下载文档原格式
把废物变成魔法的英语作文英文回答：Converting Waste into Wonder: Embracing Circularity for a Sustainable Future.Waste management has become a pressing global concern, with landfills overflowing and the environment suffering the consequences. However, what if waste could be transformed into a valuable resource, rather than a burden? Circular economy principles offer innovative solutions to this challenge by emphasizing waste reduction, resource recovery, and material reuse.Product Design for Circularity.The first step towards waste reduction is designing products with circularity in mind. This involves creating products that are easy to repair, recycle, or compost. By choosing materials that can be easily processed and byminimizing the use of hazardous substances, manufacturers can significantly reduce the environmental impact of their products throughout their lifecycle.Waste Collection and Sorting.Effective waste collection and sorting systems are crucial for successful waste management. Advanced technologies such as automated sorting machines canidentify and separate different waste materials, allowingfor efficient recycling and recovery. By encouraging proper waste disposal and providing clear instructions for sorting, communities can significantly reduce the amount of waste going to landfills.Waste Treatment and Transformation.Once waste has been collected and sorted, it can be treated and transformed into valuable resources. Organic waste, such as food scraps and yard waste, can be composted to create nutrient-rich soil amendments. Incineration, gasification, and pyrolysis are advanced technologies thatcan convert non-recyclable waste into energy or other useful products.Waste Reduction and Reuse.Waste reduction strategies focus on preventing waste from being created in the first place. This can involve measures such as reducing packaging, promoting reusable products, and encouraging repair and sharing. By adopting these practices, individuals and businesses cansignificantly reduce their waste footprint.Benefits of Embracing Circularity.The benefits of embracing circular economy principles for waste management are numerous. Reducing waste diversion to landfills not only protects the environment but also conserves natural resources and reduces greenhouse gas emissions. By recovering valuable materials from waste, we can reduce our reliance on virgin resources and create new economic opportunities.Conclusion.Transforming waste into wonder is not merely a sloganbut a necessary step towards a sustainable future. By embracing circular economy principles, we can turn ourwaste into a valuable resource, reducing our environmental impact and creating a more circular and sustainable economy.中文回答：将废物变成魔法，拥抱循环经济，实现可持续发展。
毕业设计(论文)外文资料翻译系（院）：电子与电气工程学院专业：电气工程及其自动化姓名：学号：外文出处：2007 HERE COME THE... CLEANER,GREENER CARS附件： 1.外文资料翻译译文；2.外文原文。
附件1：外文资料翻译译文2007年来了...清洁，环保汽车一个全新的领域，在柴油发动机上使用电气混合燃料电池。
这个说法是针对混合动力汽车：美国人爱他们，不过只是猜测。
一些环保人士一直在疑惑，有没有更大的混合电池组，能不能够直接插在墙上进行充电，能不能提供动力让你开车去上班，电力与小型燃气发动机使其变为可能。
这个概念最初是一个环保主义者的梦想，是来自的费利克斯克莱默，他推动了公用事业支持插件的合作。
但现在电动汽车走向市场,就像其他高科技绿色汽车当年发展的情况一样。
清洁汽车新的一天清洁和环保汽车技术正在蒸蒸日上。
可充电混合动力车,在工业发展上展现了比1900年的黄金岁月高很多的研究和开发热情。
当汽油、蒸汽、电动车在市场上进行竞争，许多公司如通用汽车、还在嘲弄像罗杰和我这样的人,是谁扼杀了电动汽车的发展?事实上,美国通用汽车公司是第一个成功制造出了可充电混合动力车的公司,他们使用了一个有趣的新方法。
他们正在研发一种全新的推进系统,在最近的底特律车展上展示,那就是雪佛兰伏特。
随着seesawing对未来石油和汽油价格的不确定性，美国人终于将注意力集中在寻找燃油经济性车辆和展望他们的下一个大型多功能运动型车。
一个由具有很大影响力的公司JD Power and Associates去年夏天对消费者的调查发现，让人吃惊的是有57%的受访者会考虑购买他们的下一个混合动力汽车，有49%的购车者会考虑E85乙醇动力汽车。
另一项由Frost＆Sullivan的调查发现约有80%的人更关注较一年前的燃油价格。
几乎有一半的人说，如果燃油价格持续上涨的话他们会考虑购买更省油的汽车或混合动力汽车。
而从居住在美国的市民的调查中发现，有五分之一的让人印象深刻的说道，他们也开始使用替代交通工具：诸如自行车，步行，公共交通和电动汽车等等。
2008年理工类押题第四部分: 阅读理解08理工新增15篇(五星级文章)第二篇: Electric Backpack c第六篇: Flying the Hyper1 Skies c第七篇: Sugar Power for Cell Phones c第十三篇: Invisibility Ring c第十四篇: Japanese Car Keeps Watch for Drunk Drivers c第十七篇: A Sunshade for the Planet c第十八篇: Thirst for Oil c第二十篇: Explorer of the Extreme Deep c第二十一篇: Plant Gas c第三十三篇: Smart Window b第三十五篇: Where Have All the Bees Gone? b第三十六篇: Listening Device Provides Landslide Early Warning b第四十八篇: ‘Hidden’ Species May Be Surprisingly Common a第四十九篇: Why Humans Walk On Two Legs a第五十篇: Black Hokes Trigger Stars to Self—Destruct a第六部分: 完形填空三篇第二篇: Biological Identification Technologies c第十二篇: Paper or Plastic? b第十三篇: Debate over the Use of Renewable Energy a--------------------------------------孙老师特别提示：亲爱的同学们，如果您真的想通过职称考试，——请牢记下列重要提示阅读理解应试技巧(职称英语)一. 牢牢抓住英文写作的“三步曲”●中心 ●举例说明 ●作者观点或文章结论。
Roadway excavation and support巷道掘进和支护Main contents1 Cross-section of Roadway 巷道断面2 Rock Roadway excavation 岩巷掘进3 Coal Roadway excavation 煤巷掘进4 Roadway support 巷道支护5 Rapid excavation 快速掘进New words about roadway excavationdrive, tunnel, excavate 掘进roadway excavation 巷道掘进drifting 平巷掘进n.sinking 沉井，凿井n.raising 上山掘进n.dip excavation 下山掘进n.1 Cross-section of Roadway1）The shape of roadway section 断面形状 The shape of roadway section depends mainly on strata conditions，the serving life and supporting materials.circular 圆形的井筒Shaft ,horseshoe 马蹄形大巷Main entry , arched,ladder-shaped, trapezoidal 梯形的上山rise,rectangular 矩形的顺槽Gateway 1 Cross-section of Roadway2）The size of roadway crosssection 断面尺寸The size of roadway section depends mainly on its usage.It is dependent on haulage or hoisting equipment or other devices used，and it is modified according to the ventilation requirement.net width；净宽net height 净高net section area 净断面面积。
第40卷 第3期吉林大学学报(工学版) Vol.40 No.32010年5月Journal of Jilin University (Engineering and Technology Edition ) May 2010收稿日期:2009207209.基金项目:“863”国家高技术研究发展计划项目(2009AA11Z215);国家自然科学基金项目(50878094);吉林省高技术产业发展项目([2009]633).作者简介:王云鹏(1966),男,教授,博士生导师.研究方向:交通环境与安全技术.E 2mail :wangyunpeng @通信作者:隗海林(1969),男,教授,博士.研究方向:交通环境与安全技术.E 2mail :khl69@基于侧倾的运输液态危险化学品的罐式半挂汽车危险状态辨识王云鹏,孙文财,隗海林,李世武,杨志发,李谷楠(吉林大学交通学院,长春130022)摘 要:针对运输液态危险化学品的罐式半挂汽车列车,建立了用以反映其运动响应特性的整车侧倾数学模型;基于实车参数在ADAMS 虚拟环境下建立了整车多体系统动力学模型;通过对移线和转向两种典型工况下整车动力学仿真,系统地分析了变装载条件、变车速条件下车辆发生侧翻的规律,得到了列车侧翻阈值。
关键词:交通运输系统工程;危险化学品运输;半挂汽车列车;adams ;侧倾中图分类号:U492 文献标志码:A 文章编号:167125497(2010)0320640205Rollover risk of tractor semitrailer with liquid hazmat tankWAN G Yun 2peng ,SUN Wen 2cai ,KU I Hai 2lin ,L I Shi 2wu ,YAN G Zhi 2fa ,L I Gu 2nan(College of T rans portation ,J ilin Universit y ,Changchun 130022,China )Abstract :Aiming at t he t ractor semit railer wit h liquid hazmat tank ,a mat hematic model of vehicle rollover is built to reflect it s movement response characteristics.Based on t he real vehicle parameter ,a dynamic simulatio n model of multi 2body vehicle is built in virt ual environment of ADAMS.Through t he dynamic simulation under two typical work condition 2lane 2changing and swerving ,t he rollover law is analyzed systematically under t he condition of different vehicle speed and different load ,t he t hreshold values of t he rollover of tractor semit railer are gotten.The research conclusion ismeaningf ul on t he t heoretical guide for t he early warning of t he risk during t he running of t he t ractorsemitrailer wit h liquid hazmat tank.K ey w ords :t raffic and t ransportation systems engineering ;hazmat transportation ;t ractor semit railer ;adams ;rollover 危险化学品公路运输车辆主要是重型罐式半挂汽车列车,尤其是液态危险化学品的运输车辆,其高质心、重载、大体积、轮距相对于车身高度过窄等特点,导致其容易发生侧翻事故。
