Titan 6: Two-Meter High Hexpod Ropbot
Abstract
The proposed work will build the world’s largest and most capable hexapod robot, called Titan-6. Our goal is to endow Titan-6 with the capability to step over large obstacles while maintaining the ability to move relatively quickly. A distinguishing feature is Titan-6 s modularity. With its modular architecture, this effort will produce a family of large legged robots. These robots will be built for both field use and upwardcompatible development in mind. To build a two-meter high robot, the proposed work will advance the state of the art in leg design and capability, creating a new system that breaks conventional wisdom on how locomotive technology scales in size. To this end, the Titan-6 design will include novel ways of using state-of-the-art, yet well-proven actuation technology. The idea here is that we want to rely on robust technologies so that we can deploy quickly, reliably, and cost effectively. The modular thrust of the proposed design will develop the hardware and software architecture to enable a field technician to quickly build (and re-build) customized large-scale hexapod systems that work intuitively. In particular, the software modularity to be designed allows us to easily mix and match system modules for ease of development, testing new features, and maintenance. Choset s group has a long history of developing modular architectures for mobile robot systems not limited to snake-like robots, hexapods (albeit small ones), tracked vehicles, etc. Finally, the Titan-6 must be agile because it will be sent into terrains where it needs to move dynamically for both its own safety as well as to overcome locomotive challenges. Standard kinematic planners for locomoting systems will thus not suffice. Even if all the relevant environmental information were available to the robot, careful kinematic planning (e.g., planning footstep locations) is doomed to be slow and brittle. We will leverage our existing planning work to overcome two key challenges: how to manage many degrees of freedom and how to handle unmodeled challenges.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Sep 19, 2018
- Source ID
- N000141812855
Entities
People
- Howard Choset
Organizations
- Massachusetts Institute of Technology
- Office of Naval Research
- United States Navy