One of the primary purposes of unmanned systems is to keep humans out of danger by performing dangerous, dirty, and dull tasks. Fires represent one of the greatest dangers to sailors, working onboard U. S. Navy ships. The Office of Naval Research (ONR) has developed a robotic firefighter to work alongside humans to fight fires. The Shipboard Autonomous Firefighting Robot (SAFFiR), which is pronounced as “safer”, performs firefighting tasks aboard the Navy ships, keeping the sailors safe and providing enhanced situational awareness for human firefighters. This unmanned ground vehicle (UGV) is a humanoid robot, which measures 5 feet 10 inches and weighs 143 pounds (Gaudin, 2016).

    The 2016 article by Sharon Gaudin talks about this amazing robot, which will potentially be a great benefit for the United States Navy. It is not only designed for the firefighting applications, but also is capable of performing basic maintenance tasks, such as checking for corrosion and leaks. By performing everyday maintenance and inspections, the SAFFiR could free up sailors for more advanced technical jobs onboard of the ship. Figure 1 depicts the prototype of the SAFFiR.

Figure 1. The SAFFiR humanoid firefighter trials. Adopted from “Making sailors ’SAFFiR’ - Navy unveils firefighting robot prototype at Naval Tech EXPO,” by T. White, 2015. Copyright 2015 by U.S. Navy.

This UGV is designed to be capable to perform autonomous operations, however, initial robot design keeps the operator in the loop, allowing the human controller to monitor and override any action of the UGV. The main goal for the SAFFiR is to allow this robot to seamlessly work alongside its human counterparts on the Navy ships, responding to verbal commands and gestures, such as pointing and hand signals (Eshel, 2015).

To enable natural collaboration with a human “fire boss”, the robot will be equipped with multimodal interfaces that will enable the robot to track and focus its attention on the human team leader. Researchers are planning to further simplify the robot interaction by using natural language commands (White, 2015).

It is designed to endure high temperature environments, recognize fire hazards, and extinguish fires using a broad range of fire suppression tools. Its upper body will be capable of manipulating fire suppressors and throwing propelled extinguishing agent technology grenades. The SAFFiR is battery powered, which gives the robot about 30 minutes of firefighting mission time, after which time the battery needs to be recharged (McKinney, 2012). As we can see, the power system design still needs to be improved to allow for longer mission endurance, when necessary. Figure 2 represent some of the features of the SAFFiR.

Figure 2. The Naval Research Laboratory's Shipboard Autonomous Firefighting Robot (SAFFiR) is a humanoid-type robot being designed for shipboard firefighting. Adapted from “NRL designs robot for shipboard firefighting,” by D. McKinney, 2012. Copyright 2012 by U.S. Naval Research Laboratory.

This bipedal robot can walk, balance, and navigate even on the moving ships, it can cross over obstacles, manipulate fire hoses, and install fire shielding equipment. It features a lightweight central aluminum construction, which allows for efficient transfers of loads throughout the UGV’s body. It’s six- axes force/torque sensor allows for strong feedback while walking.  The advanced joint movements are enabled by titanium springs installed in the robot’s “legs” (McKinney, 2012).

The SAFFiR is designed to “see” through dense smoke with a help of advanced sensor suit, including infrared stereo vision, gas sensor, and a rotating laser for light detection and ranging (LIDAR) (Gaudin, 2016). So far, the SAFFiR is in its testing stage. The first trials will take place onboard a decommissioned U.S. Navy vessel, the USS Shadwell, docked in Mobile Bay, Alabama.

The researchers are working to constantly improve and enhance the SAFFiR. The latest development for the humanoid includes a motion-planning algorithms to allow the robot to skillfully perform a variety of autonomous tasks. The U.S. Navy awarded a $600,000 grant to the Worcester Polytechnic Institute for development of the advanced motion algorithms for this UGV to work in complicated scenarios. These algorithms will allow the robot to be able to move quickly in confined spaces when working onboard a ship or submarine. It must also be able to stay balanced while the ship is moving in rough seas. Researchers are planning to improve the SAFFiR with enhanced computing power, and increase its ability to solve complicated tasks, and better communication capabilities, and longer endurance.

The main goal for the development of the firefighting robot is to prevent tragedies like the one onboard the USS Miami in May of 2012. The nuclear submarine was damaged by an onboard fire, started by a shipyard worker, while in a dry dock at the Portsmouth Naval Shipyard in Kittery, Maine. Seven people were injured during fire. Because of the degree of the damage to the vessel, the Navy inactivated the ship. (Gaudin, 2016).

Although, humanoid-type robots may seem less stable than their wheeled counterparts, the SAFFiR is showing promising results for life-saving applications, while skillfully balancing on a moving ships with the help of its advanced motion algorithms and with the constant advancements in robotic technology, humanoid-type UGVs will eventually play an important part in our everyday lives.

References:

Eshel, T. (2015). Navy scientists experiment with firefighting robots. Retrieved from http://defense-update.com/20150205_saffir.html

Gaudin, S. (2016, February 9). Navy calls on researchers to create firefighting humanoid robot. Computerworld. Retrieved from http://www.computerworld.com/article/3031657/robotics/navy-calls-on-researchers-to-create-firefighting-humanoid-robot.html

McKinney, D. (2012). NRL designs robot for shipboard firefighting. Retrieved from http://www.nrl.navy.mil/media/news-releases/2012/nrl-designs-robot-for-shipboard-firefighting

White, T. (2015). Making sailors ’SAFFiR’ - Navy unveils firefighting robot prototype at Naval Tech EXPO. Retrieved from http://www.navy.mil/submit/display.asp?story_id=85459

Sense and avoid sensor selection

Unmanned aerial systems (UAS) integration into National Aerospace System (NAS) dictate establishing regulatory standards and equipment requirements to ensure UAS operational safety. Sense and avoid technology is one of the important issues pertaining to safety of UAS. Pilots of manned aircraft are responsible to see and avoid other traffic by relying on visual detection, air traffic control radar separation, and other available sensors installed on the aircraft. Unmanned aircraft pilots often have to rely on the UAS camera sensor picture with limited field of view, which is not sufficient for the sense-and avoid requirements for operation in the NAS. That’s why proper selection, testing, and certification of sense-and -avoid technology for UAS is important.

