Sunday, December 3, 2017

Small UAS ADS-B

In the early 2000s, the Federal Aviation Administration (FAA) began researching options for updating the aging infrastructure of the National Airspace System (NAS) of the United States. The FAA’s next generation initiative, or NextGen, was announced in January 2004 in response to direction published in the Vision 100 - Century of Aviation Reauthorization Act that assigned the FAA the task of updating the NAS (Houston, 2017).  The plan called for numerous changes, most aimed at addressing safety and efficiency of increasingly crowded airspace. One of the key components of this series of upgrades is the adoption of a new and more precise aircraft tracking system. Automatic Dependent Surveillance-Broadcast (ADS-B) is the technology slated to replace the current RADAR system as the primary means of tracking aircraft in controlled airspace (Houston, 2017). ADS-B uses other systems common in today’s aircraft, like global navigation satellite systems (GNSS), to broadcast aircraft data like position, speed, planned route, and more to receiving stations on the ground and in other aircraft. When this data is received it can be collected to build an extremely accurate model of the surrounding airspace and used to bolster the situational awareness of pilots and air traffic controllers alike. In order to maximize the effectiveness of the system in the NAS, the FAA has mandated the ADS-B transponder as required equipment for flight in a majority controlled airspace by January 1, 2020.

Those responsible for crafting the NextGen plan did so with the expectation the NAS congestion would continue to present issues, and the plan did well at addressing the issues known at the time. What was not foreseen was the exponential growth in unmanned systems, specifically small unmanned aerial systems (sUAS).  The FAA Aerospace Forecast for Fiscal Years 2017-2037 published the number of sUAS estimated at 4.47 million hobbyist aircraft and 1.616 million commercial aircraft (FAA Aerospace Forecast, 2017). This number dwarfs the amount of conventionally manned aircraft. The threat of an increasing number of sUAS potentially sharing airspace with manned aircraft was significant enough to cause the federal and state regulations stymie sUAS growth just as the boom began. Fortunately, the FAA was quick to adapt with release of 14 CFR Part 107. The regulation enables commercial operators to fly sUAS within specified parameters. Thus far, there is not a stated specific requirement for sUAS to be equipped with ADS-B, but the FAA mandate was issued with consideration for sUAS and applies to all aircraft with few exceptions. However, the addition of such a capability would offer an increased margin of safety for manned and unmanned aircraft.

Until recently, the possibility of including ADS-B technology in sUAS design was not thought likely due to the size and weight of transceivers. Weight directly affects aircraft performance, and even the smallest changes in weight can have significant impact on a sUAS. uAvionix, an aircraft avionics producer, has developed a micro sized ADS-B transceiver to address this shortfall. The Ping2020 is being marketed as the “world’s smallest and lightest full range ADS-B transceiver” (Ping2020, 2016). Per the uAvionix website, the Ping2020 measures 25mm x 39mm x 12mm and weighs 20 grams, and meets the following specifications:
   Detects aircraft threats on 1090MHz and 978MHz within 100 statute mile radius in real time
   Reports threats from aircraft in a programmable spherical radius
   Transmits ADS-B on 978 MHz at 20W nominal
   Meets MOPS DO282B Class A1S
   Navigation Source (GPS and Baro) FYXnav-B option
   SMA Antenna Connector
   GDL90 combined UAT and 1090ES traffic reports
   Direct integration with Pixhawk Autopilot
   Integration with DJI via DJI Software Development Kit (Ping2020, 2016)

This sensor is ideal for incorporation as an integral component of a comprehensive sense and avoid system. While stereo cameras, LIDAR, and small aperture radar are the preferred sensor options for obstacle avoidance, their limited range neglects long range and potentially fast-moving threats. They also do not broadcast position information. ADS-B fills this capability gap and could contribute to making beyond visual line of sight (BVLOS) operation of sUAS feasible and safe. As can be gleaned from the system specifications the Ping2020 maintains capabilities similar to larger systems with little negative effect on performance due to its light weight and size. The fact that the system integrates with DJI and the Pixhawk Autopilot make it even more appealing. As a sense and avoid sensor the ADS-B is relatively new to sUAS, but stands to revolutionize the largest segment of aviation by numbers. 

