and specifically, is data indicating the conditions that are the limit of the tolerable measurement error.

[0021] The information processing apparatus 10 includes a first controller model acquisition unit 11, a control condition generation unit 12, a parameter restriction generation unit 13, a control operation generation unit 14, a second controller model output unit 15, a robustness condition generation unit 16, and a robustness condition data output unit 17.

[0022] The first controller model acquisition unit 11 acquires a first controller model 101 and error characteristic information 102. For example, the first controller model acquisition unit 11 may receive an input from a user, or may receive it from another device via a communication network or the like. Further, the first controller model acquisition unit 11 may confirm whether the acquired first controller model 101 or error characteristic information 102 is appropriate based on a predefined standard, and acquire the first controller model 101 or error characteristic information 102 when it is appropriate.

[0023] The control condition generation unit 12 generates information indicating the control conditions of the second controller model 103 based on the first controller model 101 and the error characteristic information 102.

[0024] The parameter restriction generation unit 13 generates information indicating parameter restrictions for each of the control conditions generated by the control condition generation unit 12 of the second controller model 103 based on the first controller model 101 and the error characteristic information 102.

[0025] The control operation generation unit 14 generates information indicating control operations for each of the control conditions generated by the control condition generation unit 12 of the second controller model 103 based on the first controller model 101 and the error characteristic information 102.

[0026] The second controller model output unit 15 outputs a second controller model 103 including information indicating the control conditions of the generated second controller model 103, and information indicating parameter restrictions and control operations generated for the control conditions. For example, the second controller model output unit 15 may display the second controller model 103 on a display device, or may transmit it to another device via a communication network or the like.

[0027] Based on the control conditions of the generated second controller model 103 and the information indicating parameter restrictions, the robustness condition generation unit 16 generates robustness condition data 104 for the second controller model 103. When the error characteristic information 102 is not given, that is, when the characteristics of the error are unknown, the robustness condition generation unit 16 generates robustness condition data 104 indicating the conditions that are the limits of the measurable errors that can be tolerated.

[0028] Note that even when the error characteristic information 102 has already been given, the robustness condition generation unit 16 may generate the robustness condition data 104. In that case, by comparing the robustness condition data 104 with the given error characteristics, it is possible to determine whether the given error characteristics are measurable errors that the second controller model 103 can tolerate.

[0029] The robustness condition data output unit 17 outputs the generated robustness condition data 104. For example, the robustness condition data output unit 17 may display the robustness condition data 104 on a display device, or may transmit it to other devices via a communication network or the like.

[0030] FIG. 2 is a diagram showing an example of the first controller model.

[0031] The first controller model 101 includes a set v of variables representing the true state of the control target, a safety condition, and one or more control rules. Each control rule includes control conditions, parameter restrictions, and a control operation. The control conditions are represented by the variable v. The parameter restrictions are represented by v and p i and are represented by. The control operation is represented by v and p i and a set v′ of variables representing the state after control.

[0032] FIG. 3 is a diagram for explaining a specific example of the first controller model.

[0033] A specific example of the first controller model 101 shown in FIG. 3 is a model that measures the temperature of an unstable hot spring and defines the operation of a controller that warms or cools based on the measurement results. The set v of variables indicating the state of the control target includes one variable, the temperature t. The safety condition is that the temperature t maintains a temperature of 30 degrees or more and 40 degrees or less.

[0034] There are three control rules, the first case shown in FIG. 3 cold The control rule for use is defined such that when the control condition that the temperature t satisfies t < 30 is met, the control operation (t′ = t + p heat ) is executed. Here, the parameter p heat satisfies the parameter restriction (t + p heat ∈ [30, 40] ∧ 0 < p heat ).

[0035] The second case shown in FIG. 3 ok The control rule for use is defined such that when the control condition that the temperature t satisfies t ∈ [30, 40] is met, the control operation (t′ = t + p keep ) is executed. Here, the parameter p keep satisfies the parameter restriction (t + p keep ∈ [30, 40] ∧ p keep ∈ [-6, 6]).

[0036] The third case shown in FIG. 3 hot The control rule for use is defined such that when the control condition that the temperature t satisfies 40 < t is met, the control operation (t′ = t - p cool ) is executed. Here, the parameter p cool satisfies the parameter restriction (t - p cool ∈ [30, 40] ∧ 0 < p cool ).

[0037] Also, a specific example of the error characteristic information 102 is the error characteristic (t ∈ [^t - 3, ^t + 3]). This indicates that the measurement error of the temperature t is within the range of ±3.

[0038] FIG. 4 is a flowchart showing an example of the flow of control design support processing.

[0039] When the information processing apparatus 10 receives a user operation or the like and executes control design support processing, it acquires a first controller model 101 and error characteristic information 102 (step S101). The control condition generation unit 12 generates information indicating the control conditions of the second controller model 103 (step S102).

[0040] Next, the parameter limit generation unit 13 generates information indicating the parameter limits of the second controller model 103 (step S103). Subsequently, the control operation generation unit 14 generates information indicating the control operation of the second controller model 103 (step S104).

[0041] The information processing apparatus 10 outputs the second controller model 103 generated by the processing in steps S102 - S104 (step S105).

[0042] Next, the robustness condition generation unit 16 generates information indicating the robustness conditions of the second controller model 103 (step S106). The information processing apparatus 10 outputs the generated information indicating the robustness conditions.

[0043] FIG. 5 is a flowchart showing an example of the flow of control condition generation processing.

[0044] In step S102 of the control design support processing shown in FIG. 4, the control condition generation unit 12 executes control condition generation processing. The control condition generation unit 12 adds a variable ^v representing the measured value of v to the second controller model 103 (step S201). Hereinafter, the control conditions and parameter limits included in the second controller model 103 are defined by the variable ^v.

[0045] Next, the control condition generation unit 12 selects one from all subsets S of the control conditions {case1, ···, case n} of the first controller model 101 (step S202). Specifically, the subsets S are {case1}, ···, {case n,{case1,case2},{case1,case3},···,{case1,case n ,{case1,case2,case3},···,{case1,case2,···,case n} is as follows.

[0046] Next, the control condition generation unit 12 generates an expression indicating the condition that all possible values of the true value inferred from the measured value ^v ~ v are included in the selected subset S (step S203). However, the generated conditional expression is a condition that does not include the range included in other subsets S. In this way, the control condition generation unit 12 avoids generating an expression indicating a condition that is repeatedly included in a plurality of subsets S.