Equilibrium analysis of macroscopic traffic oscillationsYu (Marco)Nie *Department of Civil and Environmental Engineering,Northwestern University,2145Sheridan Road,Evanston,IL 60208,USAa r t i c l e i n f o Article history:Received 16December 2008Received in revised form 24May 2009Accepted 4June 2009Keywords:Traffic oscillations Wardrop equilibrium Boston traffic equilibrium Freeway bottlenecksa b s t r a c tUsing a simple network model with two parallel links connecting a diverge and a merge,this paper studies under what conditions traffic oscillations may be initiated and propa-gated in a traffic stream,specially at freeway bottlenecks.Drivers are assumed to minimizeeither the experienced or instantaneous travel times,and in doing so,they settle at aWardrop (day-to-day)equilibrium or a Boston (within-day)traffic equilibrium,respec-tively.We prove that the path travel time function in our model is not monotone,and showthat this property leads to multiple Wardrop equilibria,of which only one is both stable andefficient .The paper shows that periodic traffic oscillations do not arise from Wardrop equi-libria.Trivial oscillations exist at Boston equilibria,which are caused by drivers’overreac-tion to traffic conditions.However,periodic oscillations are likely to emerge when (1)transitions between stable and unstable equilibria take place,and more importantly,(2)drivers make decisions based on out-of-date information of traffic conditions.The latterfinding is useful in guiding control practice at freeway bottlenecks and work zones to pre-vent traffic oscillations.Ó2009Elsevier Ltd.All rights reserved.1.IntroductionTraffic oscillations often arise in congested traffic flow,such as vehicular queues induced by freeway bottlenecks (for recent empirical evidences,see e.g.,Smilowitz et al.,1999;Mauch and Cassidy,2002;Ahn and Cassidy,2007).In the past,this frustrating stop-and-go motion is often explained using car-following behavior (e.g.Chandler et al.,1958;Herman et al.,1959;Treiterer and Myers,1974),lane-changing maneuvers (e.g.Gazis et al.,1962;Munjal and Pipes,1971;Daganzo,2002a,b ),and the instability predicted by higher-order traffic flow models (e.g.Kerner and Konhauser,1994;Jin and Zhang,2003).Traffic oscillations may also be triggered by macroscopic mechanisms such as queue interactions (Jin,2003).Recently,Nie and Zhang (2008)and Jin (2009)characterized this type of oscillatory traffic pattern using a two-route network with a diverge and a merge (hereafter referred to as the D–M model)(see Fig.1).These studies employ the traffic flow model of Lighthill and Whitham (1955)and Richards (1956)whereas the merge and diverge traffic follow the models of Jin and Zhang (2001)and Daganzo (1995),respectively.Periodic oscillations may occur in this model when queues formed at the merge spill back to the diverge,thereby reducing the discharging capacity of the diverging branches due to the first-in–first-out (FIFO)discipline imposed at the diverge.Noticeably,key features of such oscillatory traffic patterns ap-pear to agree with empirical evidence such as reported in Mauch and Cassidy (2002)and Ahn and Cassidy (2007).This coincidence raises an interesting question of whether such a model can be used to explain,if not predict,traffic oscil-lations often observed at freeway bottlenecks.Valid though that question may seem,it should be noted that the original 0191-2615/$-see front matter Ó2009Elsevier Ltd.All rights reserved.doi:10.1016/j.trb.2009.06.002*Tel.:+18474670502.E-mail address:y-nie@Transportation Research Part B 44(2010)62–72Contents lists available at ScienceDirectTransportation Research Part Bj o ur na l h om e pa ge :w w w.e ls e v ie r.c om /lo c at e /t rbYu(Marco)Nie/Transportation Research Part B44(2010)62–7263D–M model ignores drivers’route choice behavior,namely,oscillations occur only whenflow distribution at the diverge isfixed within a certain range.Obviously,unless an effective control device is implemented in its favor,drivers may not follow such afixed route choice.Would traffic oscillations emerge when drivers’behavior is reasonably taken into consideration?The present paper is in-tended to address this question.Well-embraced behavioral assumptions state that drivers tend to make travel choices (departure time,routes,etc.)to maximize their utility.For the purpose of this study,it suffices to focus on route choice and assume that travel time is the only factor at work in that choice.It is well-known that the traffic assignment problem (the problem of assigning traffic to shortest routes)with such behavioral assumptions can be formulated as a Nash–Cournot non-cooperative game,whose solutions are characterized by a set of traffic equilibrium states.Since the temporal evolution of trafficflow has to be considered in order to study oscillations,our equilibrium analysis falls into a class of dynamic traffic assignment(DTA)models.The reader is referred to Peeta and Ziliaskopoulos(2001)for a comprehensive review of the DTA literature.Our goal is to study the properties of dynamic equilibrium solutions to the diverge–merge network model illustrated in Fig.1a.Such a simple model allows us to derive analytical solutions that promise useful insights.More importantly,the D–M model reasonably represents a bottleneck situation where lane drop may cause vehicular queues and subsequently traffic oscillations,as shown in Fig.1b.Assumptions necessary to make such a connection are:Drivers are informed of the bottleneck by a traffic sign upstream of the lane drop.In response to this event,drivers will make a lane-changing decision at a point near the sign.That point corresponds to the diverging junction in Fig.1a.The ratio of drivers who select the shoulder lanes(i.e.,link1in Fig.1a)is denoted as r.Once passing the imaginary diverge,drivers will not change lane until they arrive at the actual lane-drop location,which constitutes the merging junction in Fig.1a.The settings in Fig.1will be frequently used hereafter.Particularly,we emphasize that links1and2in Fig.1a refer to the shoulder lanes and through lanes in Fig.1b,respectively.Two different behavioral assumptions,which lead to different equilibrium states,are considered.In thefirst,drivers want to minimize their experienced travel times.By learning from and adjusting according to daily travel experience, drivers will settle at the so-called day-to-day equilibrium,which is a dynamic extension of the Wardrop equilibrium (Wardrop,1952)and is widely used for long-term travel forecasting(e.g.Smith,1993;Friesz et al.,1993;Ran et al., 1996).However,lane-changing maneuvers as those triggered by a lane drop in Fig.1b may be too minor to be pre-dicted from such a day-to-day equilibrium.It is more likely that drivers would make those lane-changing