Of the remaining regulatory and technological issues, the goal is the certification of a system of technology, feedback, analysis and control, which reduces the risk of an air to air collision, to the same level of risk currently enjoyed for manned flight, is of paramount interest and importance. The regulations governing DSA are contained within 14 CFR 91.113 "Right of Way Rules". ASTM has published a standard, F2411-04e for "DSA Collision Avoidance" and is available for purchase from ASTM International. David Grilley of Alion Science has recently published a paper with AUVSI which describes the problem and represents an analytical framework to evaluate systems that qualify as candidates for DSA within a small UA system.
The most common term for this capability is Detect Sense and Avoid (DSA). The military uses deconfliction. Progress has been made in DSA technology development, is continuing, and more advances are inevitable. The question is - What level of efficiency is sufficient to satisfy the "Comparable to Manned Aircraft" level of safety requirement for collision avoidance for UASs?
Mid-air collisions (MAC)s are rare however they do occur and near misses happen. Most MACs occur under VFR conditions. A study, conducted by the Bureau d'Enquêtes et d'Analyses pour la Sécurité de l'Aviation Civile (BEA), covers mid-air collisions that occurred over French territory between 1989 and 1999 - 17 in total were reported. On average there were 1.5 mid-air collisions per year. Between 1978 and 1982 the FAA reports (PowerPoint) a total of 152 MACs. Dr. H. Paul Shuch writes about "Near Midair Collisions (NMAC): How Many Really Occur?" The FAA has been collecting data from pilots on NMAC since 1959 and generally gets about 1000 such incidences on an annual basis.
In the United States there is an organization called the "Bird Strike Committee" formed because bird strike accidents have caused over 195 aviation deaths since 1988 and account for over $600 million dollars in damage annually. Since many MUAVs, defined as UASs that weigh 55 lbs or less, many 15 lbs or less have mass and kinetic energy similar to a large bird, it begs the question, what is a useful analytical paradigm to certify a Micro UAV to a standard of a "Comparable Level of Safety to Manned Aircraft"?
Scott Foster, of Foster Flight, a manufacturer of a candidate optic camera system technology for deconflicting MUAVs and SUAVs has expressed concern that a standard based upon "comparable safety relative to a GA's pilots eyeballs" as an very low, perhaps insufficient, level for the DSA bar to be set at. Safety in flying a Cessna 172 with eyeballs for DSA is as much a matter of attention span as it is visual acuity. Human factors are a critical component in Mid Air Collisions according to Taneja and Weigman of the University of Illinois at Urbana-Champaign.
The Aeronautical Information Manual, published by the FAA, is an important reference document. Chapter three describes the Airspace Classification System. Chapter Five describes Air Traffic Procedures. The summary of Airspace Basics is a useful primer as is this very helpful descriptive primer. The risk associated with the operation of a UAV is directly proportional to air traffic and that can be directly correlated to the class of airspace the UAS is operating in.
Intuitively, a small 38 pound UAV operating in class B, C, D, or G airspace, at say less that 1000 ft, controlled by line of sight under VFR conditions is likely to have a lower intrinsic probability of a MAC than that of a HALE UAV operating in class A airspace along commercial routes or during takeoff and landing within Class B or C airspace. This may be true even though the larger airframe can support an effecient DSA technology mix. The metrics are not yet available, the test paradigm to certify a DSA System has not been established, and the analytical statistical probability analysis are not yet completed to determine the difference in the MAC potentiality between these two operating scenarios.
The ASTM F-38 Committee has issued a published standard for DSA collision avoidance, (F2411-04 DSA Collision Avoidance) that requires a UAV to be able to detect and avoid another airborn object within a range of + or - 15 degrees Elevation and + or -110 degrees Azimuth and to be able to respons so that collision is avoided by at least 500 ft. The 500 ft safety bubble is derrived from the commonly accepted definition of what constitutes a Near Mid Air Collission. This gives airframe and avionic/DSA electronics manufacturers a target for certification. It is likely that the ASTM standard will be incorporated by reference in eventual FAA certification requirements.
Flightglobal.com hosts a video of a near miss between a LUNA military UAS and an airlines near Kabul Afghanistan. The day will come when a UAS is responsible for causing a mid air collision that results in death.
It is critical to create a system for technology certification that minimizes the probability of this event while at the same time evaluating probability and risk assessment in realistic terms. The statistics for manned aviation on MAC, NMAC coupled with probability studies similar to those conducted at ICAT evaluating airspace class and population densities to determine acceptable levels of risk comparable to manned aviation should result in methodology to certify DSA systems without unduly retarding the growth of the UAS industry.
An article published in Aerospace America June 2002 entitled Avoiding Collisions in the Age of UAVs summarizes some of the progress that has been made to date. A second article entitled UAVs Entering the NAS published in Avionics Magazine in June 2005 issue reviews similar topics.
Military and commercial aviation safety currently benefits from several levels of redundancy starting with "eyes in the sky", pilots who have visible contact with the surrounding airspace and hands on control to react quickly in the event of emergency. "Eyes On" has historically been deemed critical to air safety. Pilots operating under Visual Flight Rules (VFR) are backed up by cooperative transponders, sensors in manned aircraft that uniquely identify each airplane with a signal recognizable to other aircraft, and air traffic control radars and controller staffs monitoring the NAS.
