Unmanned Little Bird Testing Approach - AHS Tech Specists Meeting Jan 2009

Presented at the American Helicopter Society International Technical Specialists Meeting on Unmanned Rotorcraft Systems, Scottsdale, Arizona, January 20 – 22, 2009. Copyright © 2009 by the American Helicopter Society, Inc. All rights reserved.

Unmanned Little Bird Testing Approach

Mark Hardesty1, David Guthrie2, Dino Cerchie3

Abstract

The Unmanned Little Bird (ULB) program was initiated in the Fall of 2003. Program discriminators are its flight-test friendly design as well as the unique approach for developing VTOL UAVs. First flight occurred on September 8th, 2004, with a fully autonomous multiple waypoint demonstration flight from takeoff through landing achieved six weeks later.

The ULB team succeeded in creating a powerful UAV technology development and demonstration aircraft, assisting in the rapid development and understanding of the operational concepts and requirements. The platform’s autonomous characteristics continue to be expanded through low risk testing in support of UAV subsystems development.

The Unmanned Little Bird’s design size and proven performance make it highly competitive in the evolving multi-mission VTOL UAV market. The design approach and integrated test capability that the ULB provides supports rapid development and cost avoidance in the growing VTOL UAV market. As with the fixed-wing UAVs that vary from hand-helds to over 100-foot wing spans, it is envisioned that there will be more than one VTOL UAV design world-wide once a design is demonstrated in the field highlighting the benefits provided by VTOL aircraft.

Introduction

The Unmanned Little Bird (ULB) program represents a true paradigm shift in the high-risk development of larger VTOL UAV systems. While the norm has been to develop UAV understanding on lower cost

scale models, or even step directly to the full size aircraft and the risk associated with that approach, the ULB team decided to create a full size UAV aircraft without incurring the risk associated with traditional UAV development programs.

___________________

1 Flight Test Director, Unmanned Little Bird Program, The Boeing Company, Mesa 2 Project Test Pilot, Unmanned Little Bird Program, The Boeing Company, Mesa 3 Program Manager, Unmanned Little Bird Program, The Boeing Company, Mesa

2

This approach has created the “optionally manned” UAV test aircraft. The design allows a program to safely develop the sub-systems that will lead to its initial fielding, but also support the continuing development that all aircraft experience once they are found useful.

The figure below explains the ULB design concept.

Figure 1. ULB Design Concept

A new UAV program, when developed from the concept stage, requires investing a significant part of the development cost into the platform. In reality, autonomous payload placement is the objective unless unique attributes of the platform itself contribute significantly to the mission objective. Also, once the baseline platform is complete the aircraft still needs to qualify all of the capabilities that the customer requires.

A significant cost saving has been realized in the ULB program design approach. The platform is complete with an unmodified external shell, providing internal mounting volume for the many capabilities already designed and qualified for the aircraft.

The key attributes of the ULB program are:

leverage existing technology

treat the UAV as a mission capable aircraft and provide mission planning as the basis of operations

reduce the ground station workload

make the ground station compact and mobile

make the platform autonomous so that the aircraft takes care of itself and the GCS operator can concentrate on the payload products.

Boeing Mesa has excelled in rotorcraft development over the years. The ULB program was the integration of several helicopter specific internally-funded programs in the area of control laws, avionics architecture, and human-machine interface. This allowed the program in 2004 to go from initiation to a flying prototype in 9 months, to fully expand its envelope to 18,000 feet in 4 months after first flight, and to accumulate more flight hours that any other VTOL UAV platform with payload capabilities greater than 100 pounds.

The ULB program name covers two aircraft designs, the unmanned version of the 3100-pound commercial MD 530FF helicopter that first flew in the autonomous mode in 2004, and its more capable unmanned variant; the A/MH-6M manned aircraft, more commonly known as the MELB, Mission Enhancement Little Bird, that first flew in the autonomous mode in 2006.

Figure 2. ULB program test assets

Boeing currently has one ULB aircraft (N7032C) and two UMELB aircraft (N106HX & N206HX). All three aircraft are designed to be “optionally manned” aircraft. This design feature allows these aircraft to be flown as single or dual pilot in the contiguous US in day and night VFR conditions. They are licensed by the FAA as

Airframe UAV Payload Qual

New UAV Program Cost

UAV

Leverage existing airframe Leverage existing

payloads

3

experimental aircraft for the purpose of “Research & Development/Market Survey”.

All three aircraft can also be operated as UAVs, controlled from a ground control station, with or without a pilot aboard. When flown as a UAV without a pilot, the aircraft must comply with the standard UAV rules of flight within FAA designated airspace or within restricted airspace.

The ULB aircraft has logged 576 hours of flight testing since September 2004. The UMELB design has logged 219 hours of testing since its first flight in December 2006.

