JP2913998B2 - Aircraft and other abnormality handling devices - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for detecting an abnormality in an aircraft or the like and giving an appropriate instruction to a pilot.

[0002]

2. Description of the Related Art When an abnormality such as a failure occurs in an aircraft or the like, there is a danger that the accident will develop into a major accident and damage human life.

[0003] By the way, when an abnormality is found in an aircraft or the like, generally, except for the latest aircraft and the like, a caution signal and an alarm signal of an onboard device and a BITE (Built In Tes) of an electronic device are generally used.
t) It is performed only by display, etc., and the occupants such as pilots judge the information presented by them, recognize and identify troubles, etc., and perform the necessary operations to navigate safely .

[0004] However, such a caution signal, alarm signal, BITE display, or the like, merely indicates an abnormality of an individual device and does not indicate a relation with other devices. Cannot be identified. Therefore, when an occupant has an abnormality, the occupant performs a series of procedures of judging the operation status of the system from these displays and instrument information, performing necessary operations based on the status, and judging the result. Therefore, there is a risk of making a human error (human error) based on prejudice or distraction, which is not sufficient in terms of safety.

[0005] From the above, in the latest aircraft,
EICAS (Engine Indication and Crew Alerting Syst
em) and ECAM (Electronic Centralized Aircraft Moni)
tor), and efforts are being made to reduce human error. Here, the EICAS monitors the operation status of the main system, identifies the location of an abnormality such as a failure, determines the degree of importance, and displays it. The checklist of the recovery operation is presented visually and audibly.

On the other hand, in the field of fault diagnosis, an expert system for making a diagnosis using knowledge has been studied in place of the conventional fault diagnosis system, but has not been put to practical use in the field of aircraft and the like. In addition, the expert system stores the contents of the empirical knowledge of the expert and the knowledge described in the document in a symbolic form in the knowledge base,
It performs inference for a given problem based on knowledge in the knowledge base and obtains an answer.

[0007]

The above-mentioned measures have been taken to deal with abnormalities in an aircraft or the like, but the following drawbacks have been pointed out.

[0008] In a system such as ECAM, a situation such as a failure and a checklist of operations to be taken by an occupant are displayed. When the list content is large, an operation to be actually performed must be selected from the list. Confused in an emergency,
May cause human error.

The conventional expert system takes a long time for inference and cannot perform processing in real time, so that it cannot be applied to anomaly handling of an aircraft or the like whose situation changes every moment. That is, in the conventional expert system,
Since the condition part of a large number of rules accumulated in the knowledge base is determined one by one, and the rules satisfying the conditions are executed to proceed with the inference, it takes a long time to process.

[0010] The present invention has been proposed in view of the above points, and an object of the present invention is to make it possible to display a more appropriate countermeasure by making it possible to employ an expert system, and to minimize the necessity. It is an object of the present invention to provide an abnormality countermeasure device for an aircraft or the like capable of improving navigation safety and preventing an aircraft accident beforehand by displaying the countermeasure of the above.

[0011]

According to the present invention, there is provided a data analyzing means for inputting information of sensors and the like obtained from various places such as an aircraft and performing analysis in the form of a symbol.
Multiple routes that link knowledge about
Knowledge based on the knowledge base expressed by rules including actions, a data file holding the results of inferences already made, and input information obtained through data analysis means and the contents of the data file. Transfer control of base knowledge from higher-level rules to lower-level rules
Inference means for executing inference by using the form, conflict resolution means for narrowing down to one optimal action when a plurality of actions are obtained as a result of inference, and a display for displaying the obtained action. ing.

[0012]

In the abnormality handling device for an aircraft or the like according to the present invention,
Data analysis means inputs information such as sensors obtained from various places such as aircraft and performs analysis in the form of symbols, and the inference means
Based on the input information obtained through the data analysis means and the contents of the data file holding the results of the already performed inference, use the knowledge base knowledge expressed by the rules relating the abnormal measures to the tree structure based on the rules. Perform inference. If the inference resulted in multiple actions,
The conflict resolving means narrows down to one optimal measure, and the obtained measure is displayed on the display.

[0013]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of an apparatus for coping with an abnormality of an aircraft or the like according to the present invention.

Referring to FIG. 1, this embodiment includes an aircraft 1 such as a helicopter, and an abnormality handling device 2 having a data analysis means 3, a knowledge base 4, an inference means 5, a conflict resolution means 6, a data file 7, and a display 8. It is composed of
9 is a pilot who is an occupant of the aircraft 1. The details of the functions and the like of each unit will be described through the following operations in order to avoid duplication.

