There are a lot can be written... but you just need to draw a fair line...

 

I'm not sure, at the point of preparing a 180-page of document, I don't know what is the limit. Perhaps, that is called the wisdom, where there is no limit. When you get addicted, you are just into it...

 

But, all I want is simple... I just need to graduate~!

 

Bless me on the 17th of November... I just want to graduate...

 

很久没有贴图 整个blog死气沉沉

不知不觉 秋天又来临

第九个秋天

我希望 在今年下雪之前 可以离开这个地方

我会想念秋天 因为即将到一个不怎么有秋天的地方

也会想念雪花 因为那里下雪是once in a blue moon

两年前的今天 为雪灾抗战

一年前的这个时候 为“水灾”抗战

今年的这个时候 为论文抗战

都这样熬过去了 丰富了人生经验

 

Anyway 还是会想念枫叶的

尤其照片中用久别的S45在Ottawa照下来的

看了CTT用n多！！！来发泄，我也想发泄了 - 还是不知道怎么发泄的说。

 

日子一天一天的过，睡眠还算充足有规律，1点睡8点起。9点到实验室，就对着一样的东西打。嗯，确实很恨这个Word档案才一直“打”。Copy，paste，改，删，加，整理，读一遍，不顺，改，加，整理，顺了，继续，131页，啊～！疯了。

 

偶尔看看DOW有没有掉到历史新低，或看看总统辩论，感受一下每个人还是这样不容易地熬过去的来激励一下。

 

告诉很多人说论文不是容易的事情，他们还是不了解，我想还是不问比较好。如果还是执着想明白，今天开始给自己写你过去8年的自传就会明白了，看你多久能写完。

 

磨着磨着，好像越来越多东西要做要整理，没完没了的，问我完成百分之多少，我可以说这也是一个research problem。与其“研究”还剩百分之多少，不如继续写好了。

得知一个很气愤的消息，真的很气愤。$700B磨了两周终于拨出来了，没想到该死的AIG要了钱，什么都不管，让一班executive先拿了$200M去度假，今天还向政府多要一点钱，说不够了。你说，这种公司还要来干嘛。很想告诉很多在马来西亚做AIA保险经销的朋友就不干了，投保的就退了算了。这种为自己的利益而没想到别人的公司还算什么保险公司。

 

有人问我为啥管这种事。我也给美国纳税7年了，就这样捐给Bush和Wall Street的烂人“度假”，还要被当作alien不平等看待，当然不爽。和Pat今天聊到Social Security，因为昨天的总统辩论就说到了这点，我还想起前阵子不知道那里看到说大概20-30年后Social Security会是负资产的，所以这个年代的年轻人基本上不可能退休的。晕，这是什么fact？！有时觉得怎么这个世界就是那么乱？贪，真的会害死很多人的！

 

 

没想到上一则blog是在911那天写的. 不好意思了, 美国. 不小心又碰到你的伤口. 哎, 说到伤口, 我觉得美国近来一连串的冲击, 元气大伤之下, 不知道什么时候才能恢复了. 最近美国大选campaign很热, 是不是还会和Pat聊一聊相关的事情. 最近似乎都不怎么care那两个总统候选人的campaign, 天天就守着Capitol Hill, 盼望着好消息. 或者finger crossed, 希望不会再有任何坏消息从Wall Street传出来. 美国Wall Street风风雨雨, 加上New Orleans和Houston的风风雨雨, 老百姓和我们这些"alien"也不知道还好不好过.

 

毕竟, Buffalo还是一个风平浪静的小城市, 似乎居民都没感受到什么. 是看到很多房子外面都多了一块小板, 希望原来房子的主人还活着. 前阵子很多人听到我说要Dallas的时候, 第一反应都是说那里房子便宜, 或什么时候考虑买房子的事, 我觉得现在这种危机之下这种话题还是不说好了. 这种话题在很多美国人的面前说, 估计都打击很大. 哎, 我真的还是第一次听说房地产会降值的啊! 可怜的老百姓以为投资房产会保住自己的"身家", 没想到会比自己的"新"车贬值的更快.

