外文翻譯資料----智能控制

1、<p><b>　　外文資料翻譯</b></p><p>　　INTELLIGENT CONTROL</p><p>　　Intelligence and intelligent systems can be characterized in a number of ways and along a number of dimensions. There a

2、re certain attributes of intelligent systems, common in many definitions, which are of particular interest to the control community.</p><p>　　In the following, several alternative definitions and certain ess

3、ential characteristics of intelligent systems are first discussed. A brief working definition of intelligent systems that captures their common characteristics is then presented. In more detail, we start with a rather ge

4、neral definition of intelligent systems, we discuss levels of intelligence, and we explain the role of control in intelligent systems and outline several alternative definitions. We then discuss adaptation and learn</

5、p><p>　　We start with a general characterization of intelligent systems:</p><p>　　An intelligent system has the ability to act appropriately in an uncertain environment, where an appropriate action

6、 is that which increases the probability of success, and success is the achievement of behavioral subgoals that support the system’s ultimate goal.</p><p>　　In order for a man-made intelligent system to act

7、appropriately, it may emulate functions of living creatures and ultimately human mental faculties. An intelligent system can be characterized along a number of dimensions. There are degrees or levels of intelligence that

8、 can be measured along the various dimensions of intelligence. At a minimum, intelligence requires the ability to sense the environment, to make decisions and to control action. Higher levels of intelligence may include

9、the abilit</p><p>　　The above characterization of an intelligent system is rather general. According to this, a great number of systems can be considered intelligent. In fact, according to this definition, e

10、ven a thermostat may be considered to be an intelligent system, although of low level of intelligence. It is common, however, to call a system intelligent when in fact it has a rather high level of intelligence.</p>

11、;<p>　　There exist a number of alternative but related definitions of intelligent systems and in the following we mention several. They provide alternative, but related characterizations ofintelligent systems with

12、 emphasis on systems with high degrees of intelligence.</p><p>　　The following definition emphasizes the fact that the system in question processes information, and it focuses on man-made systems and intelli

13、gent machines:</p><p>　　A. Machine intelligence is the process of analyzing, organizing and converting data into knowledge; where (machine) knowledge is defined to be the structured information acquired and

14、applied to remove ignorance or uncertainty about a specific task pertaining to the intelligent machine. This definition leads to the principle of increasing precision with decreasing ntelligence, which claims that: apply

15、ing machine intelligence to a database generates a flow of knowledge, lending an analytic form to</p><p>　　B. Many adaptive or learning control systems can be thought of as designing a control law to meet we

16、ll-defined control objectives. This activity represents the system’s attempt to organize or order its “knowledge” of its own dynamical behavior, so to meet a control objective. The organization of knowledge can be seen a

17、s one important attribute of intelligence. If this organization is done autonomously by the system, then intelligence becomes a property of the system, rather than of the system’s </p><p>　　A procedural char

18、acterization of intelligent systems is given next:</p><p>　　C. Intelligence is a property of the system that emerges when the procedures of focusing attention, combinatorial search, and generalization are ap

19、plied to the input information in order to produce the output. One can easily deduce that once a string of the above procedures is defined, the other levels of resolution of the structure of intelligence are growing as a

20、 result of the recursion. Having only one level structure leads to a rudimentary intelligence that is implicit in the thermostat, or </p><p>　　The concepts of intelligence and control are closely related and

21、 the term “Intelligent Control” has a unique and distinguishable meaning. An intelligent system must define and use goals. Control is then required to move the system to these goals and to define such goals. Consequently

22、, any intelligent system will be a control system. Conversely, intelligence is necessary to provide desirable functioning of systems under changing conditions, and it is necessary to achieve a high degree of autonomo<

23、/p><p>　　Below, one more alternative characterization of intelligent (control) systems is included. According to this view, a control system consists of data structures or objects (the plant models and the cont

24、rol goals) and processing units or methods (the control laws) :</p><p>　　D. An intelligent control system is designed so that it can autonomously achieve a high levelgoal, while its components, control goals

25、, plant models and control laws are not completely defined, either because they were not known at the design time or because they changed unexpectedly.</p><p>　　There are several essential properties present

26、 in different degrees in intelligent systems. One can perceive them as intelligent system characteristics or dimensions along which different degrees or levels of intelligence can be measured. Below we discuss three such

27、 characteristics that appear to be rather fundamental in intelligent control systems.</p><p>　　Adaptation and Learning. The ability to adapt to changing conditions is necessary in an intelligent system. Alth

28、ough adaptation does not necessarily require the ability to learn, for systems to be able to adapt to a wide variety of unexpected changes learning is essential. So the ability to learn is an important characteristic of

29、(highly) intelligent systems.</p><p>　　Autonomy and Intelligence. Autonomy in setting and achieving goals is an important characteristic of intelligent control systems. When a system has the ability to act a

30、ppropriately in an uncertain environment for extended periods of time without external intervention, it is considered to be highly autonomous. There are degrees of autonomy; an adaptive control system can be considered a

31、s a system of higher autonomy than a control system with fixed controllers, as it can cope with greater uncertai</p><p>　　Structures and Hierarchies. In order to cope with complexity, an intelligent system m

32、ust have an appropriate functional architecture or structure for efficient analysis and evaluation of control strategies. This structure should be “sparse” and it should provide a mechanism to build levels of abstraction

33、 (resolution, granularity) or at least some form of partial ordering so to reduce complexity. An approach to study intelligent machines involving entropy emphasizes such efficient computational s</p><p>　　In

34、 view of the above, a working characterization of intelligent systems (or of (highly) intelligent (control) systems or machines) that captures the essential characteristics present in any such system is:</p><p

35、>　　An intelligent system must be highly adaptable to significant unanticipated changes, and so learning is essential. It must exhibit high degree of autonomy in dealing with changes. It must be able to deal with signi

36、ficant complexity, and this leads to certain sparse types of functional architectures such as hierarchies.</p><p><b>　　智 能 控 制</b></p><p>　　智能與智能系統(tǒng)能用許多方式和從許多方面來描述。通常包含智能系統(tǒng)的特征，這些也是控制領(lǐng)

37、域所關(guān)心的特征。</p><p>　　下面，首先討論智能系統(tǒng)的幾種定義和某些基本特性。接著陳述一下具有共同特性的智能系統(tǒng)的一個(gè)簡潔的工程定義。更詳細(xì)地，我們從智能系統(tǒng)的一個(gè)非常一般的定義開始，討論智能程度，解釋控制在智能系統(tǒng)中的作用，并概括出幾種定義。然后討論智能系統(tǒng)中的自適應(yīng)和自學(xué)習(xí)、自主性和必要的高效計(jì)算結(jié)構(gòu)，來處理智能系統(tǒng)的復(fù)雜性。最后歸結(jié)出智能(控制)系統(tǒng)的基本特征。</p><p&g

38、t;　　我們從智能系統(tǒng)的一般特征開始：</p><p>　　一個(gè)智能系統(tǒng)應(yīng)具備在不可預(yù)測(cè)的環(huán)境下適當(dāng)工作的能力，在這個(gè)環(huán)境中一個(gè)適當(dāng)?shù)姆磻?yīng)能夠增加成功的可能性，從而達(dá)到系統(tǒng)最終的目的。</p><p>　　為了能讓人造智能系統(tǒng)適當(dāng)?shù)墓ぷ?，它?yīng)能模擬生物的功能和基本的人的智能。一個(gè)智能系統(tǒng)能從多個(gè)方面來描述。智能程度能從智能的各個(gè)方面測(cè)得。智能至少要具有感受環(huán)境、進(jìn)而做出決定來進(jìn)行控制的能力

39、。智能化程度比較高的智能系統(tǒng)具有識(shí)別目標(biāo)和事件、描述世界模型中的知識(shí)、思考并計(jì)劃未來的能力。在智能化程度更高級(jí)的形式中，智能具有感知和理解、理智地做出選擇、在各種各樣的環(huán)境下成功地運(yùn)行以便能在復(fù)雜的、不利的環(huán)境下生存和發(fā)展的能力。通過計(jì)算能力的發(fā)展和在復(fù)雜多變的環(huán)境中怎樣感知、決定并做出響應(yīng)的知識(shí)的積累，我們可以觀察到智能也在更新與發(fā)展。</p><p>　　智能系統(tǒng)的以上特征是非常普遍的。據(jù)上所述，很多系統(tǒng)都可

