By Denise M. Stefula
Each planet and moon in our solar system is unique, although many have characteristics in common that can be hazardous for exploration—unstable crevices, ravines and steep cliff faces, broken jagged ice, and rocky, boulder-strewn terrains—all of which are inaccessible to the currently used wheeled rovers. These locations are largely excluded from exploration planning because one mishap can end a mission.
A few years ago, a team at NASA’s Ames Research Center, supported by the NASA Innovative Advanced Concepts (NIAC) Program, established the value and feasibility of a new class of structurally compliant “tensegrity” robots. These robots are so resilient that theoretically they do not need air bags to land on other planets and can safely accommodate the risks inherent in terrains previously left out of mission planning, opening up new exploration strategies and enabling new mission profiles.
SUPERball Bot, a seedling task through 2017 under the Game Changing Development Program’s Human Exploration Telerobotics 2 project, is investigating sensing and control algorithms for these novel tensegrity robots, and building a new prototype to demonstrate the mission concept.
Tensegrity is a structural principle—based on isolated components in compression inside a net of continuous tension—that offers innovative ways of thinking about how parts and wholes interact. The compression components (bars or struts) do not touch one another and the tension members (cables or tendons) delineate the system spatially.
This characteristic property produces exceptionally robust structures for a given mass and for the cross section of components. The structure’s integrity is derived through balanced tension, or the forces at play among the cables.
The SUPERball task is establishing fundamental engineering principles for this new class of robot, paving the way for researchers to design and build tensegrity robots with increasing complexity and sophistication.
“Going forward, we will be building, testing, and then building another iteration of a single SUPERball 2.0 rod over the next several months, and then building out the full robot (i.e., six identical rods) in the spring of 2017,” says Vytas SunSpiral, task manager for SUPERball at Ames. “Next summer we will start testing locomotion controls and perform increasingly higher drop tests until we demonstrate our intended target for this prototype of landing at 7.5 m/s by rolling off the roof of a building and then proceeding to a mock science target.”
Ultimately, the goal is to demonstrate a full suite of mission relevant capabilities in one integrated terrestrial prototype. Features include the ability to deploy from a packed launch configuration, land at high speeds, move to targets of high scientific interest, and place instruments at desired science sampling locations.
SUPERball is only a single example of what is possible with this new class of robot. Others include lightweight, robust-legged robots with the terrain-access capability of a mountain goat—opening up whole new possibilities for accessing high priority science targets on the planets and moons of our solar system.