Highly Organized Multi-agent Enclosure (HOME)

This 1-year project will demonstrate multi-agent self-assembly behavior in swarms of pneumatic soft robots such that habitat construction can be autonomously performed using the mobile soft systems themselves. The swarm attaches to each other to create the habitat outer shell.\n \nObjective Statement:\nSoft, pneumatic habitat structures have long been in development and recently have had successful deployment on the ISS via work pioneered by Bigelow Aerospace. In addition, the soft robotics field of robotic systems research has demonstrated novel usability in a number of fields such as grasping, manipulation, locomotion, and human assistive bionics. \nCombining these technologies, and leveraging the in-space assembly of structure and autonomous systems developmental work at NASA LaRC, an interesting and novel approach to habitat construction and feasibility will be explored and demonstrated at the conclusion of this project.\n This project will focus on the design, programming, and implementation of 10+ small soft robotic systems, using swarm behavior algorithms, for the construction of a subscale habitat dome. It is important to note the the vehicles themselves will make up the habitat, demonstrating self-assembly in an autonomous manner.\n \nImpact:\nIt is not known whether such a novel approach has been considered in previous proposals, however, it stands that having an autonomous soft robotic entity coordinating self-assembly for habitat construction provides a unique and challenging solution for human exploration and long stay duration missions. This project will inject a new look at human-robot collaboration and habitat design, coupled with planning and execution in an autonomous fashion. It is expected to produce a stir in both industry and academia as we push the bounds in soft robotic control and multi-agent assembly.\n \nInnovation:\n This project will focus on developing:\n 1. Soft robotic entity design and construction. \n 2. Multi-agent self assembly.\n 3. Multi-agent coordination and autonomy.\n \n We feel the the soft robotic manipulator work, coupled with the autonomy systems work ongoing within LaRC enables this project in it's capability to successfully demonstrate a subscale habitat assembly of pneumatically actuated autonomous entities for the purpose of habitat construction.\n We will also use the AEON, Autonomous Entities Operations Network, developed at the Autonomy Incubator, that has proven to enable autonomous systems mission design and execution. We will also add additional self-assembly features to the framework as a result of this project.\n \nRisk:\nRisks being considered cover robot fabrication delays, multi-agent planning delays, and demonstration delays due to lack of resources. We plan to mitigate these factors by ensuring good design in the first 3 months for fabricating up to 20 robot systems. The pneumatic controllers will also need prior planning before purchase to ensure robust operations such that enough force can be applied to the material of the robots to ensure enough power and strength for subscale habitat construction. We will be using a rapid agile planning and manufacturing approach to ensure robustness to on the fly requirement changes to design and operations. Resource planning in an agile methodology will entail daily 10 minute SCRUMS to immediately deal with resource and task issues.\n \nPayoff:\nIf successful, we will prove the feasibility of using 10+ soft robotic systems to build, via using themselves as building material, a stable structure that could be used to shield underlying structures of either other soft systems framing or more traditional truss structure frames.\n \nWe are not currently aware of anyone using soft robotic systems for this type of construction project.\n \nTimeline:\n-- 3 months - Oct-Dec 2018 - Design and purchase of soft robotic systems, subscale habitat, demonstration design. Software begins\n-- 3 months - Jan-Mar 2019 - Fabrication of hardware component. Software controls and multi-agent planning testing \n-- 3 months - Apr-Jun 2019 - Testing single systems. Integrate multi-agent system construction testing.\n-- 3 months - Jul-Sept. 2019 - Prep for final demonstration and report out. Demo and papers.\n \nWe expect 2+ publications on the design and execution of this project, as well as delivering the following:\n 1. A hardware design for pneumatically controlled robot capable of linking with other similarly designed robots.\n 2. A software system capable of demonstrating self-assembly of a given subscale habitat design.\n 3. Demonstration of mutli-agent habitat self-assembly.\n 4. ~ 2 publications on the research.\n 5. Self-assembly algorithms for autonomous systems.\n 6. Novel in-space assembly approach proof of concept.

Multi-agent systems, or systems consisting of multiple autonomous entities, offer novel approaches to assistive human exploration of hazardous and remote locations both on and off planet. This project aims to develop and demonstrate a unique approach to habitat construction that provides future explorers the ability to stand up a habitation site prior to human occupation. The key technologies lie in the use of multiple soft-robotic entities that exhibit swarm behavior such that habitat construction can be initiated and completed without the oversight of a human operator. Leveraging deformable robotics technologies, soft robotics has the potential to enable habitat shielding from harmful cosmic radiation due to the ability to replace compressed air with water as a actuation fluid that also can be used as a radiation shield.\n\nBackground Information:\nMulti-agent self-assembly is a hard problem. Not only must researchers consider the hardware design components for linkable systems, researchers must also consider autonomous mission management and fault tolerance and recovery. The systems integration tasks is also a hard problem, where as design influences operation which in turn influences design as the operating environment is dynamic and stochastic.\nIn additional, there is no clear indication that self-assembling robotics systems targeting habitat construction has been seriously considered in the field of human exploration and habitation, nor in other paradigms. This presents NASA Langley with a unique opportunity to research and development a truly novel approach to off-planet habitation and sustainability, using mutli-agent self-assembling robotic platforms. Though habitat research using expandable/inflatable systems has received much attention in recent years, pointing towards the recent ISS demonstration on Bigelow Aerospace's expandable habitat. Leveraging the already existing habitat and in-space assembly work performed here at NASA Langley, this project is poised to collaborating with habitat designers, developing a unique approach to the self assembly multi agent problem. It is not known whether such a novel approach has been considered in previous proposals, however, it stands that having an autonomous soft robotic entity coordinating self-assembly for habitat construction provides a unique and challenging solution for human exploration and long stay duration missions. This project will inject a new look at human-robot collaboration and habitat design, coupled with planning and execution in an autonomous fashion. It is expected to produce a stir in both industry and academia as we push the bounds in soft robotic control and multi-agent assembly.

This is early/mid-stage (TRL 2) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.