 

我觉得对于我最牛的还是WaMu有问题的时候. 之前被它的高利息吸引开了一个帐号, 在它有宣布有问题的前一两天还涨了利息, 我还把一大笔"家产"转过去转一点利息. 没想到一下就变成Chase了. 很多人慌了, 可我不怎么慌. 有经验的人还是告诉我应该不应该慌了. 为了安全起见, 还是把一些钱转回来了. WaMu毕竟曾是美国银行业老大, 没想到还要沦落到这种下场. 至少被Chase收购总比破产让老百新更惨好. 还以为下去Dallas后可以去开一个Chase拿个$150的cash back, 现在WaMu不知不觉被"默认"成Chase了 - 我的$150啊! Ouch, 被打了一下. 没损失已经好了, 生在福中要知福嘛.

 

我觉得比较看不下去还是马来西亚的政治. 整个世界的经济局势已经深深影响到老百姓了, 却上面的领袖还在作权利斗争, 还搞内安发令抓无辜的人. 哎, 不着手把真经事做好, 只担心自己还能不能称王, 这种领袖还是不要算了. 中国那里搞个毒奶粉时间, 报章上三番两头就来一个什么食物在某地被禁了. 看着饭盒里的食物, 还是拿在手里的汉堡, 有毒么? 还有什么能吃的! 粤语有说:"大菌吃小菌, 小菌就当没菌." 或者就当"补品"吧. 随便吧, 不会那么容易出事的. 要不然就任命了.

 

接下来的时间, 我觉得这个世界的老百姓得更刻苦才能在活着了. 我觉得, 重点还是, 每个人不能太自私了, 还是得团结一点的, 才有希望把问题解决. 毕竟, 政府的能力还是有限的. 整个9月全世界波涛汹涌了一下, 10月? 看看吧... finger crossed... good luck!

下周一准备下去Dallas一下，周三回来Buffalo。H1-B申请材料给了HR，不知道何时何日可以收到。还在看如何relocate下去Dallas，很多事还很uncertain。觉得这两个月应该会是很miserable的时段。希望一切都会依照计划如愿以偿。

Gustav... please be kind...

令人“期待”的岜东埔补选成绩终于揭晓。这马来西亚国庆前夕的热门话题果真有个出乎所料的结果。与其说出乎所料，不如说安华牛得不得了的头脑，赢得也很值得。我想，世界上真的没有他那么聪明的，Time杂志说他是2008年100个最具影响力的人物之一还是没有错的。我没说我支持谁，但我觉得从被监狱放出来，到今天他重归国会的每一步，他所做过的每一件事，包括油价的辩论，都有很深的意义，都令我非常佩服。他那种不认输的精神，从不可能化成可能，令我想起Shawshank Redemption的主角。安华，你真的太牛了！佩服！

I think there are enough people who have been asking me about my job search, let me just summarize somehow. As for now, I received offer letter as a Software Engineer from Delphi. This is positive although I'm not going to choose this path ultimately since this shows I'm still able to get a job in the industry. In any case, I'm still waiting for the offer letter from Mark Spong from University of Texas - Dallas. My approach right now is not until I see the offer letter, I'm not going to make any decision based on that. So, hopefully the letter from Dr. Spong will be arriving very soon, so that all the parties don't have to keep on waiting for my decision, especially Delphi which has been constantly bugging me for decision.

 

RPI's application process wasn't a happy ending. What I was disappointing about was their decision were discussed behind me with my advisor without my acknowledgment. Till today, I haven't received any refusal letter from them. I don't feel this is very a professional way of managing things. Well, I think partly it was my fault since I didn't tell anyone that I have another favorable offer. Refusal letter to Yale has also been sent, without any reply, which made me worried about my relationship with the supervisor in the later stage.

