Damage Control for the 21st Century, or the "DC-21" project, is developing advanced technologies for shipboard damage control. The project includes the development of advanced autonomous systems to assist in discovery, control, and damage control of incipient fires. The research focuses specifically on the human-robot interaction technology that will allow a Navy firefighter to interact peer-to-peer, shoulder-to-shoulder with a humanoid robotic firefighter.
Computational cognitive models are used to create reasoning mechanisms for the robot based on required human cognitive skills. In addition, the project investigates perception, both aural and visual, and how context helps to disambiguate perceptions. Human-robot interaction is achieved through robust parsing of human speech, integrated with human gesture to disambiguate utterances.
This project is the collaborative effort of NRL Codes 5510, 5512 and 5515, along with a research association with Code 6186, Shipboard Fire Scaling. Alan Schultz is the Principal Investigator with Dr. Greg Trafton and Dr. Dennis Perzanowski as co-Principal Investigators. Research is funded by the Office of Naval Research, Warfighter Performance Department. under work request WX20805.
- Damage Control for the 21st Century: Shoulder-to-shoulder firefighting (Publication Release: 12-1231-1095)
Even in peacetime, fires represent one of the greatest risks to the U.S. Naval Fleet. To this end, the Naval Research Laboratory (NRL), with support from the Office of Naval Research, is conducting research and developing new technologies to enable shoulder-to-shoulder robotic damage control teammates. The robot in this video is a research platform for testing software for cognitive robotics and human-robot interactions. The knowledge gained from this research will be applied to firefighting, and inspection and maintenance robots used on ships. Through a combination of speech and visual recognition, the robot is able to identify trusted individuals, in this case, the human fire-fighting teammate. The human is able to provide situational information to the robot by voice and gestural commands. Here, the human partner is telling Octavia the general location of the fire before she enters the compartment. Using two infrared cameras, Octavia is able to localize the fire, allowing her to target it with the compressed air/water backpack. Ongoing work is focused on improving the naturalness of the interactions so that the human partners can interact with the robot as if it were another human teammate. Additional work is focused on recognizing and characterizing the type and behavior of the fire so that proper extinguishing techniques can be used.
Alan C. Schultz
Director, Navy Center for Applied Research in Artificial Intelligence
Naval Research Laboratory
Washington DC 20375
w5510 "at" aic.nrl.navy.mil