Planning for Temporally Extended Goals
Fahiem Bacchus Dept.of Computer Science University of Waterloo
Waterloo,Ontario
Canada,N2L3G1
Froduald Kabanza Dept.de Math et Informatique Universite de Sherbrooke
Sherbrooke,Quebec
Canada,J1K2R1
Abstract
In planning,goals have been traditionally been viewed as specifying a set of desirable?nal states.Any plan that trans-forms the current state to one of these desirable states is viewed to be correct.Goals of this form are limited as they do not allow us to constrain the manner in which the plan achieves its objectives.
We propose viewing goals as specifying desirable sequences of states,and a plan to be correct if its execution yields one of these desirable sequences.We present a logical language,
a temporal logic,for specifying goals with this semantics.
Our language is rich and allows the representation of a range of temporally extended goals,including classical goals,goals with temporal deadlines,quanti?ed goals(with both universal and existential quanti?cation),safety goals,and maintenance goals.Our formalism is simple and yet extends previous approaches in this area.
We also present a planning algorithm that can generate correct plans for these goals.This algorithm has been implemented, and we provide some examples of the formalism at work.The end result is a planning system which can generate plans that satisfy a novel and useful set of conditions.
Introduction
One of the features that distinguishes intelligent agents is their?exibility:generally they have the ability to accomplish a task in a variety of ways.Such?exibility is necessary if the agent is to be able to accomplish a variety of tasks under a range of conditions.Yet this?exibility also poses a problem: how do we communicate to such an agent the task we want accomplished in a suf?ciently precise manner so that it does what we really want.
In the area of planning,methods and algorithms are studied by which,given information about the current situation,an intelligent agent can compose its primitive abilities so as to accomplish a desired task or goal.The afore mentioned problem then becomes the problem of designing suf?ciently expressive and precise ways of specifying goals.
Much of the work in planning has dealt with goals speci?ed as conditions on a?nal state.For example,we might specify
ning systems and theories speci?cally designed to deal with temporal constraints(and sometimes other metric resources) have also been developed[V er83,Wil88,AKRT91,CT91, Lan93,PW94].
The main difference between these previous works and what we present here,lies in our use of a temporal logic that supports a unique approach to computing plans,an approach based on formula progression.The method of formula pro-gression lends itself naturally to the speci?cation and uti-lization of domain dependent search control knowledge.As shown in our previous work[BK95],the approach of do-main dependent search control offers considerable promise, and has motivated our approach to dealing with temporally extended goals.The other works that have constructed tem-poral planners have utilized complex constraint management techniques to deal with temporal information.
In[Kab90,GK91,BKSD95]similar temporal logics and similar notions of formula progression have been utilized. In this case the main difference is that here we address classical plans,i.e.,?nite sequences of actions,while these works have concentrated on generating reactive plans,i.e., mappings from states to actions(sometimes called universal plans).Reactive plans have to specify an on-going interac-tion between an agent and its environment,and thus pose a quite distinct set of problems.
To generate plans that achieve the goals expressed in our formalism we present a planning algorithm that uses the logical mechanism of formula progression.This notion was previously utilized in our TLP LAN system[BK95].In fact we have implemented the planning algorithm by extending the TLP LAN system.TLP LAN is planning system whose key feature is that it is able to utilize domain dependent search control information..This control is expressed in a temporal logic that is a limited form of the logic presented here,and it is utilized by the planner via the mechanism of formula progression.
The planning algorithm we develop is sound and complete and as we will demonstrate it is able to generate a range of interesting plans.Further work is required,however, to evaluate the planner’s performance on realistic planning problems.
In the rest of the paper we will?rst provide the details of the logic we propose for expressing goals.This logic is a temporal logic that is based on previous work by Alur et al.[AFH91].We then present our approach to planning, provide examples to demonstrate the range of goals that our system can cope with,and discuss the heuristic adequacy of our approach to planning.Finally,we close with some conclusions and discussion of future work.
Expressing goals in MITL
We use a logical language for expressing goals.The logic is based on Metric Interval Temporal Logic developed by Alur et al.[AFH91],but we have extended it allow?rst-order quanti?cation.Syntax
We start with a collection of-ary predicate(including equal-ity and the predicate constants TRUE and FALSE)function and constant symbols,variables,and the connectives(not)and (and).We add the quanti?ers and and the modal
operators(next)and U(until).From this collection of symbols we generate MITL,the language we use to express goals.MITL is de?ned by the traditional rules for gener-ating terms,atomic formulas,and Boolean combinations, taken from ordinary?rst-order logic.In addition to those formula formation rules we add:(1)if is a formula then so is;(2)if and are formulas and is an interval then so is U(the syntax of intervals is de?ned below); and(3)if is an atomic formula in which the variable is free,and is a formula then so are,and .
Notice that in our language we use bounded quanti?cation. The atomic formula is used to specify the range over which the quanti?ed variable ranges.The precise semantics are given below.
The syntax of intervals is as one would expect.The al-lowed intervals are all intervals over the non-negative real line,and we specify an interval by giving its two endpoints, both of which are required to be non-negative numbers.To allow for unbounded intervals we allow the right endpoint to be.For example,speci?es the interval of num-bers such that,speci?es the interval
,and speci?es the interval
(i.e.,the point).
Although non-negative intervals are the only ones allowed in the formulas of MITL,in the semantics and algorithms we will need to utilize shifted intervals and to test for negative intervals.For any interval let be the set of numbers such that,be the set of numbers such that,and be true iff all numbers in are less than0.For example,is the new interval ,is the new interval,and
is true.
Finally,we introduce(implication),and(disjunction) as standard abbreviations.We also introduce the temporal modalities eventually and always as abbreviations with TRUE U,and.We will also abbreviate intervals of the form and,e.g.,
will be written as and as.Finally,we will often omit writing the interval,e.g.,we will write U as U.
Semantics
We intend that goals be expressed as sentences of the lan-guage MITL.As hinted in the introduction such formulas are intended to specify sets of sequences of states.Hence,it should not be surprising that the underlying semantics we as-sign to the formulas of MITL be in terms of state sequences.