decisions en-route according to local traffic conditions.This assumption drives the system to a Boston traffic equilibrium(Friesz et al.,1993),in which drivers minimize their instantaneous travel times.The focus of the paper is,therefore,to obtain both Wardrop and Boston traffic equilibria of the D–M model and reveal their analytical properties,particularly those pertinent to oscillations.Numerical experiments will be conducted when it is difficult to get simple analytical solutions.This paper is organized as follows.Section2briefly reviews the oscillatory traffic patterns yielded from the D–M model when the route choice isfixed.Sections3and4discuss the Wardrop and Boston traffic equilibria,respectively.Section5 concludes the paper.2.Oscillatory traffic pattern with fixed route choiceNie and Zhang (2008)1provides analytical solutions of dynamic traffic flows to the D–M model depicted in Fig.1a,when diversion ratio r is treated as a fixed exogenous variable.It is shown that the evolution of the system is uniquely determined by two factors:the ratio of capacities of links 1and 2(k ),and the diversion ratio r 2½0;1 .A synopsis of their results follows.It is useful to first recall that the LWR model (Lighthill and Whitham,1955;Richards,1956),the diverge model of Daganzo (1995)and the merge model of Jin and Zhang (2001)2are used to describe traffic flow.An identical triangle fundamental dia-gram is employed to describe the flow–density relationship on all links.Assuming that the total demand is ð1þk Þc (i.e.,the capacity of link 3in Fig.1a),the outflow and inflow at the diverge arev o 3¼min ð1þk Þc ;kc ;c ;v i 1¼r v o 3;v i 2¼ð1Àr Þv o 3ð1Þwhere v o 3denotes the outflow of link 3,and v i 1and v i 2are the inflows of links 1and 2,respectively.The outflows of links 1and 2are determined byv i 4¼min f c ;D 1þD 2g ;v oj ¼v i 4D j 12;j ¼1;2ð2Þwhere v i 4is the inflow of link 4and D j is the demand for link j (i.e.,the maximum flow rate allowed to leave a link)Traffic congestion in the system may be initiated either at the merge due to the insufficient downstream supply,or at the diverge due to an imbalanced diversion ratio.The interplay of the two effects will lead the system to different terminal states depending on k and r ,as summarized in Fig.2(the reader is referred to Section 3.4(Nie and Zhang,2008)for more detailed analysis which is deemed unnecessary to repeat herein).Note that,due to the insufficient capacity,link 3will become con-gested as soon as the queues reach the diverge,regardless of traffic condition on links 1and 2.Specifically,when r P k (Re-gion I),the traffic flow pattern is always stable;when k 1þk <r <min ð0:5;k Þ(Region III),three different periodic oscillation patterns occur at different k Àr combinations;otherwise (Regions II and IV),the system will oscillate at the beginningbut eventually converge to a stable solution conforming to the initial diversion ratio,i.e.,v i 1¼v o 1¼cr while v i 2¼v o 2¼c ð1Àr Þ.Nie and Zhang (2008)noted that some periodic oscillatory traffic patterns from the above analysis (Region III in Fig.2)well coincide with the empirical observations of traffic oscillations (e.g.Mauch and Cassidy,2002).However,these oscilla-tions are not realistic in the sense that they do not reflect drivers’route choice behavior.Perhaps the most obvious flaw is that all periodic oscillations occur only when the diversion ratio is such set that link 2(through lanes)is consistently uncon-gested.Clearly those oscillations may not happen in reality since no driver would rather wait to enter a congested link 1when link 2is under free flow condition.The above observation necessitates the analysis that takes drivers’behavior into account.3.Analytical Wardrop equilibrium solutionsIf drivers are free to choose among links 1and 2(cf.Fig.1a),they would choose such that neither link is faster than the other,which will drive the system to the well-known Wardrop equilibrium (Wardrop,1952).Following the convention,we shall call such a pattern a user-equilibrium (UE).Let ½0;T be an analysis period during which departure may take place,and c i ðt ;f Þdenote the travel time on path i departing at a time t 2½0;T ,which is a function of the path flow vector f ¼½f 1;f 2 T .If1Jin (2009)gives similar but more detailed asymptotical results.2While this simplified merge scheme violates the invariance principle (Lebacque and Khoshyaran,2005),the violation does not alter the underlying nature of the oscillations.The reader is referred to Nie and Zhang (2008)for a more detailed discussion.64Yu (Marco)Nie /Transportation Research Part B 44(2010)62–72qðtÞ¼P2i¼1f iðtÞis the departure rate of drivers at time t,a UEflow pattern fÃcan be characterized by the following equilib-rium conditionsfÃi ðtÞ>0)cÃiðtÞ¼lðtÞ;f iðtÞ¼0)cÃiðtÞP lðtÞ;8i;t2½0;T ð3Þwhere lðtÞ¼minðcÃiðtÞ;i¼1;2Þ;8t.fÃmay be obtained by solving an infinite-dimension variational inequality(Friesz et al.,1993)X2 i¼1Z Tc iðw;fÃÞðf iðwÞÀfÃiðwÞÞdw P0;8f such that f iðtÞP0;i¼1;2;X2if iðtÞ¼qðtÞ;8tð4ÞFor a problem as simple as the D–M model,however,the UEflows may be obtained analytically.Proposition1.When qðtÞ¼ð1þkÞc;8t3in the D–M model,aflow pattern fÃsatisfies the UE condition(3)if it meets any of thefollowing conditions:(I)fÃ1ðtÞ¼0;8t;(II)fÃ1ðtÞ¼kqðtÞ;8t,(III)fÃ1ðtÞ¼kkþ1qðtÞ;8t,and(IV)fÃ1ðtÞ>kqðtÞ;8t,whereP2if iðtÞ¼qðtÞ;8t.Proof.We only need to show that,when any of the conditions is met,the travel time on links1and2are the same at any time t,which leads to c1ðtÞ¼c2ðtÞ8t.(I)In this case,r¼0,so v o3¼v i2¼c and v i1¼0.Since v i1;v i26c,no congestion will develop on either link.Thus,traveltimes on links1and2are always same(i.e.,freeflow travel time).(II)When fÃ1ðtÞ¼kqðtÞ,the diverge model(1)implies that v o3¼c.Link1operates at its capacity kc and link2is unde-rused.Similarly,since v i1þv i2¼c,both links are always at freeflow conditions and thus have the same travel time all the time.(III)When fÃ1ðtÞ¼k qðtÞ,(1)implies that v o3¼ð1þkÞc;v i1¼kc;v i2¼c.After the traffic arrives at the merge,queues will develop on these links because v i1þv i2>c.Theflow distribution(2)yields v o1¼kc=ð1þkÞ;v o2¼c=1þk.Not-ing that the incomingflow rate is at capacity on both links,the queues will grow at the same speed(equals the wave speed),and have the same density.Consequently,the travel time on either link will increase,but at the same pace.When the queues reach the diverge,the total outflow of link3is reduced fromð1þkÞc to c,with v i1reduced to kc=ð1þkÞ.In this case,both links1and2are congested,but the travel times on them remain the same8t.(IV)When fÃ1ðtÞ>kqðtÞ,the reader may verify that the total inflow into links is smaller than c,using the diverge formula(1).Therefore,links1and2never become congested and travel times on them are always freeflow travel time.