 

Well, I think everything wasn't that bad. I just have to refuse Delphi at some point in time, and continually work on my way to finish my dissertation.

 

I'm quite satisfied with the performance of Lee Chong Wei in the Olympiad. Finally, Malaysia doesn't have to be going home without a medal. Well, I really hope that it will be a gold. But, to this stage, I guessed it is one of the best performance in the Malaysian history in the Olympiad. Chong Wei, go get the first gold for Malaysia! Good luck!

从找资料... 到起稿... 到归纳... 到看n种不同的cover letters的写法... 到整理... 到改grammar... 到proof-read n次... 前后大概一封花2-3天的时间吧. Cover Letter还是不容易写的, 还是不错的payoff的.

 

给John Wen的:

I am writing to you to seek for the opportunity to work with you as a postdoctoral researcher. I am currently a graduate student in University at Buffalo and expected to receive my PhD degree by this September. The topic of my dissertation is “Design and Control framework for Cooperative Mobile Manipulator Collectives”, and my academic advisor is Professor Venkat Krovi. My research interest lies in the span of distributed multi-agent robotic system, analysis of parallel mechanisms, and control of mobile and locomotion systems.

 

In my graduate research work, I have been involving in the development of theoretical and implementation framework to achieve cooperative payload transport and manipulation by multiple mobile manipulators. Interest in such cooperative systems typically arises when certain tasks are either too complex to be performed by a single agent (such as carrying a huge container), or when there are distinct benefits in terms of redundancy, robustness and reliability that accrue by many simple cooperative robotic modules. However, the mechanical capability of both the individual modules as well as their interactions can significantly affect the overall system performance. Hence, in the design context, I developed the system-level models to systematically analyze the role of mixture of holonomic/nonholonomic constraints and active/passive/locked joints in affecting the collective performance in terms of feasibility, singularity, manipulability, controllability and disturbance rejection capability. On the other hand, in the control context, the proposed architecture permits natural decentralization in terms of kinematic control. This leverages the development of bi-level hierarchical control implementation that combines the upper-level planning of the kinematically compatible desired trajectories and the lower-level posture stabilization controller to accommodate, detect and correct for configuration errors. Finally, in the validation context, we employed the hardware-in-the-loop experimentation within the MATLAB/Simulink/xPC Target framework, which can potentially be extended to the future teleoperation-based deployment.

 

My long-term career goal is to obtain a faculty position in a research university. To this end, I believe that with the collaboration with you, I can further enhance my analysis framework for the cooperative robotics to a more general setting to establish a more solid research program. Specifically, as noted in some of your prior publications, due to the tight physical contact between multiple manipulators over the payload, there is a need to carefully analyze the required “information flow” between the agents (in this case, the “information” is the transmission of motion and internal forces) to achieve consensus collaborations. Furthermore, if the system is being teleoperated, the teleoperator can be analyzed as an additional agent with different behavior, but now the information flow between the systems would be the wired or wireless Ethernet-based network flow. Issues such as packet loss and time delay can then further critically affect the performance of such framework. Furthermore, there is a growing interest in the decentralization of control schemes among multiple (human and robot) agents to achieve consensus centralized goal though minimal interaction/communication. My research plan is to establish better understanding and systematic analysis to such robot-robot and human-robot interacting behaviors over different modes of networked flow/interaction, and ultimately provide theoretical guidelines to develop better decentralized controllers. I believe that subsets of the above research theme and developed experience would be relevant to your current active research programs, especially in the work in networked distributed/cooperative control systems.