[0047] For example, in the first controller model 101 shown in FIG. 3, when the measured value ^t of the temperature t is 30 degrees, since the error range is ±3 degrees, the possible values of the true value ~ t is in the range of 27 degrees or more and 33 degrees or less.

[0048] The control condition generation unit 12 generates an expression indicating the condition that all possible values of the true value ~ t are all included in the selected subset S. For example, when the selected subset S is {case cold}}, the control condition generation unit 12 generates the expression "∀ ~ t∈[^t - 3, ^t + 3]. ~ t < 30".

[0049] Next, the control condition generation unit 12 determines whether processing has been performed for all subsets S of {case1, ···, case n} of the first controller model 101 (step S204). If the control condition generation unit 12 determines that processing has not been performed for any subset S of {case1, ···, case n} of the first controller model 101 (step S204: No), it returns to the process of step S202 and selects a subset S that has not yet been selected.

[0050] When the control condition generation unit 12 determines that the processing has been performed for all subsets S of {case1, ···, case n} (step S204: Yes), the control condition generation process ends.

[0051] As a result, for all subsets S of {case1, ···, case n}, an expression indicating the control condition of the second controller model 103 is generated.

[0052] Note that the control condition generation unit 12 may not include in the second controller model 103 the subsets S for which the conditions are not logically satisfied among the expressions indicating the generated control conditions. In this case, in the process of generating the expressions indicating the parameter restrictions and control operations described later, the parameter restriction generation unit 13 and the control operation generation unit 14 may omit the processing for the subset S.

[0053] For example, in the case of the first controller model 101 shown in FIG. 3, for the subset S of {case cold , case OK , case hot}, when all possible values ~t of the true value inferred from the measured value ^t satisfy the conditions included in the selected subset S, and there is no condition that does not include the ranges included in the other subsets ({case cold , case OK , case hot}\S), an expression indicating the control condition is obtained. However, it is impossible for the range from a certain value ^t to ±3 to span all of case cold , case OK , case hot . Therefore, the control condition generation unit 12, the parameter restriction generation unit 13, and the control operation generation unit 14 may omit the processing for the subset S.

[0054] FIG. 6 is a diagram for explaining a specific example of the generation result of the control condition.

[0055] For example, case {ok,hot} The formula indicating the control condition of the usage control rule is "∀ ~ t ∈ [^t - 3, ^t + 3].( ~ t ∈ [30, 40] ∨ 40 < ~ t) ∧ ¬∃ ~ t ∈ [^t - 3, ^t + 3]. ~ t < 30 ∧ ∃ ~ t ∈ [^t - 3, ^t + 3]. ~ t ∈ [30, 40] ∧ ∃ ~ t ∈ [^t - 3, ^t + 3]. 40 < ~ t".

[0056] Figure 7 is a flowchart showing an example of the flow of the control condition string generation process.

[0057] The control condition generation unit 12 executes a control condition string generation process for generating a string indicating the control condition in the process of step S203 of the control condition generation process shown in FIG. 5.

[0058] The control condition generation unit 12 adds a string of a formula with the meaning of "All values v that can be the true value inferred from the measured value ^v have the following properties" to the string indicating the control condition of the second controller model 103 (step S301). The result of the process in this step S301 is, in the example of FIG. 6, case ~ This corresponds to the string "∀ {ok,hot} t ∈ [^t - 3, ^t + 3]." included in the formula indicating the control condition of the usage control rule. ~

[0059] Next, the control condition generation unit 12 selects one element (case i ) from the selected subset S (step S302). For example, when the selected subset S is {case1, case4}, the elements are case1 and case4.

[0060] Next, the control condition generation unit 12 adds to the string indicating the control condition of the second controller model 103, "The control condition case of the first controller model 101 i ​(v) or the v in the formula that means "or" ~ Append the string obtained by replacing v to v (step S303). The result of the process in this step S303 is, in the example of FIG. 6, case {ok,hot} The " ~ t∈[30,40]∨" in the formula indicating the control condition of the use control rule. However, "or" is omitted when the process of this step S303 is finally executed.

[0061] Next, the control condition generation unit 12 determines whether processing has been performed for all elements of the selected subset S (step S304). If the control condition generation unit 12 determines that processing has not been performed for any subset S (step S304: No), it returns to the process of step S302 and selects an element that has not yet been selected from the elements of the selected subset S.

[0062] Also, when the control condition generation unit 12 determines that processing has been performed for all elements of the selected subset S (step S304: Yes), next, the control condition generation unit 12 further selects one element (case n ) from all cases {case1, ···, case i} (step S305).

[0063] Subsequently, when case i is not an element of S, append a string of the formula that means "negation" (step S306). This can explicitly describe that it is not included in the generated control condition.

[0064] Next, the control condition generation unit 12 appends a string of the formula that means "and there exists a value with the following property in the possible value of the true value inferred from the measured value ^v ~ v" to the string indicating the control condition of the second controller model 103 (step S307). The result of the process in this step S307 is, in the example of FIG. 6, case {ok,hot} The "∧∃ ~It corresponds to the character string "t∈[^t-3,^t+3].".

[0065] Next, the control condition generation unit 12 adds to the character string indicating the control condition of the second controller model 103 the character string obtained by replacing v in the expression having the meaning of "and in the control condition of the first control model case i (v) and" with v (step S308). The result of the process in this step S308 is, in the example of FIG. 6, case ~ It corresponds to the character string "t∈[30,40]∧" included in the expression indicating the control condition of the control rule for use. However, "and" is omitted when the process of this step S308 is finally executed. {ok,hot} in the expression indicating the control condition of the control rule for use ~ t∈[30,40]∧".

[0066] Next, the control condition generation unit 12 determines whether processing has been performed for all elements of {case1, ···, case n} (step S309). If the control condition generation unit 12 determines that processing has not been performed for any subset S (step S309: No), it returns to the process of step S305 and selects an element that has not been selected from {case1, ···, case n}.

[0067] Also, when the control condition generation unit 12 determines that processing has been performed for all elements of the selected subset S (step S309: Yes), it ends the control condition character string generation process.

[0068] In this way, the control condition generation unit 12 can mechanically generate the character string indicating the control condition included in the second controller model 103 using the character string indicating the control condition included in the first controller model 101.

[0069] FIG. 8 is a flowchart showing an example of the flow of the parameter restriction generation process according to the first embodiment.

