3 ms·

I used to work on this project - surprised to see it on the front of HN. Feel free to ask me if you have any questions. This video is relatively old. Here are

I used to work on this project - surprised to see it on the front of HN. Feel free to ask me if you have any questions. This video is relatively old. Here are some more recent videos of where the technology is now: Overview of the most recent robot design: https://www.youtube.com/watch?t=64&v=te4M-b69fVs https://www.youtube.com/watch?t=64&v=te4M-b69fVs Video of the "Snake Monster" made with multiple snake robots as legs: https://www.youtube.com/watch?v=AMVO6rI5mL4 https://www.youtube.com/watch?v=AMVO6rI5mL4 Main webpage about the project: http://www.snakerobot.com http://www.snakerobot.com

deutronium 12y ago
I really love how its movement is so fluid, especially in the video you just linked to. Is there any reason you couldn't power it via battery, instead of tethering it?
- cornellwright 12y ago
  In the past we did some that were battery powered, and it's doable. As a research project, a lot of the research is on the motion and mechanism design. Given limited resources (both budget and people), battery power was removed from the scope of the project until there's a strong reason to do it. It's more of an engineering challenge than actual science if that makes sense.
  - deutronium 12y ago
    Cheers, yeah I imagine for lab work it makes more sense to have it tethered. I was just thinking how cool it'd be to let it climb up the whole tree by itself.
acadien 12y ago
How much design and algorithm design did you do in simulation as opposed to using the physical robot itself?
- cornellwright 12y ago
  It's a mix of both. In terms of simulating the shape the robot makes and visualizing its motion, that is basically all done in software the first time. Simulating interactions with the environment, however, is quite difficult. While we did have a simulator, most work was done on the robot itself.

3 ms·

I used to work on this project - surprised to see it on the front of HN. Feel free to ask me if you have any questions. This video is relatively old. Here are

by cornellwright 12y ago

deutronium 12y ago
I really love how its movement is so fluid, especially in the video you just linked to. Is there any reason you couldn't power it via battery, instead of tethering it?
- cornellwright 12y ago
  In the past we did some that were battery powered, and it's doable. As a research project, a lot of the research is on the motion and mechanism design. Given limited resources (both budget and people), battery power was removed from the scope of the project until there's a strong reason to do it. It's more of an engineering challenge than actual science if that makes sense.
  - deutronium 12y ago
    Cheers, yeah I imagine for lab work it makes more sense to have it tethered. I was just thinking how cool it'd be to let it climb up the whole tree by itself.
acadien 12y ago
How much design and algorithm design did you do in simulation as opposed to using the physical robot itself?
- cornellwright 12y ago
  It's a mix of both. In terms of simulating the shape the robot makes and visualizing its motion, that is basically all done in software the first time. Simulating interactions with the environment, however, is quite difficult. While we did have a simulator, most work was done on the robot itself.