A model for MITL is a timed sequence of states,
.In particular,a model is an in?nite se-quence of states,and each state is a?rst-order model over a ?xed domain.That is,each state assigns a denotation for each predicate and function symbol over the domain. Furthermore,there is a timing function that maps each state in to a point on the non-negative real line such that for all,and for all real numbers there exists an such that.This means that time is only required to be non-decreasing,not strictly increasing. Time can stall at a single point for any?nite number of states. Eventually,however,time must increase without bound. Let be a variable assignment,i.e.,a mapping from the variables to elements of;,,and be formulas of MITL;and be an MITL model.The semantics of MITL are then de?ned by the following clauses.
,when is atemporal(i.e.,contains no temporal modalities)and quanti?er free,iff.
iff.
U iff there exists with
such that and for all with
we have.
iff for all such that
we have.
iff there exists such that
and.
It is not dif?cult to show that any formula of MITL that has no free variables,called a sentence of MITL,has a truth value that is independent of the variable assignment.Given a sentence of MITL we say it is true in a model,, iff.
Since sentences of MITL are either true or false on any individual timed sequence of states,we can associate with every sentence a set of sequences:those sequences on which it is true.We express goals as sentences of MITL,hence we obtain our desired semantics for goals:a set of acceptable sequences.
Discussion
Intuitively,the temporal modalities can be explained as fol-lows.The next modality simply speci?es that something must be true in the next state.Its semantics do not depend on the time of the states.It is important to realize,however, that what it requires to be true in the next state may itself be a formula containing temporal modalities.MITL gets its expressive power from its ability to nest temporal modalities. The until modality is more subtle.The formula U
,for example,requires that be true in some state whose time is between5and7units into the future,and that be true in all states until we reach a state where is true.The eventually modality thus takes on the semantics that requires that be true in some state whose time lies in the
Inputs:A state,with formula label,and a time duration to the successor state.
Output:A new formula representing the formula label of the successor state.
Algorithm Progress(,,)
Case
1.contains no temporal modalities:
if TRUE
else FALSE
2.:Progress Progress
3.:Progress
4.:
5.U:
if FALSE
else if Progress
Progress U else Progress U
6.:Progress
7.:Progress
Table1:The progression algorithm.
Since actions map states to new states,any?nite sequence of actions will generate a?nite sequence of states:the states that would arise as the plan is executed.Furthermore,we will assume that part of an action’s speci?cation is a speci?cation of its duration,which is constrained to be greater than or equal to0.This means that if we consider to commence at time0,then every state that is visited by the plan can be given a time stamp.Hence,a plan gives rise to a?nite timed sequence of states—almost a suitable model for MITL. The only dif?culty is that models of MITL are in?nite sequences.Intuitively,we intend to control the agent for some?nite time,up until the time the agent completes the execution of its plan.Since we are assuming that the agent is the only source of change,once it has completed the plan the?nal state of the plan idles,i.e.,it remains unchanged. Formally,we de?ne the MITL model corresponding to a plan as follows:
De?nition1Let plan be the?nite sequence of actions .Let be the sequence of states such that,and is the initial state. is the sequence of states visited by the plan.Then the MITL model corresponding to and is de?ned to be
,i.e.,with the?nal state idled, where duration,, ,and the time of the copies of increases without bound.
Therefore,every?nite sequence of actions we generate corresponds to a unique model in which the?nal state is idling.Given a goal expressed as a sentence of MITL we can determine,using the semantics de?ned above,whether or not the plan satis?es the goal.
De?nition2Let be a plan,be a goal expressed as a formula of MITL,be the initial state,and be the model corresponding to and.is a correct plan for given iff.Generating Plans
We will generate plans by adopting the methodology of our previous work[BK95].In particular,we have constructed a forward-chaining planning engine that generates linear se-quences of actions,and thus linear sequences of states.As these linear sequences of states are generated we incremen-tally check them against the goal.Whenever we can show that achieving the goal is impossible along a particular se-quence we can prune that sequence and all of its possible extensions from the search space.And we can stop when we?nd a sequence that satis?es the goal.The incremental checking mechanism is accomplished by the logical progres-sion of the goal formula.
Formula Progression The technique of formula progres-sion works by labeling the initial state with the sentence representing the goal,call it.For each successor of the initial state,generated by forward chaining,a new formula label is generated by progressing the initial state’s label us-ing the algorithm given in Table1.This new formula is used to label the successor states.This process continues.Every time a state is expanded during planning search each of its successors is given a new label generated by progression. Intuitively a state’s label speci?es a condition that we are looking for.That is,we want to?nd a sequence of states starting from this state that satis?es the label.The purpose of the progression algorithm is to update this label as we extend the state sequence.It takes as input the current state and the duration of the action that yields the successor state.
The logical relationship between the input formula and output formula of the algorithm is characterized by the fol-lowing proposition:
Proposition3Let M be any MITL model. Then,we have for any formula of MITL,M if and only if M Progress. This proposition can easily be proved by utilizing the def-inition of MITL semantics.
Say that we label the start state,,with the formula ,and we generate new labels using the progression al-gorithm.Furthermore,say we?nd a sequence of states,
,starting at state that satis?es’s label. Then a simple induction using Proposition3shows that the sequence leading from to followed by the sequence, i.e.,,satis?es.The progression al-gorithm keeps the labels up to date:they specify what we are looking for given that we have arrived where we are. From this insight we can identify two important features of the formula progression mechanism.First,if we?nd any state whose idling satis?es its label,we have found a correct plan.
Proposition4Let be a sequence of states generated by forward chaining from the initial state to. For each state let its label be.Let the labels of the states be computed via progression,i.e.,for each state in the sequence
Progress
Inputs:A state,and a formula.
Output:True if the state sequence,where time increases without bound,satis?es.False otherwise.
Algorithm Idle(,)
Case
1.contains no temporal modalities:
if return TRUE
else return FALSE
2.:return Idle Idle
3.:return Idle
4.:return Idle
5.U:
if return FALSE
else if return Idle
else return Idle Idle
6.:return Idle
7.:return Idle
Table2:The idling algorithm.
Then iff
.
The proof of this proposition follows directly from Propo-sition3.