[0070] In step S103 of the control design support process shown in FIG. 4, the parameter restriction generation unit 13 executes a parameter restriction generation process. The parameter restriction generation unit 13 selects one from all subsets S of the control conditions {case1, ···, case n} of the first controller model 101 (step S401).

[0071] Subsequently, the parameter restriction generation unit 13 generates an expression indicating a parameter restriction that satisfies both of the following (a) and (b) (step S402).

[0072] (a) Satisfies all the parameter restrictions of the first controller model 101 in each control condition included in the selected subset S.

[0073] (b) There is an operation common to the control operations of the first controller model 101 corresponding to all the control conditions included in the selected subset S.

[0074] For example, when the selected subset S is {case cold}, the parameter restriction generation unit 13 generates an expression "∀ ~ t∈[^t-3,^t+3].( ~ t+p heat ∈[30,40]∧0<p heat )".

[0075] Next, the parameter restriction generation unit 13 determines whether processing has been performed for all subsets S of {case1, ···, case n} of the first controller model 101 (step S403). If the parameter restriction generation unit 13 determines that processing has not been performed for any subset S of {case1, ···, case n} of the first controller model 101 (step S403: No), it returns to the process of step S401 and selects a subset S that has not yet been selected.

[0076] The parameter restriction generation unit 13 performs processing for all subsets S of {case1, ···, case nWhen it is determined that the processing has been performed for all subsets S of {} (step S403: Yes), the parameter restriction generation processing is terminated.

[0077] As a result, for all subsets S of {case1, ···, case n} of the first controller model 101, expressions indicating parameter restrictions of the second controller model 103 are generated.

[0078] FIG. 9 is a diagram for explaining a specific example of the generation result of parameter restrictions according to the first embodiment.

[0079] For example, the expression indicating the parameter restriction of the parameters (p {ok,hot} , p keep ) of the case cool -use control rule is "∀ ~ t ∈ [^t - 3, ^t + 3]. (( ~ t ∈ [30, 40] ⇒ (t + p keep ∈ [30, 40] ∧ p keep ∈ [-6, 6])) ∧ (40 < ~ t ⇒ (t - p cool ∈ [30, 40] ∧ 0 < p cool ))) ∧ {} ≠ ({t′|t′ = t + p keep} ∩ {t′|t′ = t - p cool})".

[0080] FIG. 10 is a flowchart showing an example of the flow of parameter restriction string generation processing according to the first embodiment.

[0081] In the process of step S402 of the parameter restriction generation process shown in FIG. 8, the parameter restriction generation unit 13 executes a parameter restriction string generation process for generating a string indicating a parameter restriction.

[0082] The parameter restriction generation unit 13 adds to the string indicating the parameter restriction of the second controller model 103, "a possible value of the true value inferred from the measured value ^v ~The result of the process in step S501 is the case {ok,hot} The control condition of the control rule is included in the formula "∀ ~ This is equivalent to the string "t∈[^t-3,^t+3]."

[0083] Next, the parameter restriction generator 13 selects one element (case i ) (step S502).

[0084] Next, the parameter restriction generating unit 13 adds “parameter case of the first controller model 101” to the character string indicating the parameter restriction of the second controller model 103. i (v,p i ) And v of the formula ~ The character string replaced with v is added (step S503). As a result of the process in step S503, in the example of FIG. {ok,hot} The control rule for the control condition is included in the formula "(t+p keep ∈[30,40]∧p keep ∈[-6,6])∧" is a string that is ok ,case hot} out of case ok However, the "and" is omitted if the process of step S503 is executed last.

[0085] Next, the parameter restriction generation unit 13 determines whether or not the process has been performed for all elements of the selected subset S (step S504). If the parameter restriction generation unit 13 determines that the process has not been performed for any of the subsets S (step S504: No), the parameter restriction generation unit 13 returns to the process of step S502 and selects an element that has not yet been selected from among the elements of the selected subset S.

[0086] Also, when the parameter restriction generation unit 13 determines that processing has been performed for all elements of the selected subset S (step S504: Yes), next, the parameter restriction generation unit 13 adds a string of an expression meaning "and all values that may be the true value inferred from the measured value ^v" to the string indicating the parameter restriction of the second controller model 103. ~ All v have the following properties. {ok,hot} In the example of FIG. 9, the result of the process in step S505 corresponds to the string "{} ≠ ({t′|t′ = t + p keep} ∩ {t′|t′ = t - p cool})" included in the expression indicating the control condition of the case use control rule.

[0087] Next, the parameter restriction generation unit 13 selects one element (case i ) from the selected subset S (step S506).

[0088] Subsequently, the parameter restriction generation unit 13 adds a string obtained by replacing v in the expression meaning "the set of states that can be reached as a result of control in the control operation case i of the first controller model 101" to the string indicating the parameter restriction of the second controller model 103. ~ (step S507).

[0089] Furthermore, the parameter restriction generation unit 13 adds a string of an expression meaning "the common part of the above set is not empty" to the string indicating the parameter restriction of the second controller model 103 (step S508). In the example of FIG. 9, the results of the processes in step S507 and step S508 correspond to the string "{} ≠ ({t′|t′ = t + p ok,hot} ∩ {t′|t′ = t - p keep})" included in the expression indicating the control condition of the case use control rule. cool

[0090] Next, the parameter restriction generation unit 13 determines whether processing has been performed for all elements of the selected subset S (step S509). If the parameter restriction generation unit 13 determines that processing has not been performed for any subset S (step S509: No), it returns to the processing of step S506 and selects an element that has not been selected from the selected subset S.

[0091] Also, when the parameter restriction generation unit 13 determines that processing has been performed for all elements of the selected subset S (step S509: Yes), it ends the parameter restriction string generation process.

[0092] In this way, the parameter restriction generation unit 13 can mechanically generate a character string indicating the parameter restriction included in the second controller model 103 using the character strings indicating the parameter restrictions, control operations, etc. included in the first controller model 101.

[0093] FIG. 11 is a flowchart showing an example of the flow of the control operation generation process according to the first embodiment.

[0094] In step S104 of the control design support process shown in FIG. 4, the control operation generation unit 14 executes the control operation generation process. The control operation generation unit 14 selects one from all subsets S of the control conditions {case1, ···, case n} of the first controller model 101 (step S601).

[0095] Subsequently, the control operation generation unit 14 generates an expression indicating an operation common to all control operations of the first controller model 101 in each control condition included in the selected subset S (step S602).

