Optical Fiber ICRA'08

Fingertip Force Control with Embedded Fiber Bragg Grating Sensors

• Dec 10 2007 : Paper accepted
• Feb 4 2008 : Final deadline (no later than midnight, US Pacific time)
  • IEEE ICRA final submissions instructions
  • PDF format (6 pages including figures, less than 2MB)
  • Video (Mpeg4, less than 5MB)

Reviews and responses to them:

... Testing could be done more throughly using tasks that
require more than one axes of force/torque data(and
presumably involve more than one FBG strain sensors). A
good choice would be 2D or 3D contour tracking. Both
experiments described in the paper require force control in
one dimension only, which is hardly the case for dextrous
manipulation tasks.

... In addition, the paper appears to lack some depth in sensor
data analysis.

MRC: We can't really do anything about the multi-axis wish but we can say something about steady-state force accuracy and transient response (how fast it converges after a disturbance).

... The result doesn't seem to be clear how accurately the
manipulator(hand) maintains the desired force. It would be
better to elaborate the results and include more results
for better comparison.

MRC: I am not sure to interpret this, again we can address the accuracy issue.

SCR: If this means that he want to see the force tracking error during the control as the control accuracy, we can add the force error plot, and say that 'the max force error [N]' or 'the max error/desired force[%]' is less than some value after transient response like the other papers on force control.

... I'd like to point out that Trinkle's and Zexiang
Li and Guanfeng Liu's work on the finger manipulation.
Especially Liu's work on hybrid control.

MRC: Let us look this up. Offhand, I am not sure what is relevant...

SCR: I think that what we can do is just to refer to Guanfeng Liu's research on the position/force hybrid control algorithm in the begining of section IV. But, as Mark said, their researches are not seemed to be related to our paper. Most of the papers are on the Path planning or the geometric approach to the control/modeling of constrained mechanical system like multi-fingered robot's grasping. I think, it's somewhat different from the main focus of this paper, which is to show not which control algorithm we used, but, how well the optical fiber sensor can be used for robotic application as a force sensor.

The attached as-submitted for review PDF:

• PhotonicRobot_ICRA08.pdf: submitted PDF
• PhotonicRobot_ICRA08.tex: tex files for Photonic Robot paper
• PhotonicRobotRef.bib: bib file for Photonic Robot paper
• Other files are on Chewie in ylpark/PhotonicRobot/Papers/IEEE_ICRA08

The final version:

-- MarkCutkosky - 10 Jan 2008