 

Finally, although my dissertation work focuses on cooperation between multiple mobile manipulators, I have also been explored to knowledge in various fields throughout my graduate studies. I involved in attacking some other issues related to the abovementioned research topic in the form of collaboration with other labs. They include the development of differential flatness-based kinematic and dynamic control for a mobile manipulator (with Professor Sunil Agrawal at University of Delaware), and the evaluation of various real-time distributed simulation schemes for multibody dynamics with closed structure (with Waseem Khan at McGill University). I also have general knowledge in differential geometry, numerical analysis, parallel computation and optimal control. Furthermore, I possess teaching experience in the broad area of control systems, mechanism analysis and synthesis, mechatronics and computer aided design/engineering, where in some cases I was the instructor of the courses. I believe that my future postdoctoral researcher experience with Center for Automation Technologies and Systems in RPI will provide me more valuable opportunity to continue to work with great researchers to tackle more challenging multidisciplinary engineering problems.

 

I have enclosed a copy of my curriculum vitae and the names and contact information of the references. Please do not hesitate to contact me if you need further information. I will certainly respond to you as soon as possible. I appreciate your time in reviewing my applications. I look forward to hearing from you.

 

给Mark Spong的:

My research interest lies in the span of distributed multi-agent robotic system, analysis of parallel mechanisms, and control of mobile/locomotion systems. In my graduate research work, I have been involving in the development of theoretical and implementation framework to achieve cooperative payload transport and manipulation by multiple mobile manipulators. Interest in such cooperative systems typically arises when certain tasks are either too complex to be performed by a single agent (such as carrying a huge container), or when there are distinct benefits in terms of redundancy, robustness and reliability that accrue by many simple cooperative robotic modules. However, the mechanical capability of both the individual modules as well as their interactions can significantly affect the overall system performance. Hence, in the design context, we developed system-level models to systematically analyze the role of mixture of holonomic/nonholonomic constraints and active/passive/locked joints in affecting the collective performance in terms of feasibility, singularity, manipulability, controllability and disturbance rejection capability. On the other hand, in the control context, the proposed architecture permits natural decentralization in terms of kinematic control. This leverages the development of bi-level hierarchical control implementation that combines the upper-level planning of the kinematically compatible desired trajectories and the lower-level posture stabilization controller to accommodate, detect and correct for configuration errors. Finally, in the validation context, we employed the hardware-in-the-loop real-time experimentation within the MATLAB/Simulink/xPC Target framework, which can potentially be extended to the future teleoperation-based deployment.

My long-term career goal is to obtain a faculty position in a research university. To this end, I believe that, with the collaboration with you, I can further extend the above analysis framework to a more established research program, especially in involving human operator in the loop to realize remote semi-autonomous teleoperation/telemanipulation. As noted in some of your prior publications, such implementation is very attractive yet challenging since the communication between human operator and robots, and the coordination and synchronization between the robots are the key issues to affect the performance and stability of the overall cooperative system. My research plan is to treat the robots and human operator as individual agents, and unify their interactions and required control actions based on their varying grades of communication limitation in a networked framework. Such requirements can be categorized based on (a) tight cooperation, (b) loose cooperation and (c) teleoperation, each required different control schema for coordination and synchronization, but potentially using the same networked architecture. The first case corresponds to the situation when all the robots are attached to the common payload, and the information (in the form of motion and/or forces) from any agent can be immediately transmitted from one to another through the payload due to their physically tight attachment (given that the payload is rigid). Hence, their control actions can be straight-forward without delay and information loss. The second case corresponds to the situation when the robots are not attached to the payload, and the robots require some form of communication protocol (wirelessly through centralized supervisor/leader) and/or sensors (ranging/bearing sensors, omnidirectional camera, GPS, etc.) to identify each other's actions. Hence, the coordination and synchronization process in this case would require more careful control strategies that take into account the moderate delay and information loss. In the final case, the interaction between the human operator and the robots are more distant as compared to the second case, and hence experiencing significant delay and information loss. Furthermore, the input command and (force) feedback are quite different from the first two cases. Hence, my plan is to establish better understanding and systematic analysis to such robot-robot and human-robot interacting behaviors over different modes of networked interaction to ultimately provide theoretical guidelines to develop better decentralized controllers through minimal interaction and guaranteed stability. Most of your prior work has built the foundation to explore the above possible research avenue.

