For who's interested, NASA has an open source toolkit for simulating these structures, using Bullet physics:

https://github.com/NASA-Tensegrity-Robotics-Toolkit/NTRTsim

It's fun to play with.

What happens when a cable gets snagged on a rock? Or even the constant abrasion of rubbing against things could snap the cable. Interesting design though, and shows how our vehicles with a solid frame and wheels aren't necessarily the best thing ever.

I am so excited about stuff like this.

http://softroboticstoolkit.com/ was the first thing that really perked my interest. It just seems like such an obvious yet under explored line of thinking.

The robot's reminiscent of sculptures by Kenneth Snelson http://www.kennethsnelson.net/sculpture/outdoor/ , one of which, called "Mozart I", is on the Stanford Campus, where one of the creators of the robot in the article did his undergraduate studies in Symbolic Systems. I wouldn't be surprised if the sculpture was an inspiration for the robot.

definitely unique. I imagine the inverse kinematics of that structure are a nightmare but you probably don't have to worry too much about positional feedback.

Brain is whirring with questions about how reliably the tension wires can be under use, and if you have spring constants in the system do they degrade over time or are they implemented in code with the tensioner system.

These structures remind me of proteins, which are very much tangled and have to resist lots of stresses as well. A quick search yields:

Is tensegrity a unifying concept of protein folds? http://www.sciencedirect.com/science/article/pii/S0014579302...

See also: https://en.wikipedia.org/wiki/Tensile_structure

Spineless robots?

http://freefall.purrsia.com/ff200/fv00165.htm