[0096] For example, when the selected subset S is {case cold}, the control operation generation unit 14 generates an expression "t′ = t + p heat ".

[0097] Next, the control operation generation unit 14 determines whether processing has been performed for all subsets S of {case1, ···, case n}(step S603). If the control operation generation unit 14 determines that processing has not been performed for any subset S of {case1, ···, case n}(step S603: No), the process returns to the process of step S601, and an unselected subset S is selected.

[0098] If the control operation generation unit 14 determines that processing has been performed for all subsets S of {case1, ···, case n}(step S603: Yes), the control operation generation process ends.

[0099] As a result, for all subsets S of {case1, ···, case n}, an expression indicating the control operation of the second controller model 103 is generated.

[0100] FIG. 12 is a diagram for explaining a specific example of the generation result of the control operation according to the first embodiment.

[0101] For example, the expression indicating the control operation of the case {ok,hot} use control rule is “(t′∈({t′|t′ = t + p keep}∩{t′|t′ = t - p cool})”.

[0102] FIG. 13 is a flowchart showing an example of the flow of the control operation character string generation process.

[0103] In the process of step S602 of the control operation generation process shown in FIG. 11, the control operation generation unit 14 executes a control operation character string generation process for generating a character string indicating the control operation.

[0104] The control operation generation unit 14 adds a character string of an expression meaning "the result that can be reached by this control operation is an element of the following set" to the character string indicating the control operation of the second controller model 103 (step S701). In the example of FIG. 12, case {ok,hot} corresponds to the character string "t'∈" included in the expression indicating the control condition of the control rule to be used.

[0105] Next, the control operation generation unit 14 adds a character string of an expression meaning "take the intersection of these" to the character string indicating the control operation of the second controller model 103 (step S702). Then, the control operation generation unit 14 selects one element (case i ) from the selected subset S (step S703).

[0106] Next, the control operation generation unit 14 adds an expression meaning "the set of the results of the control by the control operation case i of the first controller model" to the character string indicating the control operation of the second controller model 103 (step S704). In the example of FIG. 12, the character string "{t'|t' = t + p keep}" corresponds to the character string added for case ok among S = {case hot , case ok}.

[0107] Next, the control operation generation unit 14 determines whether processing has been performed for all elements of the selected subset S (step S705). If the control operation generation unit 14 determines that processing has not been performed for any subset S (step S705: No), it returns to the processing of step S703 and selects an element that has not yet been selected from the elements of the selected subset S.

[0108] Also, when the control operation generation unit 14 determines that processing has been performed for all elements of the selected subset S (step S705: Yes), it ends the control operation character string generation process.

[0109] In this way, the control operation generation unit 14 can mechanically generate the character string indicating the control operation included in the second controller model 103 by using the character string indicating the control operation included in the first controller model 101.

[0110] In addition, in this embodiment, when the error characteristic information 102 is known, the process of generating the control conditions, parameter restrictions, and the character string indicating the control operation of the second controller model 103 has been shown. However, the information processing apparatus 10 may perform these processes on the premise that the error characteristics are unknown. In that case, the information processing apparatus 10 is the possible value of the true value in the above-mentioned character string ~ The range of t ( ~ t ∈ [^t - 3, ^t + 3]) corresponding part can be added later in a form (for example, ~ t ∈ R(^t)) is generated. Here, R(^t) is the range of possible values of the true value for the measured value ^t.

[0111] FIG. 14 is a flowchart showing an example of the flow of the robustness condition generation process.

[0112] In step S106 of the control design support process shown in FIG. 4, the robustness condition generation unit 16 executes the robustness condition generation process. The robustness condition generation unit 16 selects one from all subsets S of the control conditions {case1, ···, case n}} of the first controller model 101 (step S801).

[0113] Next, for all measured values ^v that satisfy each control condition included in the second controller model 103 corresponding to the selected subset S, the robustness condition generation unit 16 generates an expression indicating the condition that there exists a parameter that satisfies the parameter restrictions of the second controller model 103 (step S802).

[0114] For example, when the selected subset S is {case cold}}, the robustness condition generation unit 16 generates "∀^t.((∀ ~ t ∈ R(^t). ~ t < 30) ⇒ (∃p heat .(∀~ t ∈ R(^t).( ~ t + p heat ∈ [30, 40] ∧ 0 < p heat ))))」 generates an expression of the form

[0115] Next, the robustness condition generation unit 16 determines whether processing has been performed for all subsets S of {case1, ···, case n} of the first controller model 101 (step S803). If the robustness condition generation unit 16 determines that processing has not been performed for any subset S of {case1, ···, case n} of the first controller model 101 (step S803: No), it returns to the processing of step S801 and selects a subset S that has not yet been selected.

[0116] If the robustness condition generation unit 16 determines that processing has been performed for all subsets S of {case1, ···, case n} of the first controller model 101 (step S803: Yes), it sets the condition that satisfies all the generated expressions as the robustness condition (step S804).

[0117] As a result, an expression indicating the robustness condition of the second controller model 103 is generated. For example, the expression indicating the robustness condition generated based on the first controller model 101 shown in FIG. 3 is as follows.

[0118] ∀^t.((∀ ~ t ∈ R(^t). ~ t < 30) ⇒ (∃p heat . (∀ ~ t ∈ R(^t). ( ~ t + p heat ∈ [30, 40] ∧ 0 < p heat )))) ∧ ··· ∀^t.((∀ ~ t ∈ R(^t). 40 < ~ t) ⇒ (∃p cool . (∀~ t∈R(^t). ( ~ tp cool ∈[30,40]∧0 <p cool ))))

[0119] In this case, R(^t) that satisfies the above-mentioned conditional expression is the range of measurement error that the second controller model 103 can tolerate.

[0120] FIG. 15 is a flowchart showing an example of the flow of the robustness condition string generation process.

[0121] In step S802 of the robustness condition generation process shown in FIG. 14, the robustness condition generation unit 16 executes robustness condition character string generation processing for generating a character string indicating a robustness condition.

[0122] The robustness condition generating unit 16 selects one element (case i ) is selected (step S901).

[0123] The robustness condition generating unit 16 adds the character string indicating the robustness condition of the second controller model 103 with “control condition case of the second controller model 103” i The result of the process in step S902 is that the selected subset S satisfies the following property: hot In this case, the expression "∀^t.((∀ ~ t∈R(^t).40< ~ t).