Since is a formula specifying the goal,this propo-sition shows that the plan leading to satis?es the goal. Hence,if we have a method for testing for any state and any formula MITL whether or not, we have a termination test for the planning algorithm that guarantees soundness of the algorithm.We will describe an appropriate method below.
Furthermore,as long as the search procedure used by the algorithm eventually examines all?nite sequences of states the planning algorithm will also be complete.
The second feature of formula progression is that it allows us to prune the search space without losing completeness.As we compute the progressed label we simplify it by processing all TRUE and FALSE subformulas.For example,if the label TRUE is generated we simplify this to.If any state receives the label FALSE we can prune it from the search space,thus avoiding searching any of its successors.From Proposition3we know that this label speci?es a requirement on the sequences that start at this state.No sequence can satisfy the requirement FALSE,hence no sequences starting from this state can satisfy the goal and this state and its successors can be safely pruned from the search space. Termination As indicated above,we can detect when a plan satis?es the goal if we can detect when an idling state satis?es its label.This computation is accomplished by the algorithm given in Table2.
Proposition5Idle returns TRUE if and only if .That is,Idle detects if an idling state satis?es a formula.
The Planning Algorithm Given the pieces developed in the previous sections we specify the planning algorithm pre-sented in Table3.The algorithm labels the initial state with the goal and searches among the space of state-formula pairs. We test for termination by running the Idle algorithm on the Inputs:An initial state,and a sentence MITL specifying the goal.
Returns:A plan consisting of?nite sequence of actions. Algorithm Plan(,)
1.Open.
2.While Open is not empty.
2.1Remove an element of Open.
2.2if Idle Return().
2.3Successors Expand().
2.4For all Successors
2.4.1Progress duration.
2.4.2if FALSE
2.4.2.1Parent().
2.4.2.2Open Open.
Table3:The planning algorithm.
state’s formula.To expand a state-formula pair we apply all applicable actions to its state component,returning all pairs containing a successor state and the action that produced that state(this is accomplished by Expand()).We then compute the new labels for those successor states using the Progress algorithm.
It should be noted that we cannot treat action sequences that visit the same state as being cyclic.If we are only looking for a path to a?nal state,as in classical planning,we could eliminate such cycles.Goals in MITL,however,can easily require visiting the same state many times.Nevertheless, we can view visiting the same state-formula pair as a cycle, and optimize those cycles using the standard techniques. Intuitively,when we visit the same state-formula node we have arrived at a point in the search were we are searching for the same set of extensions to the same state. Proposition6The planning algorithm is sound and com-plete.That is,it produces a plan that is correct for given (De?nition2),and so long as nodes are selected from Open in such a manner that every node is eventually selected,it will?nd a correct plan if one exists.
This proposition follows from the soundness of our termina-tion test(Proposition4).
We have implemented the planning algorithm as an ex-tension of the TLP LAN system[Bac95].This allowed us to utilize many of the features already built into the TLP LAN system,including full support of the A DL formalism[Ped89] for specifying actions.
Example and Empirical Results
Types of Goals
The domain we used is a variant of the classical S TRIPS robot rooms domain[FN71].The con?guration of the rooms is illustrated in Figure1.In this domain there are objects and a robot,which can be located at any of the2locations in the corridor,or,or any of the4rooms. The robot can move between connected locations,it can
Precondition Deletes
open opened
close closed
grasp holding
release handempty
move at
holding
at
We must appropriately rewrite in terms of bounded
quanti?cation.Also it is not dif?cult to see that multiple variables
in can be handled by additional quanti?ers.Similar remarks hold
for encoding restore.
in the initial state.Notice that the semantic distinction between restore and don’t-disturb goals is made clear by our formalism.Restore goals use while don’t-disturb goals use.That is,restore goals allow the violation of during the plan,as long as these conditions are eventually restored in the?nal state.
Both of these conditions are limited special cases.MITL can express much more than this.For example,say that we want to constrain the robot to close doors that it opens. We cannot place a don’t-disturb condition closed,as this would prohibit the robot from moving into rooms where the doors are closed.If we specify this as a restore condition, the robot might leave a door opened for a very long time until it has?nished the rest of its plan.In MITL,however we can use the formula
connects
(1)
at closed open
at closed
This formula speci?es that if the robot opens a closed door(closed open),then it must go through the door(at)and then it must close the door (closed).Hence,the robot is forced to be tidy with respect to doors:it only opens doors for the purpose of mov-ing through them,and it closes the doors it opens behind it.
Timing Deadlines:MITL is also capable of expressing goals with timing conditions.For example requires the condition be achieved within ten time units. Empirical Results
We have tested different goals from each of the cate-gories mentioned above.Most of the plans were generated from the initial state in which at obj1,at obj2, at robot,handempty,object obj1,object obj2,and all of the doors are opened.
G1:From this initial state we set the goal to be at robot at obj1.This corre-sponds to the classical goal at robot at obj1. The planner generates the plan:move, grasp obj1,move,release obj1,move, move.It took the planner22sec.,expanding636 worlds to?nd this plan.
G2:From the same initial state we set the goal to be object at handempty.Now the plan-ner generates the plan:move,move, grasp,move,release.In this case it has generated a plan for a quanti?ed goal.This plan takes the planner3sec.,expanding126worlds to?nd the plan.
that has been able to produce polynomial time behavior in these domains.
As a simple illustration of the power of this using search control consider the following trivial search control formula: at at at
object holding holding
holding
If we conjoin this formula with any other goal,the planner will prune sequences in which(1)the robot grasps an object and then immediately releases it,and(2)the robot moves away from a location and then immediately moves back.For this domain these sequences serve no purpose even in plans where the robot must visit the same state more than once. Conjoining this formula with the example goals given above we obtain the following speedups.
Time New-Time
2212
212693
7718
428438
1207.75
The columns give the planning time and the number of worlds expanded,before and after we add the search control formula. Note in particular,the speedups obtained on the harder prob-lems.Furthermore,it should be noted that this is only the simplest and most obvious of control formulas for this do-main.
References
[AFH91]Rajeev Alur,Tomas Feder,and Thomas Henzinger.