40、以被認(rèn)為是智能的。事實(shí)上，根據(jù)這種定義，恒溫器盡管只是低水平的智能，但是也可以被認(rèn)為是智能系統(tǒng)。然而，習(xí)慣上當(dāng)一個(gè)系統(tǒng)具有高水平的智能時(shí)，我們才稱它為智能系統(tǒng)。</p><p>　　智能系統(tǒng)存在許多相關(guān)的定義，下面我們提到了幾種。它們提供可選擇但相關(guān)的智能系統(tǒng)的特性，這種智能系統(tǒng)著重強(qiáng)調(diào)系統(tǒng)的高程度智能。</p><p>　　下面的定義強(qiáng)調(diào)這樣一個(gè)事實(shí)，處理信息的系統(tǒng)集中在人造系統(tǒng)和智能機(jī)

41、器上：</p><p>　　機(jī)器智能是分析、組織和轉(zhuǎn)換數(shù)據(jù)成知識(shí)的過程，在這里，知識(shí)被定義為結(jié)構(gòu)化的信息，這種信息被用來消除無知或相對(duì)于智能機(jī)器來說某些特殊任務(wù)的不確定性。這種定義導(dǎo)致了增加精度而相對(duì)減少智能的原則，這種原則表示：應(yīng)用機(jī)器智能到數(shù)據(jù)庫能生成一系列的知識(shí)，通過分析形式進(jìn)行過程建模。</p><p>　　其次，智能系統(tǒng)具有自動(dòng)分配任務(wù)和在內(nèi)部自主地控制執(zhí)行機(jī)構(gòu)的特性：</

42、p><p>　　B. 許多自適應(yīng)或自學(xué)習(xí)控制系統(tǒng)被認(rèn)為是作為一種控制法則來滿足明確的控制目標(biāo)。這種行為代表著系統(tǒng)試圖組織或排列自己動(dòng)態(tài)行為的知識(shí)，來滿足控制目標(biāo)。這些知識(shí)的組織是組成智能的一個(gè)重要特征。如果系統(tǒng)能夠自主地實(shí)現(xiàn)這種組織，那么智能就成為系統(tǒng)的一種性能，而非系統(tǒng)設(shè)計(jì)者的。這意味著能內(nèi)部實(shí)現(xiàn)自組織原則的自組織控制器是智能控制系統(tǒng)。</p><p>　　下面給出智能系統(tǒng)的過程特性：<

43、;/p><p>　　C. 智能是系統(tǒng)的一種特性，當(dāng)集中注意、聯(lián)合搜索和概括等過程被應(yīng)用在輸入信息，從而產(chǎn)生輸出時(shí)，這種特性就會(huì)出現(xiàn)。你可以很容易的推斷出：一旦以上過程被定義，具有智能結(jié)構(gòu)的結(jié)果規(guī)則生成并作為新的結(jié)果。只有一種標(biāo)準(zhǔn)結(jié)構(gòu)將導(dǎo)致在恒溫器中固有的一個(gè)不成熟的智能或?qū)е鲁蔀橐粋€(gè)可變結(jié)構(gòu)的變化模型的控制器。</p><p>　　智能和控制的概念緊密相關(guān)，并且術(shù)語“智能控制”有著獨(dú)特的、可區(qū)

44、別的意思。一個(gè)智能系統(tǒng)必須定義和利用目標(biāo)?？刂票灰笥脕眚?qū)動(dòng)系統(tǒng)達(dá)到這些目標(biāo)并定義這些目標(biāo)。因此，任何智能系統(tǒng)都是控制系統(tǒng)。相反地，智能必須在條件變化的情況下提供合適的系統(tǒng)運(yùn)行過程，也必須在控制系統(tǒng)中具有高度的自主行為。因?yàn)榭刂剖侨魏沃悄芟到y(tǒng)的一個(gè)重要部分，因此在工程文獻(xiàn)中“智能控制系統(tǒng)”有時(shí)被用來代替“智能系統(tǒng)”或“智能機(jī)器”。“智能控制系統(tǒng)”強(qiáng)調(diào)智能系統(tǒng)中的控制方面。</p><p>　　下面，介紹一下智能(

45、控制)系統(tǒng)的另外一些特性。根據(jù)觀察，一個(gè)控制系統(tǒng)包括數(shù)據(jù)結(jié)構(gòu)或?qū)ο?設(shè)備模型和控制目標(biāo))和處理單元或方法(控制規(guī)則)：</p><p>　　D. 由于組件、控制目標(biāo)、加工模型和控制法則并沒有完全被定義，沒被定義的原因或是因?yàn)樵谠O(shè)計(jì)時(shí)不了解，或是因?yàn)樗鼈冊(cè)诓豢深A(yù)測(cè)地變化著。所以設(shè)計(jì)一個(gè)智能控制系統(tǒng)以便能自動(dòng)獲得高標(biāo)準(zhǔn)目標(biāo)。</p><p>　　在具有不同智能化程度的系統(tǒng)中呈現(xiàn)出幾種基本特性。

46、你可以認(rèn)為它們是智能系統(tǒng)的特征或是衡量智能程度的方面。下面我們討論在智能控制系統(tǒng)中三種非?；镜奶匦?。</p><p>　　自適應(yīng)與自學(xué)習(xí) 在智能系統(tǒng)中適應(yīng)變化多端的條件的能力是必需的。盡管自適應(yīng)不一定要具備自學(xué)習(xí)能力，但一個(gè)系統(tǒng)要適應(yīng)不可預(yù)測(cè)的各種變化，學(xué)習(xí)是最必要的。因此學(xué)習(xí)能力是高智能系統(tǒng)的一種重要特性。</p><p>　　自主性和智能 在設(shè)置和獲取目標(biāo)過程中自主性是智能控制系統(tǒng)的

47、一個(gè)重要特性。當(dāng)一個(gè)系統(tǒng)在一個(gè)沒有外界干預(yù)的不確定的環(huán)境中能正常運(yùn)行時(shí)，它就被認(rèn)為是一個(gè)高自主性系統(tǒng)。不同系統(tǒng)的自主性程度是不一樣的，自適應(yīng)控制系統(tǒng)被認(rèn)為比安裝了控制器的控制系統(tǒng)具有更高自主能力，因?yàn)樗纫粋€(gè)固定的反饋控制器更能處理不可預(yù)測(cè)的問題。盡管對(duì)低自主性來說，無智能(或“低”智能)是必然的，但是對(duì)高自主性來說，系統(tǒng)的智能(或“高”智能度)是很必要的。</p><p>　　結(jié)構(gòu)和層次 為了應(yīng)對(duì)復(fù)雜情況，一個(gè)

48、智能系統(tǒng)必須具備一個(gè)合適的功能結(jié)構(gòu)來進(jìn)行有效的分析和控制決策的估計(jì)。這種結(jié)構(gòu)是“稀少的”并且它應(yīng)該提供一種機(jī)制來建立提取標(biāo)準(zhǔn)(決議)或者至少提供某種形式的部分規(guī)則來減少復(fù)雜性。一種研究智能機(jī)器的熵方法能加強(qiáng)這種有效的計(jì)算結(jié)構(gòu)。能夠適應(yīng)的層次(或許是大致的、局部的或是組合的層次)，可以作為一種主要的工具來處理復(fù)雜性。這里的“層次”術(shù)語是指功能性層次，或者是指時(shí)空的范圍和決議，而且它并不意味著是一種層次硬件。這里面的某些結(jié)構(gòu)可能是硬件的。為

49、了應(yīng)付多變的環(huán)境，自學(xué)習(xí)能力是必要的，因此這種結(jié)構(gòu)應(yīng)能適應(yīng)重要的、不可預(yù)料的變化。</p><p>　　鑒于以上所述，具有其他任何系統(tǒng)都存在的基本特性的智能系統(tǒng)(高智能控制系統(tǒng)或機(jī)器)的一個(gè)工作特征是：</p><p>　　一個(gè)智能系統(tǒng)必須對(duì)重要的、不可預(yù)料的變化具有很高的適應(yīng)性，而且自學(xué)習(xí)也是必要的。在應(yīng)對(duì)變化因素時(shí)它必須呈現(xiàn)出高度自主性。它必須能夠處理非常復(fù)雜的問題，而且這將導(dǎo)致某些稀