Ping 2020 sUAS ADS-B transceiver. Image as published on uAvionix website. Retrieved from https://www.uavionix.com/products/ping2020/


References

FAA Aerospace Forecast. (2017). Federal Aviation Administration. Retrieved from https://www.faa.gov/data_research/aviation/aerospace_forecasts/media/FY2017-37_FAA_Aerospace_Forecast.pdf

Houston, S. (2017, June 25). NextGen in a Nutshell. The Balance. Retrieved from https://www.thebalance.com/nextgen-in-a-nutshell-282561

Ping2020. (2016). uAvionix. Retrieved from https://www.uavionix.com/products/ping2020/



Friday, November 24, 2017

Generic Control Station

Ground based robots have seen a surge in use since the rapid development in the field of unmanned systems that began within the last two decades. Robotics developers have brought numerous options to market designed for a variety of tasks. One of the most successful types of task specific unmanned vehicles are ground robots. The myriad options offer users the ability to select systems best suited for their intended use and operating environment. A shortfall in early unmanned ground vehicle (UGV) development, however, is the use of system specific control stations. Utilizing multiple types of robotic platforms once meant an equal amount of differing control interfaces, each potentially greatly different from the other. This, in turn, increases the training and skill required to employ various unmanned systems. A relatively recent trend in the field of unmanned systems in general is standardizing human-machine interfaces with generic controllers capable of being used on multiple type platforms. One such device has been developed by Roboteam, an Israeli producer of tactical ground robotics systems.
The Ruggedized Operator Control Unit - 7” Screen (ROCU®-7), seen in the figure below,  was designed as a generic, hand-held controller capable of acting as the human-machine interface for several unmanned systems (ROCU®-7, 2017). With it, an operator is enabled to control several unmanned systems: land, air, or maritime. The handheld unit is designed to be simple and intuitive, and the controls are derived from the videogame world (Eschel, 2012). Data gathered by the unmanned system, as well as system operation information, is presented to the operator via a high definition video display. Video is streamed in real time to the control unit. A unique feature, though, is the use of a secure digital datalink based on commercial-off-the-shelf (COTS) elements facilitating fully encrypted, COMSEC protected two-way datalink, with efficient spectrum utilization enabling multi-platform operations (Eschel, 2012). 

Figure. ROCU®-7  common controller image as published on Roboteam website. Retrieved from http://www.robo-team.com/products/rocu-7/#s-3




Per the published system specifications, seen in the table below, the operating system was initially intended to be Windows 7 Pro 32bit or 64bit and the processor is an Intel Atom 1.91 GHz (ROCU®-7, 2017). The use of a common operating system maximizes the likelihood of compatibility. Additionally, the user interface incorporates 2 joysticks, 8 hard buttons, 4 rockers, 3.5mm audio in/out, and radio communication and GPS antenna jacks. The high-definition display screen is also sunlight readable, night vision imaging system (NVIS) compatible, and resistive touch capable (ROCU®-7, 2017).

ROCU®-7  Specifications
Dimensions (L x W x H):
(L x W x H) 11.8 x 6.7 x 2.9 in (30.5 x 17.5 x 7.3 cm)
Weight:
Without battery: 3.9 lbs (1.76 kg) / with battery: 5 lbs (2.3 kg)
Military Standard:
Ruggedized, IP65
Operating System:
Windows 7 Pro 32bit or 64bit
Internal Memory:
64GB/128GB Fast SSD
CPU and RAM:
Intel Atom 1.91 GHz Quad Core with 4GB RAM
Screen:
7”, 1024 x 600, Resistive touch screen, Sunlight readable, NVIS compatible
Power Supply:
BB2557/BB2590 Mil. STD., 24V Roboteam battery or any 8-40V power input (Wired/Battery Pack)
Working Time:
3-6 Hours (operation mode dependent)
Interface:
USB 2.0, Ethernet RJ45, Audio in/out 3.5mm, 2 Joysticks, 8 Hard buttons, 4 Rockers, RP-TNC jack for 
radio communication, SMA jack for GPS Antenna
Operational Temperature (MIL STD.):
-4 to 140 (-20 to 60)
Communication:
Digital encrypted communication (contact (Roboteam for more information)
GPS:
Internal

Note. ROCU®-7 specifications as published by Roboteam. (ROCU®-7 Ruggedized Operator Control Unit - 7” Screen, 2017)


The level of flexibility designed into the ROCU®-7 makes it a viable alternative for tactical users to limit the requirement for excess support equipment to utilize several different unmanned platforms. Just as with larger unmanned aerial systems, the benefits of employing generic control stations speak for themselves. The form and capability of ground, maritime, and air unmanned systems changes rapidly, but the field can benefit by standardizing the control interface when possible. The efforts of Roboteam, as displayed by the ROCU®-7, are contributing to simplifying the way we employ these systems. 