Fixed Wing UAVs Success Relative to Rotorcraft UAVs

The tremendous contributions that fixed wing UAVs of all sizes have made over the last several years in both civil and combat environments are well documented. The lack of any operational VTOL UAVs for either civil or military applications is equally noteworthy. The general complexity of VTOL designs and the operational considerations for VTOL UAVs in terminal area operations continues to delay the fielding of any VTOL design.

Figure 3. Sensor performance for brownout landings

A UAV is simply an aircraft equipped with; a programmable all-axis autopilot for up and away flight; a command/control data link with remote control and monitoring station, and a method for either automated or manually controlled takeoffs and landings. The absence of an operational VTOL UAV

in either a civilian or military application may highlight the difference in maturity of manned fixed wing autopilot control relative to manned helicopters. While cruise flight is relatively straight forward and leverages fixed-wing autopilot methodology, the zero-zero landing capable helicopter autopilot represents a tougher challenge.

Only a handful of programs in the history of helicopter aviation have attempted a full authority automated flight control system. Much of the ULB’s flight test related success is due to the lessons learned during helicopter specific control law programs over the years such as the fly-by-wire AH-64 program that was flight tested in the late 1980s.

Figure 4. Landing to 16’x16’ platform

In addition to the challenge of VTOL autonomous takeoffs and landings, operation in General Airspace requires the UAV to have be able to “see and avoid” air traffic and obstacles.

Large fixed wing UAVs such as Predator operate in a similar manner to manned IFR flight operations, and generally at altitudes where airspace clutter is limited and positive control is maintained to provide separation for competing aircraft. Just like their manned counterparts, VTOL UAVs are more suited for lower altitude operations where aircraft separation may not be provided by air traffic controllers.

4

Transponder based electronic collision avoidance systems (TCAS) are certified and commonly found on commercial or higher end private aircraft. However, a barrier to UAV operation in the national airspace is the detection and avoidance of what can be called “non-compliant” aircraft which are not equipped with TCAS.

Emerging technologies that exploit IR, EO, and various radar devices as inputs to sense and avoid capability are still in prototype form and not yet ready for certification. However, the ULB program is furthering their maturity by supporting their development with safe flight testing in congested airspace. It is expected that these devices will be certified to provide an “equivalent level of safety” to a human piloted aircraft for traffic sensing. The “avoid” piece of the equation represents an additional challenge, as the response must be tied to the sensors effective field of regard.

Figure 5. Enhanced Synthetic Vision System test configuration

ULB Program Philosophy – Lower The Risk, Raise The Yield

Many aircraft designs are defined and presented as the final design even before they attempt first flight. Flight test qualification is viewed by these programs as an inconvenient step towards fielding. Therefore, many designs, and especially UAVs, do not appreciate the unique qualification tests necessary to define a safe operational flight envelope.

By 2006 the program had accumulated several hundred hours of development and the tests were considered low risk, even for a proof-of-concept demonstrator aircraft. However, to some observers the program had not answered the fundamental UAV question – could it fly without a safety pilot. The ULB program proved it could operate completely unmanned in 2006 when the aircraft flew a 20 mile mission sortie with over 700 pounds of payload for this first unmanned flight. Absolutely no operational differences existed between this flight and the preceeding prepatory sorties.

Figure 6. Purely autonomous (no safety pilot on board) demonstration flight at Yuma Proving Ground, June 2006

Development testing is characterized by changes to either the flight envelope or the aircraft capabilities. Additional unmanned flights during developmental testing adds risk as well as substantial time and cost associated with software/system validation.

The ULB program philosophy is that anything more than proving the unmanned flight capability is a reliability issue. All VTOL UAV programs are seeking to achieve platform reliability on the order of 10 or less accidents per 100,000 flight hours (similar to manned platforms). Therefore,

MPG06-93788-017.ppt

5

reliability is better addressed through analysis.

The ULB program’s “safety pilot” approach brings two notable strengths to the program. First is the ability to rapidly develop and integrate any system in a low risk environment. When the risk of losing an aircraft or sensor package is removed, the ability of the team to develop system capability is greatly enhanced. Second, it brings test pilots into the UAV program as key members of the design staff. This is true for manned programs, but rarely true for UAVs. The cockpit perspective, even when the flight data may indicate that the test was acceptable, proves invaluable when developing an optimized aviation system.

Additionally, conducting developmental flight test with a pilot onboard allows the aircraft to be operated amongst the general aviation community which avoids the costs associated with relocating the aircraft, equipment, and support personnel to locations with restricted airspace. Even inside restricted airspace, approval must be obtained from the owning agency prior to operations. The avoidance of travel expenses combined with the elimination of the delays and expense of approvals to fly UAVs in restricted airspace has allowed the ULB program to reap enormous financial benefits. Developmental flight operations conducted with the safety pilot on board better serve the industry and the customer base by more rapidly advancing the technology and developing concepts of operations, while simultaneously minimizing the timelines and associated costs.

