Newsletter April 2007


IEEE RAS Technical Committee on Algorithms for Planning and Control of Robot Motion

Newsletter April 2007

Dear Committee Members,

It was great to see many of you at our informal get-together in Rome!

This newsletter features a brief article by Calin Belta from Boston University - see #3 below. I hope that you find it interesting and that it will inspire you to submit a similar exposé about interesting research relevant to this committee.

By the way, our membership has grown to 80! Thank you for your recruiting efforts!

In this newsletter:

  1. Open Letter by Lydia Kavraki (previously sent by email)
  2. RSS early registration deadline approaching
  3. Symbolic Approaches to Motion Planning by Calin Belta
  4. Wiki Instructions

If you have not done so already, please add yourself to the directory. (How? Please see below.)

Please spread the word about this committee! Recruit new members, or suggest activities!

Yours truly,

Oliver Brock
Tsutomu Hasegawa
Steven LaValle
Thierry Simeon



1) Open Letter by Lydia Kavraki

Dear Colleagues,

I write with great sadness to let you know that Andrew Ladd passed away on March 4, 2007 in his sleep. He bravely fought a long and difficult battle with cancer.

Andrew spent almost six years at Rice and obtained his Ph.D. from Rice, before moving to the University of Michigan Ann Arbor as an assistant professor.

In his short career, Andrew has been very productive. Most of his work has been on robot motion planning. His 2005 Robotics Science and Systems paper described a novel sampling-based motion planning algorithm that became his Ph.D. dissertation work. Andrew designed an effective sampling-based planner for systems with second-order dynamics and other non-trivial dynamics, and he integrated tightly physics-based simulators with his planner. I found his work very impressive.

I will always remember Andrew for his fierce intellect, his insatiable curiosity, his bottomless well of ideas, his energy and his amazing will-power. All of us at Rice will miss him greatly.

Andrew is survived by his wife, Fumiko Chino (fumikochino@gmail.com), his mother and father, Debra and Stuart Ladd (stuart.ladd@sng.ca), and his sister, Kelly Ladd.

I would like to thank all of you who contacted us regarding Andrew during the recent weeks.

We have lost great colleague.

Sincerely,

Lydia Kavraki

P.S. For those of you have inquired, here is some information for the memorial fund his family has set up:

Andrew Ladd Rice University Fellowship - in trust c/o Schlesinger Newman Goldman 625 Rene Levesque Blvd. West Suite 1100 Montreal, Quebec H3B 1R2

Checks can be sent either in U.S. or Canadian currency.




2) RSS Early Registration Deadline Approaching

Please note that early registration for RSS ends in just one week, on April 27. To register, please follow this link.

April 28 will be the last day to book accommodation in student dorms (US$ 35 per night). To book the student dorms, please follow this link.

The organizers try to keep RSS as affordable as possible, in particular for students, to encourage large attendance, even from people that do not have a paper at the conference. Early registration for students is just US$ 99 and the registration fee includes the banquet, which will be at the Atlanta Aquarium. The program features seven invited talks, 41 technical papers, and six workshops. Workshop attendance is free with conference registration. The preliminary technical program of this year's RSS is posted on this page.

For a list of workshops, please see this page.




3) Symbolic Approaches to Motion Planning

Approaches to robot motion planning and control that involve tools such as automata, temporal logics, and grammars have been recently termed "symbolic". Currently, there are two main directions in symbolic motion planning and control. Top-down approaches emphasize the expressivity of the specification language, and draw inspiration from formal verification and temporal logics. In such approaches, high level specifications over regions of interest in the environment (i.e., "Visit A, then either B or C, then D and F, in this order, infinitely often. Always avoid obstacle G") are converted into robot control strategies [1,2,3]. However, due to sensing, control, and mechanical constraints, such strategies might not be executable by real robots. On the contrary, bottom-up approaches are focused on the construction of specification languages that explicitly capture robot constraints [4,5,6]. As a result, the expressivity of such specification languages is, in general, restrictive.

The main challenge in this emerging area is to bridge in the gap between these two types of approaches [7]. The result will be a computational framework for systematic, provably-correct control design accommodating both the robot constraints and the complexity of the environment, while at the same time allowing for expressive task specifications.


REFERENCES

[1] C. Belta, V. Isler, and G. J. Pappas, "Discrete abstractions for robot planning and control in polygonal environments," IEEE Trans. on Robotics, vol. 21, no. 5, pp. 864-874, 2005.

[2] G. Fainekos, S. Loizou, and G. J. Pappas, "Translating temporal logic to controller specifications," in 45th IEEE Conference on Decision and Control, San Diego, CA, December 2006.

[3] M. Kloetzer and C. Belta, "A fully automated framework for control of linear systems from LTL specifications," in Hybrid Systems: Computation and Control: 9th International Workshop, ser. Lecture Notes in Computer Science, J. Hespanha and A. Tiwari, Eds. Springer Berlin / Heidelberg, 2006, vol. 3927, pp. 333 - 347.

[4] M. Egerstedt and R. Brockett, "Feedback can reduce the specification complexity of motor programs," IEEE Transactions on Automatic Control, vol. 48, no. 2, pp. 213-223, February 2003.

[5] E. Frazzoli, M. A. Dahleh, and E. Feron, "Maneuver-based motion planning for nonlinear systems with symmetries," IEEE Trans. on Robotics, vol. 21, no. 6, pp. 1077-1091, December 2005.

[6] A. Bicchi, A. Marigo, and B. Piccoli, "Feedback encoding for efficient symbolic control of dynamical systems," IEEE Transactions on Automatic Control, vol. 51, no. 6, pp. 1-16, 2006.

[7] C. Belta, A. Bicchi, M. Egerstedt, E. Frazzoli, E. Klavins, and G. J. Pappas, Symbolic Planning and Control of Robot Motion, IEEE Robotics and Automation Magazine, Special issue: Grand Challenges of Robotics, vol. 14, no. 1, pp. 61-71, 2007


BIO

Calin Belta received M.Sc. and Ph.D. degrees in Mechanical Engineering from the University of Pennsylvania in 2003. He is currently an Assistant Professor in the College of Engineering at Boston University. His research interests include verification and control of hybrid systems, robot planning and control, gene and metabolic networks. Calin Belta received an NSF CAREER award in 2005, a Fulbright study award in 1997, and was the Valedictorian of his class in 1995. He received the best poster award at the Int. Conf. on Systems Biology in 2004 and was a finalist for the ASME Design Engineering Technical Conference best paper award in 2002 and for the Anton Philips best student paper award at the IEEE Int. Conf. on Robotics and Automation in 2001.




4) Wiki Instructions

Most of the pages can be edited using the password [hidden]. Please take a look at this page for some instructions on how to edit the Wiki. There is a sandbox on this page where you can try out some of the features of the Wiki. Editing the Wiki is very easy and we hope that this ease will encourage you to contribute.