The bene?ts of relaxing punctuality.In Tenth Annual
ACM Symposium on Principles of Distributed Comput-
ing(PODC1991),pages139–152,1991.
[AKRT91]J.Allen,H.Kautz,Pelavin R.,and J.Tenenberg.Rea-soning about Plans.Morgan Kaufmann,San Mateo,
CA,1991.
[Bac95]Fahiem Bacchus.Tlplan(version2.0)user’s manual.
Available via
the URL ftp://logos.uwaterloo.ca:/pub/bacchus/tlplan-
manual.ps.Z,1995.
[BK95]Fahiem Bacchus and Froduald https://www.doczj.com/doc/498543096.html,ing tem-poral logic to control search in a forward chaining
planner.In Proceedings of the3rd European Work-
shop on Planning,1995.Available via the URL
ftp://logos.uwaterloo.ca:/pub/tlplan/tlplan.ps.Z. [BKSD95]M.Barbeau,F.Kabanza,and R.St-Denis.Synthesizing plant controllers using real-time goals.In Proc.Thir-
teenth International Joint Conference on Arti?cial In-
telligence(IJCAI’95),pages791–798,1995.
[CT91]K.Currie and A.Tate.O-plan:the open planning architecture.Arti?cial Intelligence,52:49–86,1991.
Unit one My name is Gina. 1．话题： Making new friends 2．功能目标： 1）使学生学会用英语介绍自己。 2）使学生学会用英语和别人打招呼。 3．文化目标： 1）让学生初步体会用英语和别人交流的感受，培养学生学习英语的兴趣。 2）通过对英文名字的介绍，对比中西方表达名字的不同，了解中西方文化的差异。 3）通过对西方一些成功人士英文名字的介绍，使学生了解成功是由不懈的努力和艰苦的奋斗得来的。 4）通过介绍英文名字的由来，使学生产生对西方文化的好奇及兴趣。 4．认知目标： 1）词汇： clock, hello, hi, and, question, answer, look, first name, last name 2) 语法项目：一般现在时be的用法 以特殊疑问词what开头的问句 形容词性物主代词my, your, his, her What’s=What is I’m=I am My name’s=My name is 3) 语言目标： What’s your name? My name is Gina. I’m G ina. Nice to meet you. Period one I.教材分析：这是课本的第一部分，题目是My mane is Gina. 有很多学生第一次学习接触英语，为了调动他们的积极性，不产畏难情绪，尽量简单的处理教材。 Language topic: What’s your name? My name is Gina. Language strategies: Talk about the names Main vocabulary: name is, meet, his, first name, last name, answer, boy , girl. II.语言结构: present tense to be, what questions, Possessive adj. my, your, his, her. 。 III.语言功能:询问姓名及怎样回答. IV.活动设计：采访调查What’s your first name? V. Teaching steps: Step 1:Greetings. Step 2: Words: Learn the new words A: first, listen to the recorder, the students read after the recorder. B: read the new words after the teacher. Step 3: 1a Write English words for the things in the picture. How many things do you know?
八年级上册专题练习 一. 单项选择： 1. My grandparents live ______ 36 Xiaoyuan Road. A. in B. on C. at D. / 2. I don’t like this fish soup because it tastes_______ A. delicious B. good C. pretty D. salty 3. He’s a basketball fan and he wants to see Yao Ming _______ one day. A. himself B. him C. his D. he 4. ______of the students in the class would like to go to school by bike in such terrible weather. A. All B. Both C. Neither D. None 5. —Do you still have a headache, Bill? —No, it’s______. I’m all right now, mum. A. gone B. run C. left D. dropped 6. —Bob, would you like to come to our dinner party? ____________. A. Yes, I would B. Yes, I’d love to C. No, I wouldn’t D. No, I don’t go 7. You ________ be late for class. You must get to school on time. A. must B. can C. can’t D. may 8. Don’t forget to write letters. All of us hope to hear from you________. A. quickly B. fast C. soon D. quick 9. I can not afford ________ a piano. A. buying her B. her to buy C. buy for her D. to buy her 10. Is this ________ the railway station? A. the way for B. the way to C. the way from D. going for 11. It was clever______ you _______ this question. A. for; to answer B. of; to answer C. of; answer D. for; answer 12. I’ll let you know as soon as he______ A. is coming back B. will come back C. comes back D. come back 13. I am sorry for ________ yesterday. A. wake you up B. wake up you C. waking up you D. waking you up 14. Grandma couldn’t keep up with me and I had to stop ________ her. A. waiting for B. to wait for C. to waiting D. waited 15. —Look! This sweater is beautiful. A. Why not try it on B. Why not trying on it C. Why not try on it D. Why not trying it on 二. 完型填空： Dear Mum, I’m writing to you from America. You don’t know how happy I am. I __1___ many new friends here. They are very helpful. Last week I got a fever and ___2___ stay in bed for three days. They helped me a lot with my lessons. Cathy is one of my ___3___ friends at school. She has a round face ___4___ a pair of glass?es. She is two months older than I am, but a little ___5___ . She works hard and
人教版《思想品德》七年级(上)全册教案 第一单元笑迎新生活 第一课珍惜新起点——新学校，新同学 一【说教材】 【教学目标】 1、知识目标：了解自己所在的学校，了解中学与小学的不同之处，包括同学、老师、环境、课程等方面发生的新变化。认识中学的学习环境，尽快适应中学新的学习生活。 2、能力目标：通过“新学期、新面貌”的活动，更深入地思考怎样尽快适应中学新的学习生活，制定一个切实可行的计划，并与同学分享。 3、情感、态度、价值观目标：充分认识展现在自己面前的中学生活的各种新变化，尽快调整好自己的心态，以新的精神面貌迎接新的学习生活。 【教学重点】分析中学与小学的不同之处。 【教学重点】如何引导学生尽快适应中学生活。 二【学生】 三【教法】 授课理念：坚持“还课堂于学生”的教学理念,立足于学生的“学”,在开放性教学中,结合学生的个人知识、直接经验和真实体验。构建“自主—探究---合作”的学习模式。 授课思路：本框内容分目，首先学习第一课第一框第一目“初中生活新体验”。学生从小学到初中是其人生的一个重要转折。中学和小学相比，无论在同学、老师、校园环境还是在课程设置上都有很大变化，通过互动游戏把这些变化带给七年级新生新鲜感，让学生体验初中新生活，珍惜新起点。其次以活动课的形式，学习第二目“珍视新友谊”和第三目“与新朋友结伴同行”。因为在进入中学最初的阶段，能否处理好人际关系，也是能否适应中学生活的问题。而在这一方面可能存在比较明显的个体差异，有的能迅速地和同学建立良好的人际关系，有的也许历时三年也未能建立起来。因此对这一时期的学生进行相关知识的辅导是十分必要的。 【课前准备】 四【教学程序】 （一）导入新课 教师在正式组织本单元教学前，先让学生翻翻上册的教科书，让他们与同桌分享看到如此精美、漂亮的教科书的心情。并利用好本书的“前言”，简明扼要地介绍思想品德课程将要学习的内容，本书的特点及学习上要注意的问题。以上处理应该是生动活泼的，而不是严肃的、说教的。让学生对你所教的学科留下美好的印象，憧憬着将来的学习，这样你就成功了一半。（二）新课讲授 结束语 五【课堂延伸与反思】 （一）初中生活新体验（板书） （一）初中生活新体验（板书） （学生发言或小组讨论，提示：比如军训生活的紧张有趣、心理压力的增大、学习内容的复杂等等。） 师：同学们刚才都谈到了新生活带给自己的兴奋和不安。兴奋是因为那么多新奇有趣的事物等着我们去认识，不安是因为还有那么多我们未曾了解的问题。我想，同学们肯定急于想知道的东西很多，比如我们的学校是一个什么样的学校，它对我们有什么要求和期望？比如各科老师是什么样的性格，是和蔼可亲还是严厉苛刻？再比如班里的同学，谁爱学习，谁
专项训练题 短文填空篇 Unit 1 一、基础篇 A. What do students do at the Green High School ? Here are the results of the students activity survey ______(1) Green High School. Most students exercise three _____ (2) four times ______(3) week. Some students exercise once or twice a week. Some students are very active ______(4) exercise every day. ______(5) for homework, most students do homework every day. ______(6) students do homework three or four times a week. No students do homework once or twice a week. The results _____(7) “watch TV' are interesting. All students watch TV ______(8) or twice a week. Some students watch TV three or four ______(9) a week, ______(10) most students watch TV every day. 1._______ 2.________ 3. ________ 4. _________ 5. _________ 6.________ 7. ________ 8.________ 9. _________ 10._________ B. I'm pretty healthy. I exercise every day, usually _____(1) I come home _____(2) school. My _____(3) habits are pretty good. I try _____(4) eat a lot of vegetables. I eat fruit and drink milk every day. I never ______(5) coffee. _____(6) course, I love junk food too, and I eat it two _____(7) three times a week. Oh, and I sleep nine hours every night. So you _____(8), I look after my health. And my healthy lifestyle _____(9) me get good grades. Good food and exercise help me to study _______(10). 1._______ 2.________ 3. ________ 4. _________ 5. _________ 6.________ 7. ________ 8.________ 9. _________ 10._________ 二、综合提高篇 A. John and Sally are in ______(1) same school. Their home is far _______(2) school. They get up early _____(3) the morning and _______(4) the bus to school after breakfast. Sometimes they ______(5) lunch in the restaurant. John likes hamburgers ______(6) meat. ______(7) Sally likes vegetables and fruit. After lunch John goes to _____(8) volleyball and Sally goes to the classroom. She _____(9) English songs or listens _____(10) music.
