UAS payloads get smarter sensors, enhanced imagery in smaller packagesStory
July 31, 2013
Requirements for smart sensors that can see further and produce high-quality imagery in small, low-weight packages are driving Unmanned Aerial System (UAS) payloads development. Designers of these systems also face challenges such as balancing reduced Size, Weight, and Power (SWaP) requirements while increasing performance, uncertain Department of Defense funding priorities, and data link bandwidth limitations.
The advantage U.S. military forces have gained from the deployment of Unmanned Aerial Systems (UASs) is unquestioned. From the Intelligence, Surveillance, and Reconnaissance (ISR) capability of Global Hawks to the lethality delivered by armed Predator B aircraft, these platforms have been a force multiplier for the U.S. in Iraq, Afghanistan, and across the globe in the hunt for terrorists.
While these wars are winding down, the demand for the drones remains steady, but their missions are changing. In the near future – in absence of the breakout of another war – they will be repurposed for a variety of ISR, Signals Intelligence (SIGINT), and Communications Intelligence (COMINT) missions, but less for delivering weapons. However, what these missions will be and how these payload designs will be funded remains uncertain because of sequestration and other cuts in the DoD budget. What is certain is that these payload designers will produce smarter, better-performing sensors to get around low air-to-ground data links. They will also design around reduced Size, Weight, and Power (SWaP) requirements in systems they mostly develop themselves as the DoD reduces technology development spending.
“There is a drive toward flying before you buy,” says Christy Doyle, VP of Business Development at Mercury Systems in Chelmsford, MA. “In the past there was a significant government investment in payload development, but today the industry must invest before the government purchases systems. The government is also involved with flight demonstration to qualify the payload for each platform. It is moving toward a commercial business model requiring that contractors stay in lockstep with customers as we develop a system. There is a significant demand for multimission ISR payloads, which can be used on multiple platforms. In this uncertain budget environment, DoD customers are evaluating priority missions, requiring contractors to stay close to the evaluation process.”
“Requirements for Predator payloads are trending toward higher-resolution optical and infrared cameras for narrow field of view reconnaissance,” says Chris Pehrson, Director of Strategic Development at General Atomics in San Diego. “Radars and other sensors are used for observing large-area operations, covering tens or even hundreds of square kilometers. This will be especially relevant in the Pacific maritime domain for piracy and counterterrorism applications. These missions require not just the narrow field of view for detailed observation, but also wide-area surveillance for persistent situational awareness. In the maritime domain, for example, our Lynx radar is able to detect small craft and semisubmersibles over a large area and cross-cue to a high-resolution camera for close-in observation. This can help determine if there’s illicit trafficking or potential piracy activity.
“Broad area surveillance is also accomplished with sensors such as Wide Area Motion Imagery (WAMI), which can provide persistent observation of several city blocks or other large geographic areas,” he continues. “Multispectral and hyper-spectral sensors, as well as Signals Intelligence (SIGINT) sensors, are capable of observing large areas. These sensors can search for unique signatures or detect radio and other emissions of potential threats or adversaries. Many of them are modular and pod-mounted so they can be installed on the MQ-9 Reaper or the new Predator C Avenger based on mission requirements.
GA-ASI also is developing LIDAR sensors, an active imaging system that uses a laser to provide full field of view targeting, geo-rectification for accurate GPS coordinates, and foliage penetration capability. This sensor can be integrated with existing electro-optical/infrared sensors to greatly enhance their performance. When LIDAR is used for foliage penetration, any geometric shape will jump out at you.”
“The military sensor market is driving us toward high-definition and laser designation, but what everybody wants is to see further with less weight and in more compact systems – especially in the UAS and helicopter communities,” says David Strong, VP of Marketing for FLIR Government Systems in Wilsonville, OR.
(Click graphic to zoom by 1.9x)
Smaller payloads, more performance
“There has been a dichotomy regarding payload gimbal size,” Strong says. “If you want to see for long distances, you need a gimbal that is 16 or 15 inches and weighs about 100 pounds. The trade-off has been that you can’t carry that much load. Standard operating procedure is to go with smaller gimbal weights of 40 or 50 pounds. That has been the choice you had to make. It’s just a question of physics about how much optics you can pack into a system.
