Prior Research
Getting inspired and understanding the challenges
My initial inspiration for this project came from Juan Gonzalez Gomez, who over the years appears to have developed several snake style robots. I was very interested in the different modes of locomotion, the identical train of modules makes for many different possibilities. I have worked with quadcopters and simple ground vehicles (i.e. tank and 3 wheel drive) but never anything like this. The playlist of Juan’s “Cube Revolution” robot to the ___ shows the different types of motion I am referring to. I particularly like the “moving like a wheel” idea, as this overcomes the greatest challenge of most snake robots: movement speed across flat surfaces, which is typically rather slow.
Next I found this design from John Hopkins University that uses drive wheels but still maintains some of the mechanical freedom. This definitely makes for a faster moving robot, but it may not be able to navigate extremely complex obstacles like ladders.
Hopping over to Matsuno Lab at Kyoto University in Japan, we have just that: a ladder climbing snake. The researchers have gone for a different type of gait, a sort of worm gear locomotion. I thought it was an interesting and creative divergence from the typical attempts to directly mimic real snakes. You can see more on their website here, if you can’t read Japanese you’ll need to translate most of it with your browser.
Finally, there is Carnegie Mellon who have a whole lab dedicated to this type of research, CMU Biorobotics. If you are at all interested in mobile robots, there research is well worth a look. I would say their snake robot is the most well developed, but also the most expensive. They have put tons of research into different gaits that can be achieved by their powerful, aluminum machined, and brushless motor driven, robot. They have presented their work at many different technology conventions and continue to lead the field. I find their gait research the most interesting and helpful and they have tons of video documentation for reference. But their design is far too expensive for me to take more than mild inspiration from.