[0124] Next, the robustness condition generating unit 16 adds a character string of an equation meaning "there is a parameter that satisfies the following property" to the character string indicating the robustness condition of the second controller model 103 (step S903). hotWhen it is, it corresponds to the string "∃p" included in the expression indicating the robustness condition of the second controller model 103 generated based on the first controller model 101 shown in FIG. 3. cool ∈

[0125] Next, the robustness condition generation unit 16 adds a string of an expression having the meaning of "parameter restriction of the second controller model" to the string indicating the robustness condition of the second controller model 103 (step S904). The result of the process in this step S904 is that when the selected subset S is {case hot}, it corresponds to the string "(∀ ~ t∈R(^t).( ~ t - p cool ∈[30,40]∧0 < p cool ))" included in the expression indicating the robustness condition generated based on the first controller model 101 shown in FIG. 3.

[0126] Next, the robustness condition generation unit 16 determines whether processing has been performed for all elements of the selected subset S (step S905). If the robustness condition generation unit 16 determines that processing has not been performed for any subset S (step S905: No), it returns to the process of step S901 and selects an element that has not been selected from the selected subset S.

[0127] Also, when the robustness condition generation unit 16 determines that processing has been performed for all elements of the selected subset S (step S905: Yes), it ends the robustness condition string generation process.

[0128] In this way, the robustness condition generation unit 16 can mechanically generate a string indicating the robustness included in the robustness condition data 104, that is, the condition that is the limit of the measurable error that can be tolerated, using the strings indicating the control conditions, parameter restrictions, etc. generated as the second controller model 103.

[0129] Conventionally, it has been difficult to design and construct a formal model that can withstand measurement errors from scratch. This is because when measurement errors are incorporated, the elements, considerations, etc. of the model significantly increase, and it is necessary to mathematically prove that safety is satisfied for each of the increased elements.

[0130] On the other hand, according to the information processing apparatus 10 according to the present embodiment, a second controller model 103 is generated based on the first controller model 101 and the error characteristic information 102. Thereby, a controller model that can withstand measurement errors can be provided.

[0131] Further, the information processing apparatus 10 generates a character string that defines the operation of the second controller model 103 by mechanical processing using the character string that defines the operation of the first controller model 101. Thereby, a controller model that can withstand measurement errors can be easily constructed without relying on human skills.

[0132] Also, usually, the characteristics of measurement errors are unclear at the design stage and also depend on the environment of the controlled object. Therefore, it is difficult to analyze the limit of errors that the controller can withstand before constructing the controller model. Thus, there is a desire to analyze the limit of errors that the controller model can withstand and select a sensor or the like after constructing the controller model.

[0133] In response to such a desire, conventionally, it has been difficult to derive the limit of errors. On the other hand, according to the information processing apparatus 10 according to the present embodiment, the second controller model 103 is generated in a state where the characteristics of the errors are unknown, and robustness condition data 104 indicating the conditions that become the limit of the measurement errors that the second controller model 103 can withstand is generated. Thereby, it becomes easy to analyze the limit of errors that the controller model can withstand after constructing the controller model.

[0134] In addition, the information processing apparatus 10 according to the present embodiment mechanically generates a character string indicating the control conditions, parameter restrictions, and control operations of the second controller model 103 by connecting, as they are, a character string indicating the control conditions, parameter restrictions, and control operations of the first controller model 101, etc., using symbols such as "and", "or", "the following properties are satisfied for all...", and "there exists... that satisfies the following properties". Here, since the symbols used for connection are general ones that do not depend on a specific notation, the information processing apparatus 10 can make use of the notation of the first controller model 101 to be processed and can handle controller models not limited to a specific notation. For example, the information processing apparatus 10 is applicable even when the first controller model is described in natural language.

[0135] Note that the information processing apparatus 10 may be configured to store a part of the notation as a set value in order to absorb minute differences in notation. For example, in the present embodiment, an example in which the logical symbol indicating the meaning of "or" is "∨" is shown, but it may be stored as a set value and can be set, for example, as "or", "|", etc.

[0136] Next, as a specific example of the present embodiment, an example of a controller for an autonomous vehicle is shown as an example different from the above-described controller for the temperature of the hot spring.

[0137] There is a standard safety rule for autonomous vehicles called RSS (Responsibility-Sensitive Safety) (for example, Non-Patent Document 2).

[0138] In RSS, rules for which an autonomous vehicle is responsible for ensuring safety are defined for each situation. For example, RSS describes a rule that "an autonomous vehicle traveling in a situation where there is another vehicle ahead must maintain a minimum inter-vehicle distance d_min from the vehicle ahead", along with the definition of d_min.

[0139] As long as an autonomous vehicle follows the rules described in RSS, it can be considered to have fulfilled its responsibilities, and it can be seen that the cause and responsibility lie outside in the event of an accident. Also, if a control software S that reliably follows the rules described in RSS is constructed, it can be guaranteed that S will behave safely on RSS.

[0140] For example, in a model M of S, a safety condition such as "the operation of M needs to satisfy the inter-vehicle distance d_min from the vehicle in front" and a control rule consisting of a firing condition, parameter limit, and control operation such as "when the distance from the preceding vehicle is d, d = d_min + δ, and δ < ε, M decelerates with an acceleration of -f(δ)" are constructed. Then, if it is verified that the control rule of M satisfies the safety condition of M, it can be guaranteed that M will behave safely on RSS.

[0141] However, since RSS does not consider measurement errors, although S should have been safe without measurement errors, it may not satisfy safety when there are measurement errors. For example, in the aforementioned model M, the control rule and acceleration used are determined according to the distance from the preceding vehicle, but if there is a measurement error in the distance, the appropriate control rule and appropriate acceleration cannot be determined.

[0142] To address this problem, it is desired to modify S to be robust against measurement errors. At this time, the user inputs, as the first controller model 101 (M), information on what measurement errors may exist for each variable of S as error information into the information processing device 10 respectively.

[0143] For example, the first controller model acquisition unit 11 may read the file of M recorded in the auxiliary storage device, convert it into a processable form in a structured manner in the memory device, and further display the content of M on the display device. As a result, the set of variables, safety conditions, and control rules described in M are displayed on the display device.

[0144] Subsequently, the first controller model acquisition unit 11 may display a prompt on the display device to request input of information on what measurement errors may exist for each variable of S, receive the user's input, and read it into the memory device as error characteristic information.

