SAFFiR, the US Navy’s prototype firefighting robot gets baptism of fire

SAFFiR, the US Navy’s prototyp...
SAFFiR in a rain suit putting out a shipboard fire (Photo: US Navy)
SAFFiR in a rain suit putting out a shipboard fire (Photo: US Navy)
View 2 Images
SAFFiR in a rain suit putting out a shipboard fire (Photo: US Navy)
SAFFiR in a rain suit putting out a shipboard fire (Photo: US Navy)
SAFFiR is designed to operate in tight spaces below decks on ships (Photo: US navy)
SAFFiR is designed to operate in tight spaces below decks on ships (Photo: US navy)

If there's one job that a person would probably prefer to lose to a robot, it would be fighting fires aboard ships. To help make such a vision a reality, the US Navy and the Office of Naval Research (ONR) released details of demonstration exercises conducted by their Shipboard Autonomous Firefighting Robot (SAFFiR) aboard the fire training ship USS Shadwell last November.

Fire at sea is every sailor's worst nightmare, and that's doubly so for those aboard warships. On a civilian vessel, a fire means rescuing survivors and possibly taking to the lifeboats, but fighting fire on a naval ship means stopping it from spreading to ammunition stores and keeping the ship capable of fighting no matter how intense or widespread the blaze. For this reason, the US Navy is particularly keen on the idea of developing robots to take over the task of facing fires below decks.

The latest product of this effort is SAFFiR, a bipedal humanoid robot standing 5 ft 10 in (178 cm) tall and weighing 143 lb (64.8 kg). Based on the CHARLI-L1 robot created at Virginia Tech, it boasts a "super-human" range of motion for operating in tight, complex spaces and moving over uneven, obstacle-strewn floors.

SAFFiR is designed to operate in tight spaces below decks on ships (Photo: US navy)
SAFFiR is designed to operate in tight spaces below decks on ships (Photo: US navy)

"Balancing on any type of terrain that’s unstable – especially for bipedal robots – is very difficult," says Brian Lattimer, associate professor for mechanical engineering at Virginia Tech. "Whole-body momentum control allows for the robot to optimize the locations of all of its joints so that it maintains its center of mass on uncertain and unstable surfaces."

Sponsored by ONR, SAFFiR's purpose is firefighting and damage control. It uses thermal imaging to locate and identify fires and LIDAR and other sensors to map areas and see through smoke. Eventually, it will be able to walk and handle fire hoses on its own, but at present it's restricted to remote presence control. However, even when it gains such capabilities, it will take its instructions from sailors and "fire bosses" working remotely.

SAFFiR's capabilities were made public on February 4 at the Naval Future Force Science & Technology EXPO. The presentation described the SAFFiR trials that were carried out on November 3 to 5 aboard USS Shadwell, a decommissioned Navy vessel. For the exercise, the robot was clad in a rain suit to protect its electronics from getting wet as it put out a fire using a small hose.

"We set out to build and demonstrate a humanoid capable of mobility aboard a ship, manipulating doors and fire hoses, and equipped with sensors to see and navigate through smoke," says Dr. Thomas McKenna, ONR program manager for human-robot interaction and cognitive neuroscience. "The long-term goal is to keep sailors from the danger of direct exposure to fire."

According to ONR, the next step is a more advanced design with greater autonomy, better intelligence, communications, speed, and battery life. In addition, the team sees SAFFiR as having applications in ship's maintenance by carrying out routine inspections and freeing up sailors for more challenging tasks.

The video below shows SAFFiR in action.

Source: ONR

Shipboard Autonomous Firefighting Robot - SAFFiR

Bob Flint
Many obstacles to overcome speed, power, control.
Since most fires require three things, fuel, ignition, and air, I would suggest looking also at ways to self suppress certain fire potential areas, & equipment with built in robotics or autonomous systems. Perhaps combine a sprinkler system with extendable arms to increase aim, or oxygen suppression with environmental controls.
Sorry, but very unimpressive in the video. Maybe they did not show its capabilities. The video showed me a slow moving robot having difficulty maneuvering the hallway at even a normal walking pace - much too slow to address fire control. Autonomy is lso not a word that seems applicable when it is run by someone on a computer. It looks to be more of a large puppet.
It needed an overhead track to which it was attached by a cable. How many ships have these overhead tracks installed currently?
Its accomplishments in the video were akin to what could be done with a modified remote controlled toy car large enough to hold/handle the water pressure exiting the hose. Well - OK you would need to add the cable (thereby hindering loctmotion on the car), and camera so the people at the computer could see what was going on.
I sincerely hope the video was just a bad representation of what this thing is really capable of. otherwise it is just another way government has spent our tax dollars and not really accomplished much more than could be done by much less inexpensive/efficient techniques.
I am all for autonomous robots - but am hoping this is not the typical example of what our government can do when you compare it to something like Japan's Asimo.
Thank you Lbrewer42 and I've said it time and time again, the problem is the difficulty/complexity they added when they gave it a human lef design for no good reason.
All of the money, complexity, and limitations of this are pretty much completely the fault of the design decision to give it legs instead of one of the many other methods available for more efficient transportation.
Even after making all of those compromises the thing looks like it needs to be tethered to 1. an overhead support. 2. a power source, and 3. a water source and still barely moves even on a perfectly flat surface. What the hell is the point of making all those compromises when it still doesn't even climb stairs?
Biped idiots are holding back the robotics industry and wasting billions of dollars in R&D, much of it tax payer funded. If they can't come up with a better robotic transportation platform than biped legs they need to be thrown off the project. I could come up with several other ways to climb stairs.