七年级上册政治知识点 第一课：珍惜新起点 1.初中生活新体验：这里的“新”主要包括新的校园，新的老师和同学，新的友谊，新的班集体，新的学习天地等等。 2.如何创建优秀集体？（我们应怎样做才能使我们的这个集体更加优秀） （1）共同的目标，前进的动力。（2）各尽所能、发挥所长、奉献集体。（3）团结协作，互助前行。 3.团结协作对创建新集体的重要性？ 团结协作的力量是无穷的，只有团结协作，相互配合，我们才能奏出最美的班级乐章，在集体中不断成长。 4、如何为班级做贡献？ ①积极参加班集体的活动；②积极参加班级的劳动；③遵守纪律，为班争光；④平时上课认真听讲，努力完成老师布臵的作业；⑤为班级建设出谋划策。 第二课：把握学习新节奏 1.初中学习新特点有哪些？ （1）课程门类增多。（2）各门课程的内容也逐渐系统化和深化。（3）学习任务变重。（4）老师的授课方式侧重于指导和启发。 2.怎样做学习的管理者？（初中有效学习的方法与策略或如何做学习的主人）（1）学会自主学习。（2）提高学习效率。（3）科学安排时间。（4）寻求和探索适合自己的学习方法.（5）适应不同老师的教学风格。 3.如何提高学习效率？ （1）课前认真预习。（2）上课时专心听讲。（3）课后及时复习。 4.如何制定有效学习计划？（在制定学习计划时，需要遵循的原则有哪些？）（1）从实际出发。（2）区分轻重缓急。（3）可操作。（4）要有弹性。（5）适合自己。（6）自我激励。 5、如何才能将学习成绩提高？ ①自己一定要勤奋、努力；②寻求和探索适合自己的学习方法③勤问。 6.了解自己的学习方式。每个人都有自己的学习方式，不同方式各有特点。我们要不断摸索新学习方式，在不断尝试的努力过程中，找到适合自己的学习方法，把握学习的新节奏。 7、一样的学习方法，却产生了两种不同的结果。为什么？ 在不同的学习阶段，学习方法也不一样。 8.为什么说学习是苦乐交织的？ 学习是一个探究和发现的过程，需要克服困难，刻苦努力。也正是在这个过程中，我们不断地发现自身的潜能，获得一种不断超越自己的快乐。 9.享有学习的权利和机会本身就是一种快乐。我们要珍惜学习的权利与机会。 10.为什么说探索奥妙，获得新知也是一种快乐？ 当我们在学习中逐渐了解世界奥秘的时候，当我们的好奇心在学习中获得满足的时候，当我们通过学习发现自己潜能的时候，学习的快乐就洋溢在我们的心中。 11、好的学习方法建议： ①做好课前预习；②上课认真听讲；③做好课堂笔记；④独立完成作业；⑤加强课外阅读；⑥建立错题本，并经常翻看错题本；⑦课后及时复习；⑧勤学好问，
Unit1 my[mai]我的 name['neim]名字，名称 is[iz]是(be的现在式第三人称 clock[kl?k]时钟 I[ai]我 am[?m]是 nice[nais]美好的；令人愉快的 to[tu:]用于与动词原形一起构成动词不定式prep. meet[mi:t]遇见，相逢 you[ju:]你(们) what[w?t]什么 your[j?:]你(们)的 hello[he'l?u]喂，哈罗 hi[hai]嗨！(表示问候或用以 his[hiz]pron.他的 and[?nd]conj.和，与 her[h?:]pron.她的 question['kwest??n]n.问题，疑问 answer['ɑ:ns?]v.回答 look[luk]v.看 first[f?:st]a.第一的 first name[f?:st neim]名字 last[lɑ:st]a.最后的；最近的 last name[lɑ:st neim]姓氏 boy[b?i]n.男孩 girl[g?:l]n.女孩 zero['zi?r?u]num.零 one[w?n]num.一 two[tu:]num.二 three[θri:]num.三 four[f?:]num.四； five[faiv]num.五； six[siks]num.六 seven['sev?n]num.七 eight[eit]num.八； nine[nain]num.九； telephone['telif?un]n.电话 number['n?mb?]n.号码；数字 telephone number[telif?un n?mb?]电话号码phone[f?un]n.电话
七年级上册英语单词表Starter Unit1 1.好的 2.早晨上午 3.早上好 4.嗨喂 5.你好喂 6.下午 7.下午好 8.晚上傍晚 9.晚上好 10.怎样如何 11.是 12.你你们 13. 你好吗 14.我 15. 是 16.健康的美好的 17.感谢谢谢 18 .好可以 19. 硬黑 20.光盘激光唱片
Starter Unit2 1. 什么 2.是 3.这这个 4.用以 5.英语英格兰人英语的 6.用英语 7.地图 8.杯子 9.尺尺子 10.笔钢笔 11.桔子 12.夹克衫短上衣 13. 钥匙 14. 被子床罩 15. 它 16. 一 17. 那那个 18. 用字母拼拼写 19. 请 20.美国全国篮球协会
21. 停车场停车位 22. 千克公斤 Starter Unit3 1. 颜色 2. 红色 3.黄色 4. 绿色 5.蓝色 6.黑色 7.白色 8.紫色 9.棕色褐色 10.指已提到或者易领会到的人或事物 11.现在目前 12.理解明白 13.能会 14. 说讲 15.我的 16. 小号的 17. 大号的 18 .中号的
19.不明飞行物 20.中国中央电视台 第一单元单词及短语 1. 名字n. 2. 好的令人愉快的adj. 3. 用于与动词原形一起构成动词不定式 4. 遇见相逢v. 5.也又太_ 6你的你们的pron. 7.女士不指明婚否 8.他的pron. 9.和又而conj. 10.她的pron. 11.是的可以interj 12 她pron. 13. 他pron. 14.不没有不是interj 15.不没有adv. 16 零num. 17. 一num. 18. 二num.
七年级上册思想品德期中复习提纲
目要求（注意设问的答案是否有很多个小点或角度），把材料划分层次进行分析解答。 2、联系课本、材料和实际：所学知识联系题目材料和生活实践是中考考查的重点。 （1）联系课本知识：即运用课本知识答题，这要求我们在平时的学习过程中，要能使各知识点能够联系起来，例如可以按照“What，Why，How”三种问题类型对知识点进行分类；也可以按照知识之间的联系编制表格或结构图梳理知识脉络。 （2）联系材料：一般是在答出一个知识点后，在后面简要联系材料进行一两句话的说明。可借用材料词局来体现对材料的理解。 （3）联系实际：一般联系实际的是答感悟和措施，这类题大多数课本是没有完整答案，但是我们结合实际说明时一定要有课本知识。 3、构建答题思路、组织答案：“先准而后多” （1）构建思路：从与题目最直接有关的知识入手答题，围绕设问和材料回忆和选取“知识角度”，而不是将所有相关知识点都罗列出来，在“准”的基础上再“多”：多知识角度完善答案。可以一边看题目和材料一边简单在旁边写答案关键词和要点。 （2）组织答案：按规范答题——表述的规范、格式的规范。注意尽量按照设问的原句来开始答案，按照题型的答题模式来组织答案，在答案合适的地方插入一些词语，使答案逻辑通顺；学科术语要准确使用；不同主语角度或不同知识角度时，要注意分点、另起段落，即使是同一角度有不同知识点，也要分成小点。（三）不同题型的解题方法和答题模式 1、概括说明类题：（以上材料（图表、漫画）表明、说明、反映、体现了什么） =现象（材料、图表、漫画概括）+本质（体现的课本知识点） 2、谈启示=What+How 3、评析、评论、评价、谈理解、认识、感想、感悟、体会、看法、感动、理由等 =What(说明什么观点、问题；判断加行为定性)+Why（结合课本内容谈，代入法）+How（注意主语角度） 4、为什么=必要性（原因、依据，未实施前，为了解决现状不足的需要）+重要性（实施后的意义） 重要性、意义、作用=“有利于”开头，经济、政治、社会意义或对个人、
七年级上册 第一课我是中学生了 1、怎样才能成为好朋友？（相识的人成为好朋友的条件）P4 成为朋友需要彼此之间的真诚和心灵的沟通 2、什么是班风？P4 班风是指一个班级稳定的、具有自身特色的集体风范，是一个班级中大多数学生在思想、学习、生活等方面的共同倾向。 3、良好班风的作用是什么？P4 良好的班风能引导全班学生的言行，对个人的健康成长和优秀班集体的形成起着重要作用。 4、优良班风主要体现在哪些方面？P5 优秀的班风主要体现在精神风貌、学习氛围、同学关系等许多方面。 5、怎样形成良好的班风？P5—P6 优良的班风的形成要靠每一个人用行动去创造、去维护。爱心是营造良好班风的情感基础。 6、怎样做一个阳光少年？P9 做好角色转换；做好心理上的调整；主动调整生活节奏。做一个阳光少年，实质上就是要在上合生活中保持一种积极的心态。 第二课天生我才必有用 1、如何正确认识和评价自己？ ①正确的认识和评价自己关键是要实事求是，既不要高估自己，也不要低估自己。接受自己，诚实的对待自己的长处和短处，扬长避短 ②正确的认识自己，努力发现和发挥自己的潜能。 ③正确的认识和评价自己，必须掌握科学的方法，常用的有自我观察、自我反省、相互比较、咨询他人、科学鉴定等。 2、为什么要正确认识和评价自己？（正确认识和评价自己的意义）P16 ①正确的认识和评价自己，是一个人心理成熟的标志之一。 ②充分的认识自己、相信自己，超越自己是每一个成功者基本素质之一。 3、什么是自信? P17 自信是一个人成才与成长不可缺少的一种重要的心理品质，是生活中获得成功和快乐的重要因素。 4、自信心是在对自己的自我肯定和充分估计的基础上，相信自己力量的一种心理状态。 5、怎样才能获得自信？P17 ①只要找出自己的优势，挖掘自己的潜能，在现有的基础上进步，我们就会逐步树立起自信心，不断走向成功。 ②超越自卑 6、自卑的含义P18 自卑感是一种消极的自我评价和自我意识，是个人认为自己在某些方面不如他人而产生的消极情感。 7、怎样克服自卑? ①全面的认识自己 ②发现和发挥自己的长处 ③勇于正视自己的不足 第三课知耻近乎勇