[0145] Subsequently, the control condition generation unit 12, the parameter restriction generation unit 13, and the control operation generation unit 14 newly generate a list of new control rules that behave so as to satisfy the safety conditions of M even under the received error characteristics, and hold them in the memory device as the second controller model 103.

[0146] The new control rules described in the second controller model 103 generated in this way improve the old control rules described in the first controller model 101 (M) so that, for example, even when there is an error in measuring the distance to the vehicle ahead, the safe following distance defined by RSS can be maintained.

[0147] Then, the second controller model output unit 15 may display the second controller model 103 held in the memory device on the display device and record it in the auxiliary storage device.

[0148] Furthermore, the robustness condition generation unit 16 generates robustness conditions and holds them in the memory device as robustness condition data 104.

[0149] Subsequently, the robustness condition data output unit 17 may display the robustness condition data 104 held in the memory device on the display device and record it in the auxiliary storage device.

[0150] According to this embodiment, a controller model considering measurement errors can be generated based on the model M of the control software S that surely adheres to the rules described in RSS applied to the autonomous vehicle.

[0151] (Second Embodiment) The following describes the second embodiment with reference to the drawings. The second embodiment differs from the first embodiment in that, in generating an expression indicating parameter restrictions of the second controller model 103, it allows operations different from those of the first controller model 101. Therefore, in the following description of the second embodiment, the differences from the first embodiment will be mainly described, and components having the same functional configuration as those in the first embodiment will be given the same reference numerals as those used in the description of the first embodiment, and the description thereof will be omitted.

[0152] FIG. 16 is a flowchart showing an example of the flow of parameter restriction generation processing according to the second embodiment.

[0153] The parameter restriction generation processing according to the present embodiment is the same as the parameter restriction generation processing according to the first embodiment, and generates an expression indicating a looser restriction. Specifically, the difference from the parameter restriction generation processing according to the first embodiment is that the conditions that the parameter control should satisfy in step S1002 shown in FIG. 16 are as follows.

[0154] · Satisfies any one of the parameter restrictions of the first controller model 101 in each control condition included in the selected subset S.

[0155] In the first embodiment, it is a condition that all the parameter restrictions of the first controller model 101 are satisfied, but in this embodiment, it is a condition that any one of the parameter restrictions of the first controller model 101 is satisfied. As will be described later, since the control operation of the second controller model 103 in each control condition becomes any one of the control operations of the first controller model 101, if the parameter satisfies the parameter restriction, the safety condition is satisfied. Therefore, in this embodiment, the condition corresponding to (b) of the first embodiment is unnecessary.

[0156] Therefore, it can be said that the parameter restrictions generated in this embodiment are looser than those in the first embodiment.

[0157] FIG. 17 is a diagram for explaining a specific example of a result of generating parameter restrictions according to the second embodiment.

[0158] As a result of generating parameter constraints according to this embodiment, for example, case {ok,hot} The parameter of the control rule (p keep ,p cool The formula showing the parameter limit of (∀ ~ t∈[^t-3,^t+3].(t+p keep ∈[30,40]∧p keep ∈[-6,6]))∨(∀ ~ t∈[^t-3,^t+3].(tp cool ∈[30,40]∧0 <p cool ))".

[0159] FIG. 18 is a flowchart showing an example of the flow of a parameter-limited character string generation process according to the second embodiment.

[0160] In the parameter restriction character string generation process according to the present embodiment, the parameter restriction generation unit 13 adds, to the character string indicating the parameter restriction of the second controller model 103, “a value that is likely to be a true value estimated from the measured value ^v” ~ The result of the process in step S1101 is the case {ok,hot} The control condition of the control rule is included in the formula "∀ ~ This is equivalent to the string "t∈[^t-3,^t+3]."

[0161] Next, the parameter restriction generator 13 selects one element (case i ) (step S1102).

[0162] Next, the parameter restriction generating unit 13 adds “parameter case of the first controller model 101” to the character string indicating the parameter restriction of the second controller model 103. i (v,p i ) or "v" in the formula ~Append the string replaced with v (step S1103). As a result of the process in this step S1103, in the example of FIG. 9, case {ok,hot} The string “(t + p keep ∈ [30, 40] ∧ p keep ∈ [−6, 6]) ∨” included in the expression indicating the control condition of the use control rule corresponds to the appended string regarding case ok , case hot} among S = {case ok}. However, “or” is omitted when finally executing the process of this step S1103.

[0163] Next, the parameter restriction generation unit 13 determines whether processing has been performed for all elements of the selected subset S (step S1104). If the parameter restriction generation unit 13 determines that processing has not been performed for any subset S (step S1104: No), it returns to the process of step S1101.

[0164] Also, if the parameter restriction generation unit 13 determines that processing has been performed for all elements of the selected subset S (step S1104: Yes), it ends the parameter restriction string generation process.

[0165] Further, in the first embodiment, the expression indicating the control operation of the second controller model 103 is an expression indicating an operation common to the control operations of all elements of the subset S. In contrast, in the present embodiment, the expression indicating the control operation of the second controller model 103 is an expression indicating any one of the control operations of the elements of the subset S.

[0166] FIG. 19 is a diagram for explaining a specific example of the generation result of the control operation according to the second embodiment.

[0167] For example, the expression indicating the control operation of the case {ok,hot} use control rule is “(t′ ∈ ({t′|t′ = t + p keep} ∪ {t′|t′ = t - p cool})”.

[0168] FIG. 20 is a flowchart showing an example of the flow of control operation generation processing according to the second embodiment.

[0169] In step S1202 of the control operation generation processing according to the present embodiment, the control operation generation unit 14 generates an expression indicating any one of the control operations of the first controller model 101 in each control condition included in the selected subset S.

[0170] The parameter restriction according to the first embodiment is a restriction on the parameters for faithfully reproducing the control operation related to the first controller model 101. Therefore, the range of measurable errors that can be tolerated becomes small, and there is a possibility that it cannot withstand practical use.

[0171] On the other hand, the parameter restriction according to the present embodiment can be made to withstand a larger range of measurement errors by reducing the reproducibility of some control operations. Note that since the parameter restriction according to the present embodiment is required to satisfy any of the parameter restrictions included in the first controller model 101, the constraint of satisfying the safety condition is maintained. Therefore, also in the present embodiment, like the first embodiment, safety is guaranteed.