Electro optical/ infrared (EO/IR) sensor is a sense-avoid technology selected for this paper. EO/IR is a non-cooperative sensor, which means it does not rely on the technology carried by the intruder aircraft. EO/IR is a passive sensor. Due to this sensor’s size, weight and power (SWAP) considerations, it is suitable for use in the smaller size UAS (less than 55 pounds). It is capable to operate during day and night in all weather conditions. However, current EO/IR technology for SAA may have an increased rate of false alarms generated due to clutter in images and weather conditions such as fog or clouds.

EO and IR technologies are combined in a single compact sensor create a complete SAA system. The EO sensor takes images in the visible light spectrum with a charged coupled device camera. The infrared sensor works within infrared spectrum, creating images based on temperature differentiation.

EO/IR sensors have good performance in terms of detection of azimuth, elevation and traffic coverage. However, the drawbacks of this technology is that it is restricted range and has a limited field of view (FOV). It is important to mention, that a tradeoff exists between the FOV and detection range. For example, if the EO/IR sensor has a large FOV versus a small FOV to passively scan, the distance at which it can detect an object will decrease (Pearson, Moore, Ogdoc, & Choi, n.d). The particular EO/IR sensor which is suitable for smaller size UAS was developed by HoodTech Vision. Alticam AC-10 EO/IR sensor has small size and weighs only 5,700 grams and measures 25.4 cm in diameter. It is designed for both day and night and all weather operations. The sensor also features a laser pointer, a laser range finder, and a pan-over tilt. Gimbal that tilts 45 degrees up and 90 degrees down with 360 digress endless pan capability. Since power consumption is an important parameter to consider for small UAS, the AC- 10 was designed to use half the power of similar systems, freeing power for additional sensors and saving fuel for increased mission range. Power supply range is 24-32 VDC with 31 W continuous and 55 W peak consumption (HoodTech, n.d.).

It also has an increased FOV for better traffic detection: the IR sensor has a horizontal FOV of 1.7°- 22° and EO imager features 1.1°- 31.5° FOV. Since this particular sensor is enhanced with laser pointer and range finder, its application for as a sense-and-avoid sensor is greatly improved. The laser rangefinder operates in the 30 to 3000 meter range and it is eye-safe. The slew rate of the gimbal is 60° per second.

Figure 1. Alticam AC-10 EO/IR sensor. Adapted from “Alticam AC-10 specifications,” by HoodTech. (n.d.). Copyright by HoodTech.

 The concept of operations of a laser enhanced EO/IR system for sense-and avoid is as follows:

1. The EO/IR sensor detects potential intruders.

2. The laser subsystem confirms the azimuth and elevation angles of potential traffic and estimates range of the targets.

4. The gimbal with laser sensor slews to the target bearing angle detected by the EO/IR system.

5. After this information has been analyzed and, in case the intruder traffic presence is confirmed, the bearing angles from the EO/IR and the range from the laser ranger are fused to estimate the position and velocity of the intruder. In scenarios involving multiple intruders, the gimbals/scanners may be employed to slew the laser from one intruder to another (Ganguli, Avadhanam, Yadegar, Utt, & McCalmont, 2011).

Additional specifications details of AC-10 sensor are presented in Table 1.

Sensor	Wavelength	Horizontal FOV	Pixels	Video output	Zoom
IR imager	3-5 μm	1.7°- 22°	640 x 480	NTSC	Optical 13X; digital 2X
EO imager	0.4-0.9 μm	1.1°- 31.5°	1280 x 720	NTSC	Optical 30X, Digital 0.5-2X

Table 1. EO/IR sensor specifications. Adapted from Adapted from “Alticam AC-10 specifications,” by HoodTech. (n.d.). Copyright by HoodTech.

 The EO/OR sensor with laser range founder is a suitable solution for smaller UAS sense-and-avoid requirements. It is possible that in the future most UAS will require installation of cooperative technologies for traffic sense-and avoid. Such systems as transponder based Traffic Alert and Collison Avoidance system (TCAS) or GPS-based Automatic Dependent Surveillance Broadcast (ADS-B) or other similar technology will greatly enhance the sense-and-avoid capabilities for UAS. With the rapid technological advancements and miniaturization of sensor technology, weight, size, and power requirements of many sensors are decreasing while there technological capabilities are increasing. Installation of passive non-cooperative sensor such as the one presented in this paper in combination with cooperative active sensor such as TCAS will greatly enhance UAS sense and avoid capabilities.

References

Ganguli, A., Avadhanam, S., Yadegar, J., Utt, J., & McCalmont, J. (2011). Multiple intruder tracking using a laser enhanced EO/IR sense and avoid system. Retrieved from American Institute of Aeronautics and Astronautics: http://enu.kz/repository/2011/AIAA-2011-1478.pdf

HoodTech. (n.d.). Alticam AC-10 specifications. Retrieved from http://www.hoodtechvision.com/products/detail/alticam-ac-10

Pearson, J., Moore, Z., Ogdoc, J., & Choi, F. (n.d). Design of sense-and-avoid standards for RQ-7B Shadow under loss-link. Retrieved from http://seor.gmu.edu/projects/Capstone_2013/UAS_IEEE_Final_24Apr13.pdf