I believe that subsets of the above research theme and developed experience would be relevant to your current active research programs. I also believe that my future postdoctoral researcher experience with you and your co-workers in University of Texas at Dallas will provide me valuable opportunities to tackle more challenging multidisciplinary engineering problems.

 

给CSU的:

I am writing to you to express my interest in applying for the full-time lecturer position advertised in the Summer 2008 issue of the ASEE PRISM Magazine. I am currently a graduate student in University at Buffalo and expected to receive my PhD degree in Mechanical Engineering by this August. I believe that my extensive teaching experience in mechanical engineering, specialized in controls, robotics/mechanisms and computer aided design/engineering, and my research experience in the design and control of cooperative mobile manipulator collectives have provided me with the background necessary to make me a strong candidate for the lecturer position.

 

As shown in my curriculum vitae, I have given excellent opportunities to be an instructor for 2 semesters to teach both undergraduate and graduate level courses during my graduate studies. I was given full responsibility to design and organize the course structure to meet the syllabus need. I used the online course management system effectively not only to post lecture notes and assignments, but also interact with the students using the forum-based discussion board in interactive and collaborative manner. I also successfully implemented several game-based group-learning teaching methodologies in the class to allow the students to learn fast and play the role as a team member. I find that, to effectively teach students in mechanical engineering, it is vital to illustrate real application examples in the lecture so they can remember forever, and to enhance the interactions between students and instructor both inside and outside the classroom.

 

I have also given opportunities to be a teaching assistant for 6 semesters, where I worked closely with the professors and other teaching assistants as a teaching “team”. When teaching System Analysis, I not only teach in the hands-on lab course to introduce the students to the transient dynamics of mechanical systems in different domain, I also took the role as a “Head TA” to lead other junior TA’s to get up to speed to teach in different sections. I also taught more advanced control courses in Digital Control and Mechatronics, where the students implemented computer and microcontroller to control various electromechanical systems – allowing me to design and setup the laboratory equipments independently. I also taught Machines and Mechanism course, and co-developed an online tutorial that allows students to learn virtual prototyping skills at their own pace. The results of the tutorial was published in both conference and journal articles. Finally, I was also invited to give multiple lectures on numerical methods in the course in robotics, and to give multiple presentations on robotics and mechatronics in various high school workshops. I believe the above cumulated multi-disciplinary experience will be valuable as a successful lecturer.

 

I have enclosed a copy of my curriculum vitae and the names and contact information of the references. Please do not hesitate to contact me if you need further information. I will certainly respond to you as soon as possible. I appreciate your time in reviewing my applications. I look forward to hearing from you.

说说Mark Spong吧。个人是非常欣赏这头大牛的。想起来，我买的第一本Robotics的书，就是他写的一本很经典的Robotics教材。那本书还是在马来西亚的时候不知为啥会在Kinokuniya买下的，比起美国的价格便宜多。可惜上次“水患”吧那书给毁了。后来赔回来的是那本书的再版，很新很新。他算是Robotics方面的先锋，很多今天很重要的Robot控制系统都是他导出来的。看看这些吧：

Mark W. Spong is Donald Biggar Willett Professor of Engineering, Professor of Electrical and Computer Engineering, and Research Professor in the Coordinated Science Laboratory at the University of Illinois at Urbana-Champaign. He received the B.A. degree, magna cum laude, in mathematics and physics from Hiram College, Hiram, Ohio in 1975, the M.S. degree in mathematics from New Mexico State University in 1977, and the M.S. and D.Sc. degrees in systems science and mathematics in 1979 and 1981, respectively, from Washington University in St. Louis.

