3D Multispectral Vision System for Robotic Fire-Fighting Systems

Period of Performance: 06/25/2012 - 04/26/2013


Phase 1 STTR

Recipient Firm

Lynntech, Inc.
Firm POC
Principal Investigator

Research Institution

Carnegie Mellon University
5000 Forbes Avenue
Pittsburgh, PA 15213
Institution POC


Advancements in humanoid robotics led the U.S. Navy to recognize the potential for humanoid robotic fire-fighting systems such as the Shipboard Autonomous Fire Fighting Robot (SAFFiR) to address the dangerous task of onboard fire suppression and reduce costly time-to-response. A vision subsystem capable of imaging through low-visibility fire smoke and generating a 3D model of the local environment is an essential enabling technology for such a robotic firefighter. Typical stereo-pair camera techniques might be applied but visible imagery is obfuscated by smoke and infrared imagery poorly conveys edge detail and clear geometry definition in extreme thermal environments. LADAR is inhibited by scattering from smoke particles. Lynntech s 3D Multispectral Vision System for robotic fire-fighting systems fuses smoke-penetrating LiDAR, thermal infrared, and visible imagery data to generate a geometrically-accurate model of the environment with surface-temperature data in real time. This unique combination of sensing modalities addresses the challenges imposed by a shipboard fire scenario and provides unprecedented situational awareness for operation in such environments.