First, various kinds of information from sensors and the like arranged at various places of the aircraft 1 are inputted to the data analysis means 3 of the abnormality coping device 2, and the data analysis means 3 adapts the information to a later inference process. In the form of a symbol. Here, as input information of the data analysis means 3, flight instruments,
Some are obtained from instruments such as an engine monitoring instrument and a main transmission instrument, and others are obtained from a control device, a switch, and the like to indicate the state.

Next, the inference means 5 comprises the data analysis means 3
The inference is performed using the knowledge of the knowledge base 4 on the basis of the input information obtained in symbolic form via the data file and the contents of the data file 7. Further, if necessary, a situation is recognized from input information obtained in a symbolic form via the data analysis means 3.

Here, the data file 7 contains information on the measures and diagnosis results already obtained by inference and displayed on the display 8, that is, to which group and rule (described later) in the knowledge base 4 the inference has progressed. Information is stored.

Further, the knowledge base 4 stores in advance information on emergency operation procedures and limitations on aircraft, and knowledge on abnormalities possessed by experts such as aviation engineers and pilots. Such knowledge is created in advance by collecting manuals (documents) for dealing with abnormalities and opinions of experts.

FIG. 2 shows an example of the contents of the knowledge base 4. A plurality of rules are associated with each other by a tree structure to form a group, and the group similarly forms a tree structure with other groups. Knowledge base 4
Is composed. Here, one group basically corresponds to each abnormal event such as a failure, and forms one unit. Note that a portion drawn with a dashed line indicates identification of a situation such as a failure, and this information is inherited by lower nodes. The underlined portion indicates that the group is shifted to another group.

As described above, the advantage of forming the knowledge base 4 in a tree structure is that the inference processing can be performed at a high speed, and that the description of rules can be facilitated. That is, in the conventional expert system, the rule to be adopted is selected by determining the condition part of all rules.
When a tree structure is adopted as in the present invention, only the lower rules of the currently selected rule need to be determined, so that the processing can be performed at a very high speed. In addition, since each rule can describe the condition part on the premise that a higher-level condition has already been satisfied, there is no need to duplicate conditions with the same content, simplifying the rule description. can do.

FIG. 3 shows a specific example of a rule, which includes a condition part indicated by IF〜, an execution part indicated by THEN〜, and other necessary information. This example is a part of the knowledge of the emergency operation procedure which describes the operation procedure and the situation to be taken by the pilot 9 when a failure occurrence, an alarm, a sign or a situation of lighting a warning light in the aircraft 1 appears, and Based on a representative document for the emergency operation of It should be noted that this type of emergency operation procedure focuses on displaying the minimum necessary danger avoidance instructions to the pilot 9 rather than identifying the fault.
Since all abnormalities must be dealt with, there is often a feature that a lot of conflicting knowledge exists when an abnormality occurs.

FIG. 4 shows an example of a specific description of a rule, and shows an example in the LISP language. Here, "setq ~" declares that a rule is defined,
"Priority ~" indicates the priority of this rule,
"Name ~" is the person (maintainer) about this rule
"Group ~" indicates the name of the group to which this rule belongs, and "type ~" indicates that the second argument is the position of the node ("serial" indicates that the rule exists in the middle between the top and the end). The third argument indicates a predicate for inheriting information from the parent node when the rule is other than the top node. Also, "if
~ "Indicates a condition for this rule to be satisfied, and" th
"en ~" indicates the content of the action to be taken when the firing is performed while satisfying the conditions of this rule, "answer ~" indicates the status of the abnormality to be inherited by the lower node, and "sequenc"
"e-" indicate predicates for inheriting information to lower nodes.

The expression form of FIG. 4 can express rules for knowledge for performing situation recognition, in addition to describing rules indicating measures as described above. That is, when expressing a rule for recognizing such a situation, an object to be recognized is described after "group", and "type ~" is a rule for recognizing the situation. It is described. The recognition of the situation based on such rules is performed by the inference means 5.
It is performed by

Returning to FIG. 1, the inference means 5 performs the inference.
At first, the condition part is determined from the rule of the top node of each group of the knowledge base 4 (see FIG. 2), and the control is shifted to the lower rule associated with the rule that satisfies the condition. Outputs the described action.
Since there are generally a plurality of rules that satisfy the conditions, there are many cases where a plurality of measures are taken.