There is currently internal discussion within the FAA about whether existing Detect See and Avoid (DSA) technology including ground based RADAR and airborne synthetic aperture radar (SAR) can meet the needs for collision avoidance.
Sandia National Laboratory has a long (40 year) history of developing Radar systems and provides an intuitive technical description about what SAR is and some of the applications that it has studied for SAR systems.
It is noted that light sport aircraft, Para foil, or parachutists and many UAVs are made of composite material that have a small radar footprint and that some sort of EO/IR “eyes in the sky” camera system with video feedback to the UAV operator may be necessary.
A paper from the MIT Lincoln Laboratory titled "Safety Analysis Methodology for UAV Collision Avoidance Systems" was presented at the 6th USA / Europe Seminar on Air Traffic Management Research and Development(June 2005). The MIT ICAT has published a series of relevent studies by Weibel and Handsman entitled "Safety Considerations for Operation of Unmanned Aerial Vehicles in the National Airspace System"(2005) and "Safety Considerations for Operation of Different Classes of UAVs in the NAS" (2004)- both papers discussing models developed to address relative safety (risk) for mid air collisions and ground impact (airworthiness). A third Weibel and Hansman paper, presented at the 2005 AIAA conference entitled "An Integrated Approach to Evaluating Risk Mitigation Measures for UAV Operational Concepts in the NAS" also sheds light on this topic. Finally a paper by the same authors entitled " Safety Considerations for Operation of Small Unmanned Aerial Venhcles in Civil Airspace" discusses risk associated with small UA systems.
The USAF Research Lab, Air Vehicles Directorate, Control Sciences Division, Systems Development Branch at Wright Patterson AFB has published an article attempting to define Sensing Requirements for UAVs. The Neil Taylor and the University of Cambridge have written on the Challenge of Integrating UAVs into Mixed User Airspace.
The USAF has recently issued a Broad Agency Announcement (BAA) intended to identify and qualify contractors and technologies for development and testing of DSA technology for the Predator Class UAV - the BAA response closes in July 2007. As of the September 2005 Fact Sheet issued by the FAA "The design of many UAVs makes them difficult to see and adequate “detect, sense and avoid” technology is years away." Further, The Netherlands Organization (TNO), has initiated a two year study on UAS DSA.
So the race is on to determine a winner. As in all aspects of UAVs for commercial use, cost is a principal factor affecting usage and market penetration. A key for successful development will be adapting or repurposing FAA certifiable military technologies for use with small lightweight commercial UAVs, at costs the market can absorb.
Competing technologies and vendors include the following companies and products:
PBS's NOVA website on Spies That Fly has an overview of Synthetic Aperture Radar, how SAR creates an image and how an image is analyzed.
NASA and Scaled Composites, LLC in 2003conducted a series of tests Final Report using the Proteus UAV at Scaled Composites Mojave facility to evaluate a Detect See and Avoid System without the benefit of cooperative transponder equipped systems, following a series of tests in Las Cruces in 2002 . The Amphitech OASys high frequency radar system operating on the 35 GHz (Ka band) was used to detect any approaching aircraft on a potential collision course within 7 nautical miles. The Proteus was challenged by a variety of aircraft including an F/A 18 jet and a hot air balloon.
PSL at the TAAC within the University of New Mexico, in conjunction with the 46th Test Group at Holloman AFB and the Air Force Research Labs, conducted DSA collision avoidance tests at White Sands Missile Range in November of 2005. An Aeronautics Defense Systems Aerostar equipped with a Mode C transponder was used as a test be platform. More tests are planned.
Northrop Grumman is helping the USAF develop a collision avoidance system for UAVs (see also).
General Atomics LYNX AN/APY-8 - SAR/GMTY Radar System is a high resolution Synthetic Aperture Radar System developed for use on Predator class UAVs. This LYNX White Paper describes the system in detail.
The UAV Applications Center, NASA Ames Research Center, Moffett Field, CA has developed a Radar based system and specialize software algorithms for use as a ground based, mobile, area of operations portable solution.
Sandia National Laboratories contribution is the MiniSAR a lightweight, 30 pound radar, with an operating range of 15 kilometers versus the 35 Km range of their larger synthetic aperture radar systems.
Global Aerial Surveillance has signed an NDA with a well known developer of laser systems and is conducting an evaluation phase on a system for military use.
The Tactical Endurance Synthetic Aperture Radar powered by Mercury Computer Systems Inc. RACE ++ Processing Systems is operational in the Predator UAV.
Amphitec in cooperation with ERAST and NASA has demonstrated their OASys DSA radar technologies ability to detect, track and report a variety of aircraft in flight at ranges up to 8 NM. See report.
EADS is developing its QuaSAR lightweight Synthetic Aperture Radar for MALE or Tactical UAV applications.
Flight Safety Technologies' UNICORN System, using an Ultra Wide Band Radar (UWB) for high resolution and a micro strip patch array antennas, is under development.
MIT researchers have written a paper of interest to this field. UAV Trajectory Design Using Total Field Collision Avoidance calls for a distributed computational approach to forecasting and preventing collisions of N UAVs. As N increases to a large number the potential for collision grows exponentially.
UAVM.com will update this page regarding news in the area of development DSA technology applicable to the Commercial Use of UAVs.