ÃWe note that Condition IV actually contains infinite number of solutions,because f1ðtÞcan take an arbitrary value be-tween kqðtÞand qðtÞfor any t without violating the UE condition.Consequently,the D–M model demonstrates that multiple user equilibria corresponding to different linkflow patterns exist in the DTA problems with physical queues.Multiple equi-libria are often a result of the lack of monotonicity of the cost function cðt;fÞin the VI problem.In the static traffic assign-ment problem(Beckmann et al.,1956;Sheffi,1985),for example,multiple equilibria may occur if non-separable cost functions(such as those involving intersection delays)renders the failure of monotonicity(Smith,1982).It is known that monotonicity of path travel times may not hold in the DTA problems even when physical queues are not considered.For example,Mounce(2003)showed,based on a point-queue traffic model,that path travel times may not be monotone when a path contains more than one bottleneck.As demonstrated in the following,non-monotonicity in our model arises not due to the interaction between bottlenecks,but rather from the link interactions imposed through the merge and diverge models. Daganzo(1998)showed that DTA solutions may be chaotic when queue spillover is permitted,because a small perturbation of network inputs can override an over-saturated equilibrium with an under-saturated one,or vice versa.The multiple-equi-libria phenomenon revealed from the D–M model is along the line of Daganzo’s discovery(in the sense that they both have to do with interactions between links),but has a different cause.Let usfirst present the following lemma which characterizes the relationship between the path travel time and diversion ratio in the D–M model.Lemma1.When any of the Conditions(I)–(III)in Proposition1is satisfied,c1ðtÞand c2ðtÞare identical and attain the same minimum value cðtÞ;8t.If0<f1ðtÞ<kc=ð1þkÞ;8t;c1ðtÞ< cðtÞ<c2ðtÞalmost everywhere in t2½0;T ;if kc=ð1þkÞ< f1ðtÞ<kc;8t;c1ðtÞ> cðtÞ>c2ðtÞalmost everywhere in t2½0;T .Proof.When any of the Conditions(I)–(III)is met,because the bottleneck(the merge point)always operates at its capacity c, the total travel times experienced by travelers departing at any time must be the lowest(note that no otherflow pattern could reduce travel delay),and thus identical.Further,because these cases imply equilibrium,travelers should experience3This condition is introduced to simplify the analysis.One may argue that qðtÞequals the capacity of link3only when the queue does not reach the origin. We note that this need not undermine the analysis if traffic is allowed to queue at the origin.Yu(Marco)Nie/Transportation Research Part B44(2010)62–7265the same travel time whether they take path1or2.Therefore,c1ðtÞand c2ðtÞcorrespond to the minimum possible travel time cðtÞ.When0<f1ðtÞ<kc=ð1þkÞ;8t.Link2is always congested as soon as traffic reaches the merge,because v o2<c¼v i2. Although link1may also become congested4before the system converges to a stable solution,the density is always lower and accordingly the queue grows more slowly.Thus,the travel time on link1is lower than that on link2except at instants with zero Lebesgue measure when they are identical such as when t¼0and t¼T.This implies c1ðtÞ<c2ðtÞalmost everywhere. Moreover,note that the bottleneck still operates at capacity in this case.Therefore,the total travel time experienced by travelers departing at any time remains to be the lowest possible.That is,c1ðtÞf1ðtÞþc2ðtÞf2ðtÞ¼q t cðtÞ)c1ðtÞ< cðtÞand c2ðtÞ> cðtÞ.When kc=ð1þkÞ<f1ðtÞ<kc;8t,link1always operates at its capacity according to the diverge model(1),and is subject to queuing delay as soon as traffic arrives at the merge.Depending on whether k<0:5or not,the system may or may notenter into a periodic oscillation(Fig.2).Nevertheless,if f1ðtÞ>ðkþk2Þc1þkþk2;8t,link2is never congested(again,cf.Fig.4in Nie andZhang,2008);otherwise,link2will become congested but its queue has a lower density and grows more slowly.Thus, c1ðtÞ>c2ðtÞalmost everywhere in t2½0;T .This in turn implies c1ðtÞ> cðtÞ>c2ðtÞ.ÃFig.3visualizes the result of Lemma1,illustrating how the travel times on paths1and2associated with a departure time 0<t<T may change with the diversion ratio r.We are now ready to show that c iðt;fÞis not a monotone mapping.This is formally stated as follows.Proposition2.The cost function c iðt;fÞin the VI problem(4)is not monotone if trafficflow is described by the LWR model,the diverge model(1)and merge model(2).Proof.Monotonicity of cðt;fÞimplies that for any feasible f and g,the following inequality always holdsX2 i¼1Z T½c iðw;fÞÀc iðw;gÞ ½f iðwÞÀg iðwÞ dw P0ð5ÞWe only need tofind one case in which the above does not hold.Let f1ðtÞ¼kc;8t,and g1ðtÞ¼r0c;8t;k=ðkþ1Þ<r0<k. Thus f1ðtÞ>g1ðtÞand f2ðtÞ<g2ðtÞ.According to Lemma1,however,c1ðt;fÞ<c1ðt;gÞand c2ðt;fÞ>c2ðt;fÞalmost everywhere. Thusðc1ðt;fÞÀc1ðt;gÞÞðf1ðtÞÀg1ðtÞÞ<0;ðc2ðt;fÞÀc2ðt;gÞÞðf2ðtÞÀg2ðtÞÞ<0almost everywhere.So the inequality(5)does not hold for this f and g.ÃThe existence of multiple equilibria is a bad news for the purpose of travel forecasting since one has to determine which equilibrium is more likely to prevail in reality.Nevertheless,different equilibria may be distinguished by some properties,of which the most important are stability and efficiency.Definition1(Stability).An equilibrium pathflow pattern fÃis stable if the following two conditions are met when an arbitrarily small number of traveler switch from one path to another.1.Travelers who stay on their original paths do not have a better alternative after the perturbation.2.The travel time experienced by any traveler who changes their paths is strictly longer.Intuitively,the above definition ensures that travelers who have changed their paths have the incentive to return imme-diately to the previous equilibrium.A more rigorous definition of stability requires that all eigenvalues of the Jacobian matrix cðt;fÞhave positive real parts(e.g.Pappalardo and Passacantando,2002).Since the Jacobian of cðt;fÞusually does not have an analytical form,it is difficult to establish the stability of the system using its Jacobian.4Link1becomes congested in this case only when f1ðtÞ>kc=ð1þ2kÞ.The reader is referred to Fig.4in Nie and Zhang(2008)and related discussions.66Yu(Marco)Nie/Transportation Research Part B44(2010)62–72Yu(Marco)Nie/Transportation Research Part B44(2010)62–7267 Definition2(Efficiency).An equilibrium pathflow pattern fÃis efficient if there is no other equilibrium in which at least one driver has a lower travel time and no driver has a higher travel time.Now we are able to characterize the four equilibria given in Proposition1using these properties.Proposition3.Of the four equilibria described in Proposition1,Equilibria I and II are efficient but unstable;Equilibrium IV is neither stable nor efficient;Equilibrium III is both efficient and stable.Proof.That equilibria I,II and III are all efficient directly follows from Lemma1.Equilibrium IV is inefficient because the bottleneck always operates below the capacity c.At Equilibrium I,if a tiny number of drivers switches from path2to path 1,these drivers will experience a lower travel time,as shown in Fig.3.Similarly,a driver switch from path1to path2at Equilibrium II may reduce its travel time.This violates the second condition for stability.The stability of Equilibrium III is easy to verify using Lemma1,noting that any small change of pathflow at that equilibrium will change path travel times in the same direction.Finally,Equilibrium IV is unstable because the second condition in Definition1is violated whenflow slightly shifts from path1to2,and hence increases the bottleneck capacity and reduces travel time.