-- MarkCutkosky - 18 Jul 2003

7-18-03 8:30am PST telecon

('pardon the formatting for pasting into the ClimbingRobot web)

Given that only MarkCutkosky, KellarAutumn and RonFearing? were present, we focused on dry adhesion, toes, lamellae and conformal substrate concepts.

Topics:

  • Force sensing
  • Dry adhesive patches
  • Lamellae and conformal substrates
  • Action items

1. Force sensing Interested in deployment, engagement, detatachment forces for dry adhesive pads, claws, etc.

Kellar's group is using Kistler piezoelectric load cells. Small single-axis cells with sensitivies of milliNewton scale, larger 3-axis load cell with range from 1 gram to 100s of Newtons. Excellent accuracy, range but the usual piezo issues of high impedance, drift, for static measurements. Should compare notes (MotohideHatanaka and WillP) with Kellar.

ATI sensors were considered, but not sensitive enough.

What about building something that suits our purposes? Mark mentioned that Will had talked to him about this -- it is a bunch of work but could be worthwhile if amortized over the several groups that will be needing it. (Action item: follow up with Kellar, Ron, Will.)

2. Dry adhesive patches

Ron - working on patches. Some early samples. need better conformal substrate. superglue may be stiffening and making less smooth too much.

looking at gel packs too but they were TOO compliant. adhesive is stronger than gel pack so it tore.

sheets 50 micron polyurethane sheet with hairs sticking out on top up to 60 microns long. more like 30-40 typically.

3. Lamellae and conformal substrates

Venetian blind idea - the lamellae - adhsive with a leaf spring. underside faces rear of toe. leading edge faces nail. push toe down - springs provide compliance on rougher surface. this is atop the liquid bladder. tension : liquid bladder not enough in this case. maybe lamellae + bladder needed for tension.

Russell's hypothesis - gecko anatomy guy at Calgary. hypothesis is lam are for drag; liquid is for preload. but if is mechanisms for compliance in both tension and compression then this hpothesis is not quite right.

OK - what about lateral loads? How best to uniformly distribute? This might really be what the 'ventian blinds' are about.

trailing edge pressure and shear traction distribution.

one kind of gecko has the lam rotated 90 deg. does this help decouple propulsion from adhesion.

cavity with liquid bladder membrane silicone gel; distribute forces. Sinus blood. hydrostatic? preload adhesive - if develop pressure one area - const vlume better than ear plug idea - more uniform redsitrbution.

Would like to have some simple foot prototypes (see action items below) to test with force sensor -- measure best initial preload force, adhesion force, shear force, detachment force.

4. Action items

  • Send samples of materials we are investigating for SDM etc. to Kellar
  • Arrange visit in next few weeks to Lewis & Clark -- Will? who else?
  • Planning meeting to create concept designs for a couple of feet incorporating conformal substrate
    • For example: silicone gel in a bag or conformal foam + dry adhesive patch from Fearing + claws at periphery. When pad presses against surface, the claws are deployed. (This is similar to a concept we have discussed before and similar to what we see in ants.)
    • Alternatively, try to build (SDM) some toes with a few lamellae.
    • Alternatively, array of micro pins on compliant surface. (Micro pins per Kellar's pointer -- need to obtain some.)

 
This site is powered by the TWiki collaboration platformCopyright &© by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback