search

personalrobotics / ada

Software for ADA, the Assistive Dextrous Arm developed by the Personal Robotics Lab at CMU.
1 stars 9 forks source link

Cable management solution for Structure IO and uEye cameras #36

Open siddhss5 opened 9 years ago

siddhss5 commented 9 years ago

We really need a better solution than the rat's nest of cables we have now would precariously around the robot. What should we do?

shenlirobot commented 8 years ago

Sorry guys, I broke the camera cable USB extender today when I tested block sorting on ada... Do we have another cable to replace it? Or just solder it? I can buy one to replace it.

romesco commented 8 years ago

I accept partial responsibility too. I promise I have good soldering skills...

siddhss5 commented 8 years ago

@mikedh, could you please survey the damage and help us fix this?

mikedh commented 8 years ago

Sure, I'll take a look at it this afternoon. Fully solving the cable issue could be pretty involved but potentially could be done by Kinova as part of a joint project/grant. I wrote up a quick RFI document for how one might approach the issue: https://docs.google.com/document/d/18aaiOQ_LNFZ60GW1Co5qZraywzLCYIUkqVzeXL81HXc/edit?usp=sharing

Stefanos19 commented 8 years ago

It appears only the cheap USB extension cable broke, which is really cheap: http://www.amazon.com/gp/product/B00N9HRVH6?psc=1&redirect=true&ref_=oh_aui_detailpage_o03_s00 The structureIO cable looks fine.

ClintLiddick commented 8 years ago

We should have extra extension cables in the lab, I'll dig one out for now. USB passthrough would be great though. @mikedh, I think the SensorIO has a dedicated power/recharing cord as well (not powered through USB), any chance of routing that as well?

romesco commented 8 years ago

Yeah, there was a spare extension cable laying right next to ADA, so we are using that (and being more careful with letting it plan crazy twisting joint motions). The StructureIO does have a dedicated power cord and it would be great to have that in the proposed pass-through. As I mentioned at the last lab meeting, the point-cloud depth data vanishes if the Structure is left in a non-charging state for too long.