ÃWhile multiple equilibria exist for the D–M model,only Equilibrium III is likely to be observed in the long run.The unsta-ble Equilibria I,II and IV are short-lived,if not unobservable,because any small perturbation in the proper direction may initiate an irreversible transition from any of the unstable equilibria to Equilibrium III.Suppose,for example,that all drivers initially merge to the through lane(link2)and leave the shoulder lane(link1)empty.This corresponds to Equilibrium I. However,some aggressive drivers may wish to cut in line by taking advantage of the shoulder lane.Because these‘‘pioneers”do improve their travel times,they will be followed by more drivers who are now at a disadvantage by staying the course. Consequently,this trend will continue until link1becomes as congested as link2.From a behavioral point of view,Equilib-rium III is likely to become a dominant pattern through drivers’day-to-day learning process.In this process,drivers may gradually recognize that switching to the main line only at the real merge point is the best strategy to choose in dense traffic. However,oscillatory traffic will inevitably appear when transitions between equilibria take place.Particularly,a non-equi-librium state rest between Equilibria II and III is associated with periodic oscillations.The existence of multiple equilibria in the D–M model suggests that such a situation is bound to arise in more general DTA models which can only be solved numerically.Since forecasting traffic is a major application of DTA models,it is important to ensure that numerical solutions yielded from such models are both stable and efficient.Unfortunately,unlike its static coun-terparts,most general DTA models can only be solved with heuristic algorithms that do not guarantee convergence,owing to mostly the lack of monotonicity.In practice,these algorithms are typically unable to tightly converge to an equilibrium solu-tion.While an approximated equilibrium may be acceptable for many practical purposes,it is a considerable challenge to verify its stability and efficiency,which is critical to its realism.4.En-route decision and oscillatory traffic patternsThe previous section suggests that traffic oscillations described in Nie and Zhang(2008)may not manifest in the D–M model should drivers observe any of the Wardrop equilibria.In fact,any oscillatory traffic pattern in the D–M model neces-sarily implies different travel times on links1and2which cannot be satisfied at a Wardrop equilibrium.As shown in Fig.3, in the oscillation regionðk=ð1þkÞ<r<k and0<r<1=ð1þkÞÞ,travel times on the two paths never equilibrate with each other.However,we caution that Wardrop equilibria may not accurately predict driver’s behavior when a bottleneck necessitates lane-changing maneuvers.For one thing,drivers may have to travel through a number of congested freeway bottlenecks in their journey.Therefore,where and how to perform lane-changing at each bottleneck is a rather minor choice compared to other choices such as routes and departure times.Moreover,due to limited availability of data,the inherent randomness in both supplies and demands,and the individual’s perception error and capacity of short-term memory,it is unlikely that day-to-day equilibrium could predict those micro choices with acceptable accuracy.Consequently,it is more reasonable to as-sume that drivers would make such lane-changing decisions‘‘en-route”based on local conditions.Such an assumption leads to the so-called Boston traffic equilibrium(BTE),in which‘‘network users do not settle into a day-to-day equilibriumflow pattern,but rather try to optimize their route and departure time choices based on current information”(Friesz et al., 1993).At a BTE,drivers constantly seek to minimize their journey times through en-route rerouting.They are assumed to know the accurate and up-to-date network-wide traffic information in order to make those rerouting decisions.Although drivers seem to make the best choice at each instant of time in BTE,they are myopic in that they do not properly account for other drivers’decisions.Consequently,drivers may experience a journey time totally different from what they anticipate at each decision point.It seems that oscillations may happen to a BTEflow pattern because BTE could produce differences in experienced travel times.In the following,we obtain the BTE solution for the D–M model based on the concept of reactive traffic assignment. Suppose that link travel times are updated according to real-time traffic conditions.Then,for each diverge point(where a rerouting decision is required),shortest paths leading to the destination are computed.Flows approaching a diverge are as-signed to the up-to-date shortest path.Note that the above procedure is just a sequence of all-or-nothing assignments decomposed in time and space,which does not guarantee to yield a BTE.Nevertheless,like the incremental assignment。
专利名称：Articulated vehicle发明人：Brown, David Patrick,Brown, John Bowes申请号：EP83307741.5申请日：19831220公开号：EP0112712A2公开日：19840704专利内容由知识产权出版社提供专利附图：摘要：A main chassis (2) is pivotally supported by front and rear sub-chassis (1, 6)having at least one drive axle (7, 17, 23). One of the sub-chassis mounts the engine (14).Constant velocity power is communicated between the sub-chassis (1, 6) by a mechanism including an intermediate shaft (20) parallel to the plane containing the pivot axes (27, 30)of the sub-chassis (1,6). The ends of the shaft (20) have universal joints equidistant from the pivot axes (27, 30). Means are provided for controlling pivoting of the sub-chassis (1, 6) so that, during turning, the angles between the centre-lines of the sub-chassis (1, 6) and the centre-line of the main chassis (2) are equal and opposite.申请人：Multidrive Limited地址：Handley Close Preston Farm Industrial Estate Stockton on Tees Cleveland TS18 3SD GB国籍：GB代理机构：Godwin, Edgar James更多信息请下载全文后查看。