[0172] Specifically, it is conceivable that the parameter restriction in the first controller model 101 includes restrictions that are stricter than the safety condition due to environmental considerations or the like. In this case, the parameter restriction generation unit 13 according to the first embodiment generates an expression indicating a parameter restriction that satisfies such a restriction stricter than the safety condition. On the other hand, the parameter restriction generation unit 13 according to the second embodiment generates an expression indicating a parameter restriction that does not satisfy a restriction stricter than the safety condition in some operations. Thereby, the second controller model 103 with improved practicality can be constructed.

[0173] In the above-described embodiments, an example of a controller model including parameter restrictions as elements independent of control conditions and control operations has been shown, but the parameter restrictions may be included in the control conditions or control operations.

[0174] (Hardware configuration example according to this embodiment) The information processing apparatus 10 can be realized, for example, by causing a computer to execute a program describing the processing contents described in this embodiment. Note that this "computer" may be a physical machine or a virtual machine on the cloud. When using a virtual machine, the "hardware" described here is virtual hardware.

[0175] The above program can be recorded on a computer-readable recording medium (such as a portable memory), saved, or distributed. It is also possible to provide the above program through a network such as the Internet or email.

[0176] FIG. 21 is a diagram showing a hardware configuration example of the above computer. The computer in FIG. 21 includes a drive device 1000, an auxiliary storage device 1002, a memory device 1003, a CPU 1004, an interface device 1005, a display device 1006, an input device 1007, an output device 1008, etc., which are mutually connected by a bus B.

[0177] A program for realizing the processing on the computer is provided, for example, by a recording medium 1001 such as a CD-ROM or a memory card. When the recording medium 1001 storing the program is set in the drive device 1000, the program is installed from the recording medium 1001 to the auxiliary storage device 1002 via the drive device 1000. However, the installation of the program does not necessarily have to be performed from the recording medium 1001, and it may be downloaded from another computer via a network. The auxiliary storage device 1002 stores the installed program and also stores necessary files, data, etc.

[0178] When there is an instruction to start a program, the memory device 1003 reads and stores the program from the auxiliary storage device 1002. The CPU 1004 realizes the functions related to the device according to the program stored in the memory device 1003. The interface device 1005 is used as an interface for connecting to a network. The display device 1006 displays a GUI (Graphical User Interface) or the like according to the program. The input device 1007 is composed of a keyboard, a mouse, buttons, a touch panel, etc., and is used to input various operation instructions. The output device 1008 outputs the calculation result.

[0179] Note that the above computer may be provided with a GPU (Graphics Processing Unit) or a TPU (Tensor Processing Unit) instead of the CPU 1004, or may be provided with a GPU or a TPU in addition to the CPU 1004. In that case, the GPU or the TPU may execute the processing that requires special calculations, and the CPU 1004 may execute the other processing, so that the processing may be shared and executed.

[0180] As described above, the present embodiment has been described, but the present invention is not limited to such a specific embodiment, and various modifications and changes are possible within the scope of the gist of the present invention described in the claims.

[0181] The information processing apparatus and program of the present embodiment may be configured as the information processing apparatus and program shown in the following items. Further, the following control design support method may be implemented.

[0182] <Configuration related to the present embodiment> (Item 1) An information processing apparatus for outputting a second controller model that can withstand a measured value including a measurement error based on a first controller model including information indicating a control condition based on a measured value and information indicating a control operation that defines the operation of a control target when the control condition is satisfied. A control condition generation unit that generates information indicating control conditions included in the second controller model based on information indicating the control conditions included in the first controller model; A control operation generation unit that generates information indicating control operations included in the second controller model based on information indicating the control conditions and control operations included in the first controller model, and an information processing apparatus. Information processing apparatus. (Item 2) The control condition generation unit selects one from all subsets of the control conditions included in the first controller model, and generates information indicating a condition that all possible values of the true value inferred from the measured values are included in the selected subset. The information processing apparatus according to Item 1. (Item 3) The control operation generation unit selects one from all subsets of the control conditions included in the first controller model, and generates information indicating an operation common to all control operations of the first controller model in each control condition included in the selected subset. The information processing apparatus according to Item 1 or Item 2. (Item 4) The first controller model further includes information indicating parameter restrictions indicating conditions that parameters included in the control operations should satisfy. The first controller model further includes a parameter restriction generation unit that generates information indicating parameter restrictions included in the second controller model based on information indicating the control conditions, the parameter restrictions, and the control operations included in the first controller model. The information processing apparatus according to any one of Items 1 to 3. (Item 5) The first controller model includes information indicating a safety condition that is a condition of safety to be guaranteed. The parameter limit generation unit selects one from all subsets of the control conditions included in the first controller model, satisfies all the parameter limits of the first controller model in each control condition included in the selected subset, and there is an operation common to the control operations of the first controller model corresponding to all the control conditions included in the selected subset, and generates information indicating a parameter limit that satisfies the safety condition in the state controlled by the common operation. The information processing apparatus according to claim 4. (Claim 6) The first controller model further includes information indicating parameter limits indicating conditions that parameters included in the control operation should satisfy, and information indicating a safety condition that is a safety condition to be guaranteed. The first controller model further includes a parameter limit generation unit that generates information indicating a parameter limit of selecting one from all subsets of the control conditions included in the first controller model and satisfying any of the parameter limits of the first controller model in each control condition included in the selected subset. The control operation generation unit selects one from all subsets of the control conditions included in the first controller model and generates an expression indicating any of the control operations of the first controller model in each control condition included in the selected subset. The information processing apparatus according to claim 1 or claim 2. (Claim 7) The information indicating the control conditions, the parameter limits, and the control operations included in the second controller model is generated in a format that can later add the characteristics of the measurement error. The information processing apparatus according to any one of claims 4 to 6. (Claim 8) The information indicating the control conditions, the parameter limits, and the control operations included in the second controller model is generated based on the information indicating the characteristics of the measurement error. The information processing apparatus according to any one of claims 4 to 6. (Claim 9) Based on the control conditions and the information indicating the parameter restrictions included in the second controller model, further comprising a robustness condition generation unit that generates robustness condition data indicating conditions that are the limit of the measurable error that the second controller model can withstand. The information processing apparatus according to any one of Items 4 to 8. (Item 10) The robustness condition generation unit selects one from all subsets of the control conditions included in the first controller model, and for all measured values that satisfy each control condition included in the second controller model corresponding to the selected subset, generates information indicating that the condition that there exists a parameter that satisfies the parameter restrictions included in the second controller model is satisfied for all subsets, as the robustness condition data. The information processing apparatus according to Item 9. (Item 11) A control design support method executed by an information processing apparatus for outputting a second controller model that can withstand measured values including measurement errors based on a first controller model including information indicating control conditions based on measured values and information indicating a control operation that defines the operation of a control target when the control conditions are satisfied, Based on the information indicating the control conditions included in the first controller model, generating information indicating the control conditions included in the second controller model; Based on the information indicating the control conditions and the control operation included in the first controller model, generating information indicating the control operation included in the second controller model. Control design support method. (Item 12) In a computer included in an information processing apparatus for outputting a second controller model that can withstand measured values including measurement errors based on a first controller model including information indicating control conditions based on measured values and information indicating a control operation that defines the operation of a control target when the control conditions are satisfied, Based on the information indicating the control conditions included in the first controller model, generating information indicating the control conditions included in the second controller model; Generating information indicating a control operation included in the second controller model based on the information indicating the control conditions and the control operation included in the first controller model; A program for causing execution.

[0183] This international application claims priority based on Japanese Patent Application No. 2021-087649 filed on May 25, 2021, and incorporates the entire contents of the Japanese patent application herein by reference.

Explanation of Reference Numerals

[0184] 10 Information processing apparatus 11 First controller model acquisition unit 12 Control condition generation unit 13 Parameter restriction generation unit 14 Control operation generation unit 15 Second controller model output unit 16 Robustness condition generation unit 17 Robustness condition data output unit 101 First controller model 102 Error characteristic information 103 Second controller model 104 Robustness condition data 1000 Drive device 1001 Recording medium 1002 Auxiliary storage device 1003 Memory device 1004 CPU 1005 Interface device 1006 Display device 1007 Input device 1008 Output device

Claims

1. A first controller model acquisition unit that acquires a first controller model including information indicating a control condition based on a measurement value and information indicating a control operation that defines an operation of a control target when the control condition is satisfied; A second controller model output unit that outputs a second controller model capable of withstanding a measurement value including a measurement error, The second controller model includes information indicating a control condition based on the information indicating the control condition included in the first controller model, and information indicating a control operation based on the information indicating the control condition and the control operation included in the first controller model. The information indicating the control condition and the control operation included in the second controller model is generated based on information indicating the characteristics of the measurement error. An information processing apparatus.

2. The information processing apparatus further includes a control condition generation unit that selects one from all subsets of the control conditions included in the first controller model, and generates information indicating a condition that all possible values of a true value inferred from the measurement value are included in the selected subset, as the information indicating the control condition included in the second controller model. The information processing apparatus according to claim 1.

3. The information processing apparatus further includes a control operation generation unit that selects one from all subsets of the control conditions included in the first controller model, and generates information indicating an operation common to all control operations of the first controller model in each control condition included in the selected subset, as the information indicating the control operation included in the second controller model. The information processing apparatus according to claim 1.

4. The first controller model further includes information indicating a parameter restriction indicating a condition that a parameter included in the control operation should satisfy. The information processing apparatus further includes a parameter restriction generation unit that generates information indicating a parameter restriction included in the second controller model based on the information indicating the control condition, the parameter restriction, and the control operation included in the first controller model. The information processing apparatus according to claim 1.

5. The first controller model includes information indicating a safety condition that is a condition of safety to be ensured. The parameter limit generation unit selects one from all subsets of the control conditions included in the first controller model, satisfies all the parameter restrictions of the first controller model in each control condition included in the selected subset, and there is an operation common to the control operations of the first controller model corresponding to all the control conditions included in the selected subset, and generates information indicating a parameter restriction that satisfies the safety condition in the state controlled by the common operation. The information processing apparatus according to claim 4.

6. The first controller model further includes information indicating a parameter restriction indicating a condition that a parameter included in the control operation should satisfy, and information indicating a safety condition that is a safety condition to be guaranteed. A parameter restriction generation unit that selects one from all subsets of the control conditions included in the first controller model and generates information indicating a parameter restriction that satisfies any one of the parameter restrictions of the first controller model in each control condition included in the selected subset as information included in the second controller model. The apparatus further includes a control operation generation unit that selects one from all subsets of the control conditions included in the first controller model and generates, as information indicating the control operation included in the second controller model, an expression indicating any one of the control operations of the first controller model in each control condition included in the selected subset. The information processing apparatus according to claim 1.

7. The information indicating the control conditions, the parameter restrictions, and the control operations included in the second controller model is generated in a form that allows the characteristics of the measurement error to be added later. The information processing apparatus according to any one of claims 4 to 6.

8. The information indicating the parameter restrictions included in the second controller model is generated based on the information indicating the characteristics of the measurement error. The information processing apparatus according to any one of claims 4 to 6.

9. The apparatus further includes a robustness condition data output unit that outputs robustness condition data indicating conditions that are the limits of the measurement errors that the second controller model can withstand based on the information indicating the control conditions and the parameter restrictions included in the second controller model. The information processing apparatus according to any one of claims 4 to 6.

10. Select one from all subsets of the control conditions included in the first controller model, and for all measurement values that satisfy each control condition included in the second controller model corresponding to the selected subset, generate information indicating that the condition that there exists a parameter that satisfies the parameter restriction included in the second controller model is satisfied for all subsets, as the robustness condition data, further comprising a robustness condition generation unit. The information processing apparatus according to claim 9.

11. A control design support method executed by an information processing apparatus, obtaining a first controller model including information indicating control conditions based on measurement values and information indicating a control operation that defines the operation of a control target when the control conditions are satisfied; outputting a second controller model that can withstand measurement values including measurement errors, wherein the second controller model includes information indicating control conditions based on the information indicating the control conditions included in the first controller model, and information indicating a control operation based on the information indicating the control conditions and the control operation included in the first controller model; the information indicating the control conditions and the control operation included in the second controller model is generated based on information indicating the characteristics of the measurement error; Control design support method.

12. A program for causing a computer included in an information processing apparatus to obtain a first controller model including information indicating control conditions based on measurement values and information indicating a control operation that defines the operation of a control target when the control conditions are satisfied; output a second controller model that can withstand measurement values including measurement errors, wherein the second controller model includes information indicating control conditions based on the information indicating the control conditions included in the first controller model, and information indicating a control operation based on the information indicating the control conditions and the control operation included in the first controller model; the information indicating the control conditions and the control operation included in the second controller model is generated based on information indicating the characteristics of the measurement error; Program.

Images