“So what we did is turn that trade-off on its head and changed the geometry in how we put together the payload – because we want to have the performance of a big gimbal but at half the weight or better in a 10-inch gimbal,” he explains. “It was a question of geometry, but you have to redesign everything that goes inside. We can do that because we make all the components. The result was the Star SAFIRE HDc – the C is for compact (Figure 1). It has the performance of a big gimbal at half the weight. It is a 15-inch diameter gimbal while the typical large gimbal has a height of between 18 and 20 or more inches. We reduced it to less than 14 inches in height. This cut the weight in half, but the package still has the same number of multiple sensors, long focal lengths, big apertures, etc.”
Figure 1: The Star SAFIRE HDc from FLIR has Short Wave Infrared (SWIR) technology and continuous zoom lenses for thermal, color, and low-light viewing.
(Click graphic to zoom)
Managing SWaP, costs through standards, FPGAs
Not every small system requires custom designs and many integrators will turn toward standards such as VPX for their payload designs. Larger UAS platforms such as the Predator will use 3U and 6U VPX systems, but smaller UASs do not have enough room and that’s where Small Form Factor (SFF) designs fit in, says Ray Alderman, Executive Director of the VITA Standards Organization in Fountain Hills, AZ.
“There are always SWaP challenges in UAS payloads. Many use custom SFF designs to solve them, yet this can be costly not only up front, but over the lifetime of the product,” says Bill Ripley, Director, Business Development at Themis Computer in Fremont, CA. “Leveraging standards such as VITA 74 SFF enables integrators to procure designs often priced 50 percent less than similar 6U products or custom designs, while maintaining military environmental specifications. We are seeing an increased desire from industry to stay with standards as opposed to custom form factors. Some custom form factors can come in with lower initial costs and initial prices, but for cost savings over the life of a program, standards-based systems are still the best choices. For conventional-sized UAS applications, Themis offers the NanoATR VITA 74 product. For larger applications, we offer 3U VPX systems.
“When we first started looking at VITA 74, the obvious market niche was UASs, and at about the same time there was a big push recapitalizing ground vehicles and building new advanced tactical ground vehicles,” Ripley continues. “This heated up because the numbers associated with them were large. For a while, ground vehicles enabled development in the UAS market more than in their own sector. A lot of the development cost was spent on the hardware used in UAS platforms. The UAS market had a need but didn’t have the numbers of assets at that time. Now the ground vehicle market is quieting down when the UAS market is potentially heating up.”
Mercury Systems engineers leveraged FPGA technology for the signal processing functions and reduced the electronic footprint in their smaller payload designs. “Our COMINT payload is an FPGA-based system with an antenna array design and electronic system self-contained in a small pod,” Doyle says (Figure 2). “This small pod was designed specifically to facilitate field retrofits, so it can be rapidly mounted under the wing or fully integrated into the fuselage.”
Figure 2: A Communications Intelligence (COMINT) payload from Mercury Systems can be rapidly mounted under the wing or fully integrated into the fuselage.
(Click graphic to zoom by 1.8x)
“The use of FPGAs enabled us to provide tremendous signal processing power in the same weight and DC power footprint,” says Chris Michalski, Technical Director, ISR Systems at Mercury Systems. “In previous systems most of the processing was performed in general purpose CPUs. There were not a lot of FPGAs involved in front end processing. Today we offer systems with ruggedized embedded boards that leverage FPGAs. This enables us to offload processing onto the FPGAs at a much lower level of DC power consumption than you would get by running a standard PowerPC device. FPGAs such as the Xilinx Virtex-6 have gotten quite good not only in terms of performance but also in lowering power consumption because the feature sizes are small.”
(Click graphic to zoom by 1.9x)
Data link bandwidth problems
Another reason designers are packing so much signal processing horsepower next to the sensor is to get around the bandwidth limitations of the air-to-ground data links. “They have got to move the processor closer to the sensor and preprocess the data before sending it to the ground,” Alderman says. “It is similar to satellite payloads that have a video camera and use a frame grabber to send images to the ground. Well if nothing happens you are sending the same frame three or four times. The data needs to be compressed so that only images that differ from the previous image are sent. This requires large processing systems such as VPX.”
“Getting the warfighter access to actionable intelligence is a challenge with the large amount of data generated onboard on a UAS platform,” Doyle says. “Determining which data are truly actionable can require a long time, so if you can automate recommendations at the payload level, it frees up the warfighter to focus on other tasks.”
“For small payloads, the data link bandwidth to the ground is even worse than for larger systems,” Michalski says. “A lot of these platforms won’t even have a link to ground or an imagery link. Sometimes the link bandwidths on the aircraft are smaller than what the SIGINT payload provider uses. It is also an asset management issue as the military doesn’t want to add more assets or ground personnel to operate new payloads while still operating the older payloads.”