专利名称：ACTUATOR FOR A VEHICLE CLUTCH 发明人：HEDMAN, Anders,ZETTERSTRAND, Lars 申请号：EP05704712.8申请日：20050114公开号：EP1841979B1公开日：20110406专利内容由知识产权出版社提供摘要：The invention relates to an actuator (413) for a clutch arranged between an engine and a transmission and comprising an annular cylinder part (458) located concentrically with an input shaft to the transmission, and having a cylinder chamber (427) and an annular piston (426) connected to a throwout bearing (416). A part of the hydraulic piston arrangement (454, 554) or a position-sensing mechanism (460) remote from said hydraulic chamber (423) adjoins a fluid chamber (469), which is connected to or forms a part of a connecting duct (428, 469) or cylinder chamber (427), so that in the fluid chamber (469) the control pressure from a valve arrangement (422) acts on the hydraulic piston arrangement (454, 554) with a force (470) that is opposed to the force (442) with which the fluid pressure in said hydraulic chamber (423) acts on the hydraulic piston arrangement (454, 554). Assembly and maintenance are facilitated. The friction surfaces of the clutch are well sealed off from the hydraulic fluid. Embodiments with overload protection and measurement of the thickness of the clutch plates are demonstrated.申请人：VOLVO LASTVAGNAR AB地址：SE国籍：SE代理机构：Fröhling, Werner Otto 更多信息请下载全文后查看。
Abstract —In this paper, a novel crawler mechanism for sideways motion is presented. The crawler mechanism is of circular cross-section and has active rolling axes at the center of the circles. Conventional crawler mechanisms can support massive loads, but cannot produce sideways motion. Additionally, previous crawler edges sink undesirably on soft ground, particularly when the vehicle body is subject to a sideways tilt. The proposed design solves these drawbacks by adopting a circular cross-section crawler. A prototype has been developed to illustrate the concept. Motion experiments confirm the novel properties of this mechanism: sideways motion and robustness against edge-sink. Motion experiments, with a test vehicle are also presented.Keywords: Tracked Vehicle, Sideways Motion, Circular Cross-Section, Crawler, Pipe InspectionI. INTRODUCTIONonventional crawlers cannot move sideways. Therefore they usually (i) lack enough maneuverability to move in narrow spaces such as in Fig. 1(a). For example, It is not so easy to set the position of the crawler vehicle to trajectory number 5 in Fig. 1(a). In addition, when a conventional crawler tilts sideways on soft ground, (ii) the edge of the crawler unit might sink undesirably as shown in Fig. 1(b). In this paper we present a mechanism that solves these two issues. A crawler mechanism that realizes sideling motion is presented and the application of a pipe inspection robot is examined.A . The weak point of crawlers which realize the sidewaysmobilityIn order to realize holonomic omni-directional motion, there exist many commercial wheels which are based on small passive rotational wheels[1]-[8]. Some of them are similar to a crawler-like mechanism. A particularly accomplished example of this is the VUTON[9] developed by Hirose, or the vehicle developed by M. West et al.[10] and the mechanismUniversity of Electro-Communications 1,AIST 2, Tohoku University 3, Massachusetts Institute of Technology 4 , tadakuma@ , 1-5-1 Chofugaoka, Chofu-shi, Tokyo, 182-8585 JAPANdeveloped by Chen et al[11]. However, these crawler-like mechanisms have many numbers of small passive rotational rollers, and are not generally capable of overcoming steps or ground discontinuities typical in environments such as houses, offices or hospitals (e.g. the gap at an elevator opening). This limitation stems from the fact that the diameter of the passivewheel is much smaller than the diameter of the whole wheel.[ Front View ][ ObliqueView ](a)Difficulty to Adjust the Position,Going Into Narrow Space (Top View )(b)Sinking from the Edgein Soft Ground Narrow SpaceBody of theV ehicleCrawler Unit Soft GroundObstacleSoft GroundObstacleObstacleCrawler Unit Bodyof theV ehicle1234245A BC CCFig. 1: Weak points of Conventional Crawler MechanismII. B ASIC C ONCEPT OF THE C RAWLER WITH C IRCULAR C ROSS -S ECTIONThe basic concept of the proposed crawler with a circular cross-section is shown in Fig.2. The crawler module has an active rotational axis; which allows it to realize the required sideways motion.Additionally, this configuration has another distinctive feature, shown in Fig. 3.Fig. 2: Basic Schema of Omni-Crawler with Circular SectionCrawler Vehicle with Circular Cross-Section Unitto Realize Sideways MotionKenjiro Tadakuma 1, Riichiro Tadakuma 2, Keiji Nagatani 3, Kazuya Yoshida 3, Steve Peters 4, Martin Udengaard 4, Karl Iagnemma 4C2008 IEEE/RSJ International Conference on Intelligent Robots and Systems Acropolis Convention Center Nice, France, Sept, 22-26, 2008Basic Motion Direction of the Vehicle(a) Inner Pipe (b) Outer PipeFig. 3 Fitting by design to various pipe sizes (Front View) When a conventional crawler moves inside a pipe (or on its outer surface), the edge of the crawler belt contacts the surface with a small area: an edge line. Or conventional one should change their inclining angle in roll axis adopting to the surface of the pipe[18]. On the other hand, when the proposed crawler with a circular cross-section moves on the inside (or outside) of a pipe, the contact area is significantly increased due to the shape and elasticity of the circular crawler belt. In addition, the circular cross-section reduces the problem of the crawler unit sinking into the pipe surface.Robots mounted with the proposed mechanism move in a direction perpendicular to the passive wheel axis, as shown in Fig. 2. The maximum step which the mobile robots can overcome is significantly small relative to the size of the whole wheel because of the small diameter of the passive wheels.III.M ECHANISM OF THE C RAWLER WITH C IRCULARS ECTIONIn this section, the basic configuration of the proposed cylindrical crawler mechanism is described. First, the mechanism of the Omni-Ball[12][13] which authors developed before (which inspired the cylindrical crawler design) is explained.A. Basic Configuration of “Omni-Ball”A spherical mechanism such as the spherical wheel [14] can be used to achieve omni-directional motion. However, this wheel has many mechanical parts, including rollers, mechanical guides, etc. One of the key developments presented in this paper is a novel mechanism for AGV which can realize omni-directional motion symmetrically.The basic 3D-concept model is shown in Fig.4. In Fig. 4, two hemispheres rotate passively, and the active rotational axis lies in the center of the Omni-Ball. In order to rotate, both the hemispheres are passive. Note that each passive rotational axis is independent, so that each rotation of a hemispheric wheel is also independent.When the active axis rotates, the Omni-Ball produces a propelling force in a direction perpendicular to the active rotational axis, as shown in Fig.4. At the same time, the wheel does not produce a propelling force in the horizontal directionin Fig.4, so that this mechanism can similarly move in an arbitrary direction by a combination of three propelling forces.Passive Rotational AxisGeared Motor(Rotary Actuator)Passive MovementFig.4: Basic Structure of the “Omni-Ball”B.Basic Configuration of “Omni-Crawler”The comparison between the Omni-Ball and Omni-Crawler designs is shown in Fig.5.The Omni-Crawler is a an extension of the Omni-wheel from a sphere to a cylinder with spherical