References

Chuffart, V. (2015, June 9). The Iron A-Team: Unmanned Robots in Harsh Environments. Kontron. Retrieved from https://www.kontron.com/blog/embedded/roboteam-the-iron-a-team


Micro Tactical Ground Robot (MTGR). (n.d.). Combating Terrorism Technical Support Office. Retrieved from https://www.cttso.gov/?q=MTGR

ROCU®-7 - Integrated, Highly Intuitive & Secure Wireless Common Controller. (2017). Roboteam. Retrieved from http://www.robo-team.com/products/rocu-7/#s-0


ROCU®-7 Ruggedized Operator Control Unit - 7” Screen. (n.d.). Roboteam. Retrieved from http://www.robo-team.com/wp-content/uploads/2016/08/ROCU7_Web_2017_1.pdf

Sunday, November 12, 2017

Unmanned System Data Protocol and Format: DJI Inspire 2

Commercial unmanned aerial systems (UAS) available to consumers vary greatly in size and capability. Purpose built systems are offered by producers for a variety of missions. For aerial photography and cinematography, Dà-Jiāng Innovations Science and Technology Co., Ltd (DJI) has established itself as an industry leader in small UAS (sUAS). The technology firm offers a range of options for aerial camera drones that span from those appropriate for amateur users to professionals. One of their most capable and well-rounded platforms is the Inspire 2 filmmaking drone.

The Inspire 2 (see Figure 1) is an sUAS designed with professional photography and cinematography in mind. The aircraft design form is an electrically powered quadcopter. The drone is satellite navigation enabled, has an obstacle avoidance system, and can function with a high degree of autonomy or can be configured to use two remote controls; one pilot and one camera operator. At first glance, the system looks very similar to the previous iteration. There have been many changes, however, that have contributed to evolve it to a more capable and safe aerial cinematography sUAS.

Figure 1. DJI Inspire 2 with Zenmuse X5S Camera image as published on https://www.dji.com/inspire-2/info

Perhaps the most innovative advances on the platform lie in the cameras and data storage system. The drone incorporates a first-person view (FPV) camera for the streaming video to the pilot, as well as a customizable primary camera payload. The highest resolution camera compatible with the platform is the Zenmuse X7 (see Figure 2). The camera features a Super 35 sensor, and is able to shoot 6K video and 24-megapixel still photography (Zenmuse X7, 2017). Its 24-megapixel sensor is rated at 14 stops of dynamic range and can collect continuous RAW images in burst shooting at a rate of 20 frames per second (Zenmuse X7, 2017). The lenses can be changed to meet the photography or cinematography needs. DJI offers four specifically designed for the image sensor; 16mm F2.8, 24mm F2.8, 35mm F2.8, and 50mm F2.8 (Zenmuse X7, 2017).

Figure 2. DJI Zenmuse X7 Camera image as published on http://www.dji.com/zenmuse-x7/info#specs

To manage the high-resolution imagery, the camera integrates with the CineCore 2.1 Image Processing System (Esulto, 2017). The system records in two formats; CinemaDNG and Apple ProRes (Zenmuse X7, 2017). Both formats allow for compression of very high-resolution image files. Though not as condensed as other formats, the reason for using them is the ability to maintain a high dynamic range and overall better image quality. The drawback is a very notable increase in file size.

The increased file size is an issue that had to be addressed for the Inspire 2. Data storage is a challenge with the size of files for high-resolution imagery. The aircraft does use the industry standard of Micro-SD cards, but for greater demand, DJI has introduced the use of optional solid-state drives (SSD) (Inspire 2, 2017). The use of a SSD, CINESSD as it is called by DJI, allows operators to collect video footage at the camera’s highest resolution. DJI offers CINESSD drives in sizes ranges from 120GB to 480GB. Filming video the highest quality video requires a generous amount of storage space. For this application, quality takes precedence over compression.


Until another data format is developed that will allow for high-resolution imagery to be compressed without degradation of quality, the current alternatives should be expected to remain the standard. We have seen image processing on platforms move from the camera to the vehicle, allowing for increasingly capable cameras to be fitted to the gimbal mount and yet maintain smaller form factor. I see the most notable changes in the near future to this platform to be in data storage. The CINESSD is a step in the right direction, but even the largest drive available has its limitation when shooting in the highest resolution. I recommend DJI continue to develop storage technology to meet the demands of today’s high-end camera systems. This may be achievable with next generation SSD technology like that used in Intel and Micron Technology’s 3D XPoint.

Case Analysis Effectiveness

Communicating the detailed points of an area of study can be a challenging task. Developing sound methods and collecting research data fo...