I【词汇】 家庭成员：family, grandfather, grandpa ,grandmother, grandma, grandparent(s), father, dad, mother, mom(mum),parent(s) ,son, brother, daughter, sister, uncle, aunt ,cousin 代词： this复数 these, that复数those， who, he复数they 同义词：picture-photo, bye-goodbye, mom-mum, parent-father/mother ,grandparent-grand-mother/ grandfather 对应词：grandfather-grandmother, grandpa-grandma, father-mother ,dad-mom/mum, daughter-son, brother-sister, uncle-aunt, these-those, boy-girl 其他： have, day,here, of, next, dog, well, oh II.【词组】 1.my parents 2.my sister 3.my brothers 4.my family 5.your parents 6.my father 7.my mother 8.in the first photo 9.in the next picture 10.a family photo=a photo of family11.in my family 12.the name of my dog=my dog’s name III.【句型】 介绍人物： 1.单数句：This / That is... 这是...... (那是......) 2.复数句：These/Those are...这些是......(那些是......) Who引导的特殊疑问句及其答语形式： 3.单数句-Who’s he/she ? 他是谁？（她是谁？） -He’s/She’s...他是......(她是......) 4.复数句-Who’re these/those/they? 这些（那些、他们）是谁？ -They’re...他们是...... 5.Here is/are +主语. 这有......,这是...... IV．【交际用语】 1.This/That is my ... 2.-Nice to meet you. -Nice to meet you, too. 3.-Who’s he/she?-He /She is my ... 4.–Who’re these/those/they?-They’re my... 5.–Have a good day!-Thanks！You ,too. V．【语法】 1.指示代词
八年级英语上册完形填空专项练习 1 When Mr Smith retired(退休), he bought a small 1 in a village near the sea. He2 it and hoped to live a quiet life in this house. But to his great surprise, many tourists came to see his house in summer holidays, for it was the most 3 building in the village. From morning to night there were 4 outside the house. They kept looking into the rooms through the windows and 5 of them even went into Mr Smith’s garden. This was too much for Mr Smith. He decided to ask the visitors to 6 . So he put a notice on the window. The notice said, “If you want 7 your curiosity(好奇心)，come in and look round. Price: twenty dollars.” Mr Smith was sure that the visitors would8 coming, but he was wrong. More and more visitors came and Mr Smith had to 9 every day showing them around his house.“I came here to 10 not to work as a guide（导游）,”he said angrily. In the end, he sold the house and moved away. （）1.A. garden B. shop C. house D. school （）2. A. liked B. hated C. sold D. built （）3. A. big B. interesting C. small D. clean （）4. A. children B. students C. parents D. tourists （）5. A. no B. none C. many D. much （）6. A. come B. leave C. stay D. play （）7. A. to satisfy B. satisfy C. to satisfying D. satisfying （）8. A. go on B. stop C. continue D. not （）9. A. take B. cost C. spend D. pay （）10. A. play B. work C. watch D. retire 2 One day, a boy had a fight with one of his classmates. Then he went to his 11 and told him his sto ry angrily. “He is really bad,” the boy said, “and I 12 him.” The grandfather said, “13 me tell you a story. When I was a boy, I too, sometimes hated others for what they did. ..” As the boy 14 carefully, the grandfather went on, “There are always t wo tigers inside my heart. One is 15 and kind. He gets on well 16 everything around him. But 17 is bad and unfriendly. Even the smallest thing will make him angry. He fights with everyone all the time, and for no reason. He can’t think carefully 18 he always hates others. It is difficult to live with these two tigers inside my heart. They both try to control(控制) me. ” The boy looked into his grandfather’s 19 and asked, “20 tiger always controls you, Grandfather?” The old man said slowly and serio usly, “The one that I feed. I always feed the good and kind tiger, so I never hate others and seldom(很少) get angry now. ”
g o f o r i t七年级英语上 册U n i t单元整合 文档编制序号：[KK8UY-LL9IO69-TTO6M3-MTOL89-FTT688]