When a plurality of measures are output, the conflict resolving means 6 narrows down to one optimal measure, and displays the resulting measure on the display 8 as a human interface.

FIG. 5 shows an example of processing in the conflict resolution means 6.

That is, the conflict resolving means 6 performs the selection based on the priority as the first stage (Step S).
1). That is, the priorities given to the rules describing the measures are compared, and the highest one is selected.
This is because priorities indicate the importance of avoiding danger, and measures with higher priority should be selected with priority.

[0029]

Next, as a second step, a selection is made by deleting the already displayed knowledge (step S 2 ). This is because no new countermeasure is presented.

Next, as a third stage, priority is given to selection of a node other than the top node (step S 3 ). This is in order to avoid leaving room for selection of subsequent measures, since measures other than the top node are ongoing.

Next, as a fourth stage, a complicated condition part, that is, a condition part having a large number of condition items in the IF part is selected (step S 4 ). This means that having more criteria has a broader impact,
This is because it should be adopted first.

Each of these steps is executed sequentially until one measure is narrowed down, and if one is narrowed down, the next step is not performed.

Returning to FIG. 1, the pilot 9 looks at the measures displayed on the display 8 and performs the operation of the instructed contents to shift the aircraft 1 to a safer state.

These processes are continued when the end of the tree structure being traced in the knowledge base 4 is reached, that is, until the handling for a failure or the like is completed or the aircraft 1 is safely landed.

[0036]

As described above, the apparatus for dealing with abnormalities of an aircraft or the like according to the present invention has the following effects.

Since a tree structure is used to represent the rules of the knowledge base, it is only necessary to make a judgment on rules lower than the already selected rules, and since there are few rules to be judged, high-speed real-time processing becomes possible. An expert system can be employed for dealing with abnormalities of this kind of aircraft and the like. Therefore, it is possible to give an appropriate instruction to an occupant such as a pilot in response to the situation of an aircraft or the like that changes from moment to moment, and it is possible to greatly improve navigation safety.

By performing the conflict resolution, a minimum necessary countermeasure is displayed, so that it is difficult to make a selection in an emergency and a quick operation can be performed. That is, by reducing the workload of the pilot, it is possible to prevent human errors due to preconceptions or distractions, such as misjudgments that are likely to be made in the process of judging the situation from a lot of information.

Since the knowledge base can be expressed by the rules of the tree structure, it is easy to describe the knowledge, it is easy to obtain useful knowledge from experts, and it is easy to maintain, so that a highly reliable knowledge base is constructed. it can.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing one embodiment of an abnormality handling device for an aircraft or the like according to the present invention.

FIG. 2 is a diagram showing an example of the contents of a knowledge base.

FIG. 3 is a diagram illustrating an example of a rule.

FIG. 4 is a diagram showing an example of a specific description of a rule.

FIG. 5 is a diagram illustrating an example of processing of a conflict resolution unit.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Aircraft 2 ... Abnormality coping device 3 ... Data analysis means 4 ... Knowledge base 5 ... Reasoning means 6 ... Conflict resolution means 7 ... Data file 8 ... Display 9 ... Pilot

──────────────────────────────────────────────────続 き Continued on front page (58) Field surveyed (Int.Cl. ⁶ , DB name) B64D 45/00 B64F 5/00

Claims (2)

(57) [Claims] 1. A data analysis means for inputting information of sensors and the like obtained from various places such as an aircraft and analyzing the information in the form of symbols, and a plurality of rules for associating knowledge on abnormality handling with a tree structure.
Knowledge based on a knowledge base expressed by rules that include actions, a data file that holds the results of inferences already performed, and input information obtained through data analysis means and the contents of the data file. Base knowledge ,
Inference means for executing inference by transferring control from rules to lower-level rules, and conflict resolution means for narrowing down to one optimal solution when multiple inferences are obtained as a result of inference An abnormality countermeasure device for an aircraft or the like, comprising a display for displaying a countermeasure. 2. A conflict resolution means, and the selection of the deal by the priority, and the selection of the deal by the deletion of the displayed knowledge in already, and the selection of a preferential deal with anything other than the top-level node, a complex of the condition part The abnormality handling device according to claim 1, further comprising a configuration in which the selection of a response is sequentially performed until the response is narrowed down to one .