From nogo to go in three tries NSF

From no-go to go in three tries NSF CAREER Workshop April 4 -5, 2011 Sean B. Andersson Mechanical Engineering Division of Systems Engineering Boston University www. bu. edu/anderssonlab

About me and my research • Systems and control (dynamics in nanometer-scale systems) • CAREER from NSF-CMMI-Control Systems Program Non-raster scanning in AFM Tracking in confocal microscopy

Every CAREER starts with a good idea • New direction: control in AFM and confocal microscopy • Ph. D: Geometric control theory • Many months of • • Idea development Idea refinement Idea scrap-ment/restart-ment Preliminary work Tracking in SPM High-speed AFM Non-raster imaging Single particle tracking in confocal Direct tracking of molecules Need to develop a five year PLAN

Preparing to write…and then writing • Read MANY successful and unsuccessful CAREERs • Asked for examples from friends, colleagues • Sought common features to emulate and common mistakes to avoid • Spent significant time on outreach • Ensured it was tightly coupled to research program • Drafted, edited refined…and then sent to colleagues • Re-edited, re-refined…and then sent to colleagues • Ph. D advisor, postdoctoral advisor, friends, senior members in the department, junior members that recently won the CAREER Four to five months on writing

So, how’d that work out for me? • Scores: V, V, E, G • Fund if possible. . . But it wasn’t possible • Proposal was well-written with good ideas but: • Ideas were too vague • (But it’s a five year CAREER! How can they be? ? ) • Connection to program goals (control) were weak • Note: outreach was well-received • Lessons learned: • Have a tightly focused (but broad!) research program • Provide sufficient detail • Don’t use your first CAREER as your first proposal!!!

The second round – revising the ideas • Received result in March – started again in June • Maintained overall structure, concepts, goals • My response to reviews: add (a bit more) control • bit of nonlinear here, dash of adaptive there • Read/reviewed by Ph. D advisor and departmental colleague • Only one round • After all, first proposal was well-received, right?

So how did it work out? • Scores: E, VG/G, G, G • Fund if possible – but it wasn’t again • Well-written with good blend of theory and experiments but… • Lack of technical details • Role of the advanced control material was not clear • Why was it needed? Is it better than other methods? How will it be done? • Lessons learned • • This was the hardest; I was clearly my own undoing Be sure to understand panels concerns – and address them Be tightly focused in context of grander plan Do not add vague ideas!!!

Taking it to the next level • Talked to PM • Spent sufficient time to really understand panels concerns. New things came out: • AFM work hard to fund, budget was a concern, lack of details was the killer • Discovered and attended this workshop • Came in uncertain what the workshop could really do • Clearly I was a victim of randomness • Left with: • Attitude change: I was in charge and there were clear deficiencies to address • New ideas on organization

Back to the drawing board • Invested time in rethinking research plan • Narrowly focused, specific research objectives with details • Nowhere for reader to ask how? why? • Expressed focused problem in context of broader control research • Coherent organization- for each topic (including education) • Intro – State of the art – My prior work – Research plan WROTE A SUBSTANTIALLY NEW PROPOSAL • Sent to colleagues, edited, sent to colleagues • Foreshadow: most noticed third goal was not as well-described or well-planned • My reaction: That’s in the fourth and fifth years so of course… • Organized on-campus mock panel for further feedback

So how did it work out? • Scores: E, V, V, G • Fund if possible – Ack! • Panel noted particularly • Some details (particularly on third goal!!!) were missing • Proposed approaches lacked some supporting theory • However, overall was strong proposal • Even the G review was positive • NOTE: had also submitted more focused proposal to standard program • More specific goals, more details on approach • PM was going to fund it and then… ARRA to the rescue!

Tips from experience: DON’T • Rush • Brush off/ignore advice • Leave unanswered questions • Forget it’s both a research proposal and CAREER plan • Fret over stuff you cannot control • Blame results on that stuff

Tips from experience: DOs • Have a solid 5 year plan • shouldn’t you anyway? • Read as many CAREER proposals as you can • both successful and unsuccessful • Think carefully about organization • It doesn’t happen (well) holistically • Get your draft done early so that you can… • Get as many people to read it as you can • Use some new people after each round of edits

And one more don’t • Don’t approach this as make it or break it • If successful – great! • If unsuccessful – you may not have a CAREER but you can still have a career Good luck!