Robotic Assembly of Digital Structures using Deployable Elements

The work I’ve performed over the past four years has explored how a building-block based engineering approach can create robust, scalable systems that are easier to develop and iterate on than their traditional counterparts. I’ve explored the impacts of this approach to the design of nano-satellite platforms, distributed sensor systems, robotic inspection platforms, and human exploration vehicles. Previous work in building-block based systems have either taken the form of modular robotics or digital cellular solids. In modular robotics, a robot is composed of many simple modules with specific roles (such as motor, power, communication, etc) that, when combined, form a robot that can perform complex tasks. In digital cellular solids, many identical mechanical components are joined together to form ultra-high performance materials with tunable properties. One of the first projects I worked on was a modular satellite platform called OuroboroSat, where each module could locally store power and transfer it to its neighbors. Since the electric power system (EPS) is the most failure-prone part of the satellite, I saw discretizing the EPS as a way to increase the robustness of the system. I created a test article and demonstrated its ability to transfer power and route around failed modules, and, with Greenfield Trinh at NASA Ames, tested and launched the article on a sounding rocket. The next project I did was designing a robot that could traverse the interior of the CubOct lattice. Most robots that traverse a truss structure are designed to climb something like the Eiffel tower, with many different strut lengths, cross-sections and angles between them. Working in collaboration with Benjamin Jenett, we explored how we can use the periodicity of the CubOct structure to simplify the complexity of the robot, while also increasing the accuracy of the robotic positioning. We coined a new class of robot from this approach that we call “Relative Robots”, and created a demonstration robot that could climb in the lattice. We tested this robot, MOJO, on a zero-gravity flight twice last year, and demonstrated the ability to climb in any direction through the lattice. The previous two projects taught me how to design circuits and program microcontrollers, something I had only basic experience working in when I started my PhD. The next project I worked on was the culmination of these skills, and involved the development and implementation of a distributed sensing platform that could augment a digital cellular solids lattice. Usually, sensor networks are sparse systems with wireless connections between them and (relatively) undefined physical locations. This work examined how the regularity of the digital cellular solids lattice could be used to simplify the communications network topology and hardware (by using a fixed topology wired network rather than a variable wireless) and how the lattice could be used to physically position the nodes with an accuracy that wouldn’t be possible if the nodes were free to move. Working with Nick Cramer, we implemented and tested this system in the 14x22 wind tunnel at NASA Langley Research Center, on an experimental aircraft designed at NASA Ames, and demonstrated a 20-node network collecting pressure and orientation data at 100 Hz across 74 pressure ports and 3 IMUs. In total we collected 2 GB of data, which we are currently analyzing, and plan on expanding this work to allow the sensor nodes to distribute the complex calculations on-board. Finally, I have studied digital cellular solids extensively, and tried to design new structures and materials that are amenable to robotic construction. This design work culminated in a NIAC submission (and companion paper) written with Benjamin Jenett that explored how we might one day create a ship capable of ferrying 10,000 people from Earth to Mars every synodic cycle. I also wrote and submitted a paper to Aeroconf 2017 that described the construction of the large pressure vessels that would compose the vessel.