He has been at the University of Illinois at Urbana-Champaign since 1984. From 1981 to 1982 he was with the Department of Electrical Engineering at Lehigh University. From 1982 to 1984 he was with the School of Electrical Engineering at Cornell University. He has held visiting positions at the National Polytechnic Institute of Mexico, the University of Waterloo, Ontario, Canada, the Laboratoire d'Automatique de Grenoble, France, The Lund Institute of Technology, Lund, Sweden, the Katholieke Universiteit, Leuven, Belgium, The Universite d'Tecnologie de Compiegne, France, the National University of Singapore, and the Technical University of Munich, Germany.

Dr. Spong was the 2005 President of the IEEE Control Systems Society. He was Vice President for Publication Activities of the IEEE Control Systems Society from 2000-2002, Editor-in-Chief of the IEEE Transactions on Control Systems Technology from 1997-2000, and a member of its Board of Governors from 1994-2002. His research interests are in nonlinear control theory and robotics. He has published over 200 technical articles in control and robotics and is co-author of four books, Robot Dynamics and Control, John Wiley & Sons, 1989, Robot Control: Dynamics, Motion Planning, and Analysis, IEEE Press, 1992, Robot Modeling and Control, John Wiley & Sons, 2005 and The Reaction Wheel Pendulum, Morgan & Claypool Publishers, 2007.

Dr. Spong is a Fellow of the IEEE, a Distinguished Member of the IEEE Control Systems Society and a member of Phi Beta Kappa. His recent awards include the 2007 IROS Fumio Harashima Award for Innovative Technologies, the Senior Scientist Research Award from the Alexander von Humboldt Foundation, Bonn, Germany, the IEEE Third Millennium Medal, and the O. Hugo Schuck Award and John R. Ragazzini Award from the American Automatic Control Council. In addition, he is President of Mechatronic Systems, Inc., a company he founded in 1996.

怎么说呢？我觉得我在他的手下压力还是很大的。看了他很多博士生的论文的Acknowledgment，觉得他还是一个很善良的老师，是个优点。我觉得比较容易从这种老师学到很多。他的成就也真的太夸张了。现在去开会，我常常得说我是Krovi的学生，有些人还认识。有时得加说Krovi是Kumar的学生我才占得了一点名气。如果和Spong的话，我想我不需要这样了吧，就说我是Spong的postdoc，全世界都知道是谁了。晕了，明显我在发梦了。

又有一点点进展：

1. CSU-Fullerton的phone interview还算顺利。本来说只聊半个小时却聊了一个小时。本来以为只和dean聊，后来听到一些背景笑声，觉得应该是phone conference了，虽然其他人都没说话。Anyway，觉得一个小时聊得还不错，只是好像不能达成多大的共识，因为他们是存搞education的，不做任何research。而且，他们好像都要教一些很奇怪的科目，而且要我做好心理准备教一些根本和我背景没有相关的科目。个人觉得即使拿到了，也得三思。不过这周会有答复。拿到的话也算是一种成就。

2. 决定了要去Yale参观一下了。和那里的人聊得不错，他们好像觉得我是要去那里工作了。Anyway，旅馆订好了，准备开车下去。就要准备一下presentation什么之类的。

3. 这个就可大了。今早突然接到RPI的电话，说要招postdoc了。啊，John Wen显灵了！聊了大概半个小时，共识挺多的。他们要做construction robot，基本上就和mobile robot没啥区别，还不错。不是和John Wen聊，和research associate聊。后来我们还聊到teaching方面的东西。他还说他们最近走了一个教mechatronics的老师，需要一个人来教，可以找我。啊！晕了。他还说看了很多其他的简历，觉得我比较趋近他们的方面，比较favor我。最后还决定了，8月头去他们那里给个talk。啊！啊！太突然了，我没推，反正又多一个选择了。

4. 寄了简历给UPenn的Dr. Kumar，虽然是老板的老板，其实不抱多大希望，还是试试看吧。

5. USC的老师正式寄了一封信把我给推了，看来加州不去了。

6. Dr. Spong至今没反应，也不知道所有的reference letter到了没。寄n多信也没人回。我想，算了。

好了，我得加油，再好好考虑这些options了。