ends. From a front view, as in Fig 5[a-1] and 5[b-1], the two mechanisms are identical. The difference is evident, however, from the bottom view, as in Fig 5[a-2] and 5[b-2], where there is a cylindrical section inserted between the spherical ends of the Omni-crawler. While the surface of the Omni-ball consisted of two solid hemispheres, the surface of the Omni-crawler is composed of a belt that runs over both the cylinder and spherical ends. The belt is described in the next section.As stated in the previous section, the hemispherical wheel of the Omni-Ball is rotated passively in Fig.5. The Omni-Crawler, however, can produce propelling force not only laterally, but also longitudinally. If a rotary actuator is mounted on the hemispherical wheel of the Omni-Ball, its longitudinal velocity must be adjusted according to the angle formed by the axis of the hemispherical wheel to the ground, because the relative radius of the wheel will change based on the inclining angle of the wheel mechanism. Fig.5(b) shows a schema of the Omni-Crawler: the crawler mechanism to realize sideways motion. Note that in contrast with the omni-ball, the velocity of the crawler in forward direction isindependent of the inclining angle of the driving unit, asshown in Fig. 5[b-1] and [b-2]. θωr o P Passive Active [ b-2 ] Omni-Crawler [ a-2 ] Omni-Ball(Bottow View)(Bottow View)Active RotationActive RotationT r a c t i o n D i r e c t i o no 1P 1o 2P 2Active Rotation Active Linear MotionTraction DirectionTraction Direction Contact Line ContactPoint P a s s i v e M o v e m e n tLinearDriving Areaωr ωr P [ a-1 ] Omni-Ball (Front View)o θr R P [ b-1 ] Omni-Crawler(Front View)o θr RFig.5: Basic Principle of the “Omni-Crawler”The velocity of the vehicle with Omni-Ball is described by,V ob = R {cos(π/2)-θ}*ω =( R * sin θ) *ω (1)and it is dependent on the incline angle of the Omni-Ball. On the other hand, the velocity of Omni-Crawler “Voc” is as follows (linear contact with ground hypothesis)V oc = R* ω (2)This is the basic principle of the Omni-Crawler.Summarizing: (i) There is a singularity line on the Omni-crawler that doesnot allow any longitudinal traction force to be generated. (ii) The system can move longitudinally by standard motion of the track.When the side belts are actively moved in the same direction, the generated motion by the crawler is changed before and after crossing the singularity line, therefore, therotational motion direction of the sprocket should be changed, if the crawler unit should move in the same longitudinal direction. During half rotational motion of the crawler unit, it can realize following two, although it is better that thelongitudinal and the sideways motion are conducted individually basically. (iii)The system can move omni-directionally (bycombining the rolling motion in roll axis).(iv)It can adjust its velocity in any arbitrary direction.In order to realize the mechanism based on this idea, the actual prototype model has been developed as explained in next section.C. Actual Prototype of “Omni-Crawler” (c-1) Inner Mechanism of the CrawlerThe inner mechanism of the crawler is shown in Fig. 6. The geared motor rotates the inner shaft and bevel gear set changes the rotation direction to the vertical axis of the shaft of the motor and sprocket is rotated at the end. In the end of the front and rear of the inner mechanism unit are the shafts for the sideling motion of the crawler. By actively rotating these shafts in the rolling axis, lateral motion can be realized. The tensioner for the belt is mounted at the rear. It is much better to put a slip ring for wiring at the rear shaft.AAA Batteryfor size comparisonGuide of the BeltSprocketGeared MotorCouplingRolling Axis for Sideling MotionTensionerBevel Gear SetScrewTensionerFig.6: Mechanism inside the Crawler Unit(c-2) Crawler Belt with Circular LugIn order to realize smooth sideling motion, the shape of the section of the crawler has to be a circle. The lug of the crawler is shown in Fig. 7. At the surface of the lug module, rubber is set to the lug supporter. The material of the lug supporter is SUS304 (metal). The overview of the belt of the crawler is shown in Fig.8.Lug Supporter (Bent SUS304)Surface Rubber of the Circular LugScrew HoleFig.7: Lug of the Crawler BeltLugsFig.8: Belt of the Crawler (top one is reverse side of the crawler)(a) Oblique View the Crawler UnitLug of the Crawler(b) Front View(c) Access to the tensioner AAA Battery for size comparisonFig.9: Actual Prototype Model of “Omni-Crawler” UnitNote that the width of the crawler unit “Wc”and height “Hc”and diameter of the crawler unit “Dc”are satisfied the following equation and as shown in Fig.9 and table 1, the actual model satisfy these relations.Wc = Hc = Dc (3)IV. Vehicle with Crawler Units4.1 Configuration of the VehicleThe realizable configurations of the vehicle with sideways motion are shown in Fig. 10. There are basically three configurations as follows.(a) Rectangle Configuration with 4 Crawler UnitsThis layout of the crawler vehicle was adopted into the configuration of the VUTON[9]. The rotational motion of the crawler unit is passive while the longitudinal motion of the crawler is active.(b)Radial Configuration with 4 Crawler UnitsThe rotational motion of the crawler unit is active while the longitudinal motion of the crawler is passive. (c)Parallel Configuration with 2 Crawler UnitsTwo crawler units are aligned in parallel. Both of the rotational and the longitudinal motions are active.Finally, the selected configuration as a prototype model is (c), 2-unit parallel type in order to realize better pipe inspection task easily in low cost and light weight this time. The whole configuration of the vehicle is shown in Fig. 11 and specification of the vehicle is shown in table2.There is the motor to rotate both axes of rolling shaft from the right and left crawler units for realizing sideling motion. The battery “Pocket Moba MV”[17] and electronic circuit boards are mounted on top of the body.If the vehicle must be thrown to reach its search target, its body will be made smaller than the space between the crawler units.