I【词汇】 家庭成员：family, grandfather, grandpa ,grandmother, grandma, grandparent(s), father, dad, mother, mom(mum),parent(s) ,son, brother, daughter, sister, uncle, aunt ,cousin 代词： this复数 these, that复数those， who, he复数they 同义词：picture-photo, bye-goodbye, mom-mum, parent- father/mother ,grandparent-grand-mother/ grandfather 对应词：grandfather-grandmother, grandpa-grandma, father-mother ,dad-mom/mum, daughter-son, brother-sister, uncle-aunt, these-those, boy-girl 其他： have, day,here, of, next, dog, well, oh II.【词组】 1.my parents sister brothers family parents father mother the first photo the next picture family photo=a photo of family my family name of my dog=my dog’s name
七年级思想品德上册复习提纲 第一单元走进中学 1.2－认识新伙伴（主题词：克服闭锁心理，热情开朗，友谊） 1、什么是闭锁心理？P11 闭锁心理也叫含羞草心理。一方面渴望与别人沟通了解，一方面把自己封闭起来。表现为沟通时紧张、拘谨的心理状态。 2、闭锁心理有什么危害？ 难以适应新环境，危害身心健康；影响学习；不利于建立良好的人际关系。闭锁心理如果不及时克服，就容易发展为自闭心理，把朋友拒之门外。 3、怎样克服闭锁心理？ ①因为人际交往是双向性的。如果我们想获得理解，就要袒露自己的真实思想；如果我们渴望人际交往中的温暖，就要先信任对方。②要主动适应新环境，建立自信心；（3）培养热情开朗的性格。 4、如何才能结识新伙伴，在同学中建立新的友谊呢？（结识新伙伴的方法）（1）打开“心锁”，克服闭锁心理；（2）养成热情开朗的性格；（3）建立新友谊。 5、热情开朗的性格的作用：（为什么要养成热情开朗的性格？） ①热情开朗的性格会对人的生理、心理产生积极的影响； ②容易被人接受，建立良好的人际关系； ③在集体和社会中营造融洽的人际氛围，并且容易得到快乐。 6、怎样培养热情开朗的性格？P11~12 （1）每天保持一种好心情： （2）对老师、对同学态度友好； （3）与人交往主动、豁达，不计较个人得失。 7、朋友和友谊对我们有什么意义？（朋友的作用？）P12 （1）朋友和友谊是人生最富贵的财富。 （2）朋友能温润我们的心田，能分享我们的欢乐，能驱散我们的烦恼，能给我们支持和力量。 8、怎样建立真正的友谊？P12 （1）友谊要靠真诚和宽容“播种”； （2）友谊要靠平等与尊重“护理”； （3）友谊要靠信任与奉献“浇灌”； （4）善交益友，不交损友，乐交诤友。
义务教育教科书《英语（Go for it!）》七年级上册单词表 Starter Unit 1 good 好的 morning 早晨；上午 Good morning! 早上好！ hi 嗨；喂 hello 你好；喂 afternoon 下午 Good afternoon! 下午好 evening 晚上；傍晚 Good evening! 晚上好！ how 怎样；如何 are 是 you 你；你们 How are you? 你好吗？ I 我 am 是 fine 健康的；美好的 thanks 感谢；谢谢 OK 好；可以 HB （铅笔芯）硬黑 CD 光盘；激光唱片 BBC 英国广播公司 Alice 艾丽斯（女名） Starter Unit 2 what 什么 is 是 this 这;这个 in 用(表示使用语音；材料等)用；以 English 英语；英格兰的；英语的 in English 用英语 map 地图 cup 杯子 ruler 尺；直尺 pen 笔；钢笔 orange 橙子 jacket 夹克衫 key 钥匙 quilt 被子；床罩 it 它 a(an) (用于单数可数名词前，表示未曾提到的) 一（人、事、物）that 那;那个 spell 用字母拼；拼写 please 请
NBA （美国）全国篮球协会 P 停车场；停车位 Kg 千克；公斤 Starter Unit 3 color (=colour) 颜色 red 红色(的) yellow 黄色(的) green 绿色(的) blue 蓝色(的) black 黑色（的） white 白色（的） purple 紫色（的） brown 棕色（的）；褐色（的） the 指已提到或易领会到的人或事物 now 现在；目前 see 理解；明白 can 能；会 say 说，讲 my 我的 S （尤指服装的尺码）小号的 M （尤指服装的尺码）中号的 L （尤指服装的尺码）大号的 UFO 不明飞行物 CCTV 中国中央电视台 Unit 1 name 名字；名称 nice 令人愉快的；宜人的 to 常用语原形动词前，表示该动词为不定式 meet 遇见；相逢 too 也；又；太 your 你的；你们的 Ms. （用于女子的姓氏或姓名前，不指明婚否）女士his 他的 and 和，又，而 her 她的 yes 是的；可以 she 她 he 他 no 不；没有；不是 not 不；没有 zero 零 one 一 two 二 three 三 four 四 five 五
八年级上册英语专项训练之单选部分 二、单项填空（本大题有20小题，每小题1分，共20分） 26. ___________ do you exercise? Twice a week. A. How often B. How far C. How long D. How much 27. The boy is too fat. He eats _________ food. A. much too B. too much C. many too D. too many 28. She looks ________ because she has a _________ vacation. A. relaxed, relaxed B. relaxing, relaxing C. relaxed, relaxing D. relaxing, relaxed 29. ___________ did it _______ you to finish the work? One hour. A. How often; spend B. How far; spend C. How long; take D. How much; take 30.—Can you come to my birthday party tomorrow evening? —________________. A. I can. B. That’s all right. C. Sure, I’d love to. D. That’ right. 31. The two boys ___________ Chinese. A. are both B. both are C. are all D. all are 32. Some of my friends like playing computer games, but I am _______them. I like to reading books. A. the same as B. different from C. interested in D. angry with 33. I think you should eat _________ food and take ________ exercise. A. a little; a few B. less; more C. many; much D. much; many 34. She is leaving Zhuhai ___________ Hongkong tomorrow. A. at B. for C. on D. in 35. _______ does your mother go to work every day? She takes the bus. A. When B. Where C. How D. Why 36. –What are you doing _______ Sunday morning? –I’m studying _______ a math test. A. on; in B. on; for C. in; for D. in; in 37. He is ___________ than me at school. A. popular B. much popular C. most popular D. more popular 38. These days Mrs Li looks_______ healthy. A. nice B. pretty C. heavy D. cool