EigenBot: Intelligent Modular Robotics
Abstract
EigenBot: Intelligent Modular RoboticsSOW: To enable field technicians to rapidly assemble customized robotic solutions on-the-fl"y from a ""bag of modular components , that include actuators, sensors, and passive components, and then seamlessly deploy the config"ured systems with intuitive user interfaces. Modularity represents the degree to which a system’s components (e.g., actuators, sensors, etc.) may be separated and combined to form new configurations that address different task objectives. In the context of robotic systems, many researchers have promised to develop modular systems architectures. However, few, if any, have successfully delivered hardened and truly flexible solutions. The proposed work thus develops the next generation of modular robotic systems that makes possible the seamless composition of “eigenmodules,” i.e., atomic modular components, into task specific configurations. An algorithmic approach will be developed that automates the process for composing new sets of these eigen-modules. Once configured, the proposed work will endow the systems with the high-level intelligence necessary to enable non-specialists to intuitively control them to meet the needs of their individual tasks at hand. Furthermore, the modular architecture will also make it possible for a technician to simply swap out a malfunctioning module in the field such that the composite system is immediately restored to its nominal functionality, all without the need to change a single line of code.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Apr 24, 2019
- Source ID
- N000141912224
Entities
People
- Matthews Travers
Organizations
- Massachusetts Institute of Technology
- Office of Naval Research
- United States Navy