(a) Rectangle Configuration(b) Radial Configuration (c) 2-Units Parallel ConfigurationCrawler Unit Active PassiveFig.10: Configurations of the Vehicles (Top View)[ Oblique View][ Front View ][ Side View ]Fig.11: Whole View of the Crawler Vehicle4.2 Control System of the VehicleThe control system of the vehicle is shown in Fig. 12. The two motors in each crawler units and one motor for sideways motion are controlled with radio controller unit by manually. And Signals are transmitted to each motor through the electronic circuit board. The vehicle is controlled by manually so far, but automatic control is going to be developed and rotational direction of the motor in the track will be changed in near future by using semi-automatic control.MotorMotor forCrawler UnitCrawler UnitFig.12: Control System of the Vehicle V. E XPERIMENTS OF THE C RAWLER V EHICLEIn this section, we describe a set of experiments conducted to confirm the performance of a prototype of this crawler vehicle with the omni-crawler drive mechanism.A. Omnidirectional MotionAs one of the basic mobility criteria of this robot, the ability to produce omnidirectional motion should be confirmed. One example of such a motion is shown in Fig.13. It was observed that this prototype model has the ability to move in an arbitrary direction smoothly. Please see the movie attached to this paper.Fig.13: Omnidirectional Motion on FloorIn order to measure the actual speed of the crawler in the forward-backward direction, the experimental device including one crawler unit have been developed and tested as shown in Fig. 14. The device has the box-like flamed shape and passive casters are mounted on the bottom of the device. During the measurement, the one crawler unit moved to forward and backward direction in just only the fixed angle in its each roll axis.CasterFlame Geared Motor1 - Omni-Track UnitV e l o c i t y (m m /s )Inclining Angle in Rolling Axis of the Crawler Unit(degree)020406080100120140160020406080Fig.15: Comparison of the Speeds in the forwarddirectionFigure 15 shows that the longitudinal speed of the crawler is almost constant. The speed does not depend significantly on the inclination of the rolling axis of the crawler unit. See also Fig.5.B. Step ClimbingThe ability of the vehicle to climb steps was also confirmedas shown in Fig.16. The height of the step is 33.5mm. It was observed that a prototype with the Omni-Crawler mechanism can climb step not only longitudinally, but also laterally.123Fig.16: Example Motion of Step ClimbingWhen the crawler vehicle needs much higher ability to climb steps, the configuration can be set the joint mechanism like connected crawler vehicle “Soryu[15]” by removing the front supporter.C. Moving on PipeThe vehicle’s mobility on the outside edge of a pipe was also confirmed. It was observed that the prototype with the Omni-Crawler mechanism can traverse along small and large pipes without any adjustments as show in Fig.17. Similarly, motion along the inside of a pipe was also observed, and it was confirmed the vehicle could maintain smooth motion without requiring any kind of adjustment. See Fig.18. The diameter of the outer pipe is 513mm and inner diameter of the pipe is 490mm.0[s]12342[s]4[s]6[s]PipeMotion DirectionMotion Direction(c) Example MotionFig. 17: Moving on Pipes: small and big0[s] Vehicle 12 3 41[s]2[s]3[s]54[s]PipeMotionDirectionMotionDirection65[s]76[s]87[s]98[s]Fig. 18: Moving on the inside of a PipeD. Moving on Soft GroundsFig. 19: Moving on soft ground (sand)The vehicle’s ability to move on soft ground was also confirmed, as shown in Fig. 19. We used “Toyoura” sand as the soft ground. It was observed that this prototype with the Omni-Crawler mechanism can move on soft ground smoothly with a low level of sinkage.VI.C ONCLUSIONIn this paper, we showed a new crawler mechanism inspired by the “Omni-Ball” that has a circular cross-section and can realize sideling motion. An omnidirectional vehicle that uses this wheel mechanism was also presented. We confirmed the basic characteristics of the crawler mechanism and the motions of the crawler vehicle through experiments.In future works, we plan to optimize the mechanism of the crawler as well as the materials of the circular lug includingthe suspension mechanism.A CKNOWLEDGMENTThe authors would like to thank Prof. Shigeo Hirose for his contribution and abundant advice. And we would like to thank Mr. Hiroaki Kinoshita and the staff of Ono-denki Corp. for their contribution.R EFERENCES[1] A control system for an omnidirectional mobile robot, Paromtchik,I.E.; Asama, H.; Fujii, T.; Endo, L.;Control Applications, 1999.Proceedings of the 1999 IEEE International Conference on Volume 2, 22-27 Aug. 1999 Page(s):1123 - 1128 vol. 2[2]Atsushi Yamashita, Hajime Asama, Hayato Kaetsu, Isao Endo andTamio Arai: “Development ofOmniDirectional and Step-ClimbingMobile Robot”,Proceedings of the 3rd International ConferenceonField and ServiceRobotics (FSR2001), pp.327-332, Espoo(Finland), June 2001.[3]/ (at the present day, April 8th)[4]Low vibration omni-directional wheel, Patent number: 6547340, Filingdate: Dec 6, 2001, Issue date: Apr 15, 2003, Inventor: Donald Barnett Harris, Assignee: Airtrax Corporation, Primary Examiner: S. JosephMorano, Secondary Examiner: Long Bao Nguyen, Attorney: Foley & Lardner[5]Omni-directional munitions handling vehicle; Patent number: 6668950,Filing date: May 9, 2002, Issue date: Dec 30, 2003, Inventor: Andrew D.Park, Primary Examiner: Kevin Hurley, Attorney: Schwartz Law Firm, P.C.[6]http://www.kanto-aw.co.jp/en/[7]"Omnidirectional Vehicle", Applicant: Kanto Auto Works , Inventor:Hideki Toda, Japanese Patent application for a patent:2003-203645,date of application:30th, July, 2003, publication number: Patentdisclosure2005-47312, date of publication of unexamined patentapplication: 24th, Feb, 2005,[8]"Omnidirectional Wheel and Mobile Platform", Applicant: ShinichiFuji, Inventor: Shinichi Fuji, Japanese Patent application for apatent:2004-367219, date of application:20th, Dec, 2004, publicationnumber: Patent disclosure2006-168659, date of publication ofunexamined patent application: 29th, Jun, 2006[9]S. Hirose and S . Amano, ‘The WTON Hi& Payload, High EfficiencyHolonamic Omni-Directional Vehicle,” in Proc. 6th Int. Symp. OnRobotics Rerearch, (1993).[10] “Design of a holonomic omnidirectional vehicle”, West, M.; Asada,H.;Robotics and Automation, 1992. Proceedings., 1992 IEEEInternational Conference on 12-14 May 1992 Page(s):97 - 103 vol.1 [11] Chen,P.,S.Mitsutake,T.lsoda,andT.Shi,Omn robot and adaptivecontrol method for off-road running,IEEE Transactions onRobotics and Automation,volume18 (2002), number pp.25 1.2561[12]Tadakuma, K.; "Tetrahedral Mobile Robot with Novel Ball ShapeWheel", The First IEEE/RAS-EMBS International Conference onBiomedi-cal Robotics and Biomechatronics, 2006. BioRob 2006.February 20-22, 2006 Page(s):946 – 952[13]“Development of holonomic omnidirectional Vehicle with“Omni-Ball”: spherical wheels,” Kenjiro Tadakuma,; RiichiroTadakuma,; Berengeres, Jose; IROS 2007. IEEE/RSJ InternationalConference on Intelligent Robots and Systems, Oct. 29 2007-Nov. 22007 Page(s):33 - 39[14]“Design and control of ball wheel omnidirectional vehicles”, West, M.;Asada, H.; Proceedings., 1995 IEEE International Conference onRobotics and Automation, Volume 2, 21-27 May 1995 Page(s):1931 - 1938 vol.2 Digital Object Identifier 10.1109/ROBOT.1995.525547 [15]Toshio TAKAYAMA, Shigeo HIROSE; “Development of “Soryu I &II” -Connected Crawler Vehicle for Inspection of Narrow and Winding Space-”, Journal of Robotics and Mechatronics, Vol.15, No.1,Feb.2003, pp61-69[16]"Robot Mechanisms and Mechanical Devices Illustrated", Paul e.Sandin, pp220-221, The MCGraw-Hill Companies, ISBN 0-07-141200[17]http://www.valuewave.co.jp/moba/mv.htm[18]/versatrax150.htm。