The “lessons learned” by operating at a high operational tempo with multiple flight experiments flown in a single day are simply not available to programs that only fly occasionally. On one technology integration and demonstration program, the ULB flew an average just over 8 hours per day for 3 consecutive days in support of both day and night mission related testing back in 2005.

The safety pilot also provides immediate feedback regarding the “machine like” or

“human pilot like” flight capabilities of the VTOL UAV. The advantage of a smooth ride should result in longer sensor component life due to the reduced vibration present in steady state and maneuvering flight. This will be significant factor in the total operational budget when the sensor cost may approach or even exceed the cost of the baseline aircraft.

In short, the presence of a safety pilot accelerated the VTOL UAV development process. If errant autonomous mode behavior was presented during the development phase of the aircraft, the safety pilot was able to quickly take manual control of the aircraft and debrief the test team. In contrast, a program working without a safety pilot may have spent additional time and budget determining the cause of an incident as well as how to avoid future occurrences. It was not uncommon during the first autonomous mode flights to find an issue during the morning flight, debrief, make a software change, and validate the software change during the afternoon flight.

Figure 7. Safety pilot allows different mission concepts to be explored

6

Design Concept

The rapid prototyping philosophy of the ULB team dictated that off-the-shelf actuators would be exploited to develop the UAV concept. Additional actuators were added to manage throttle and collective control inputs, and a patented method of combining the load relieving low bandwidth control force trim actuators along with the high bandwidth autopilot actuators has resulted in a design that flies in a manner virtually identical to a skilled human pilot.

The UAV actuators are installed in parallel to the mechanical control system. When the aircraft is in the autonomous mode, the safety pilot can disengage the system or mechanically over-ride the UAV system, taking full flight control authority. The pilot’s treat the UAV system as a student pilot, always alert to the possibility of an improper control input at any time.

Figure 8 depicts the cockpit of the original ULB test aircraft. The red button on the cyclic stick is what instantly transforms the UAV to a pure mechanical aircraft. The display mounted in the console in front of the pilot allows the crew to disengage the system, modify gains in the control laws, and then re-engage the system. The system optimization is measured in days rather than months. In the early days of development, the flight computer would launch with two distinct software loads to test in one sortie. It was also not unusual to fly one software load in the morning and the fix in the afternoon. These were flights that were expanding the flight envelope or adding a new aircraft capability. The software tools are mature enough that a change would be made, tested in ground simulations and loaded into the aircraft in a span of time less than two hours.

Figure 8. Original ULB Cockpit

The 8.4” diagonal touch screen on the left side of the cockpit is driven by a hardened PC, streamlining integration issues for PC based payloads and test hardware. Due to this design strategy nothing added to the aircraft is flight critical. This opens the door for flight testing prototype, non-flight qualified hardware to help finalize its design and performance prior to the expensive qualification process.

Technological Development & Demonstration Capability

The strength of the ULB design is that it can safely investigate a new capability in a very short period of time and bring that new capability within the aircraft’s already proven capabilities for the unmanned design. There is no question that the operational design will be offered in a purely unmanned configuration. The ULB has just provided what appears to be a more efficient, cost-effective solution on how to get to the end point and develop new capabilities without losing program assets or precious prototype payloads along the way. Many of the payloads tested are nearly the cost of the aircraft itself. Therefore, knowing that the payload will be tested and returned safely is a key feature of the ULB design.

The safety pilot approach also allows the ULB to test anywhere in the U.S. Unlike other UAV programs that must reside within restricted airspace or defined COAs, the

7

ULB can go to where the test is rather than forcing the test to come to a UAV. This becomes very powerful when participating in large scale exercises where the UAV may not be the most important element. Figure 9 illustrates this point, showing the ULB aircraft conducting testing with other manned aircraft in the local airspace at Boeing Mesa. Figures 3 and 5 show the testing of an Enhanced Synthetic Vision System for brown-out landing conditions, conducted at a low cost test area in rural Arizona.

Figure 9. Manned / unmanned testing.

Figure 10 documents the test configuration for a ground based hostile fire flash detection system as a prototype for an aviation installation. Again, this test was conducted in local airspace adjacent to a civilian weapons firing range. The results from this testing provided the design data to

yield a smaller optimized aviation unit with the required performance.

Figure 10. Testing a prototype hostile fire flash detection system.

Figure 11 shows a developmental test that was conducted using the ULB to host a high-bandwidth UAV communication system. The antenna mounted to the bottom side of the aircraft was a phased-array antenna previously mounted to a fire station tower in the mountains of California. The un-altered assembly was installed on the ULB aircraft and flown in a demonstration at Ft Campbell, Kentucky.

8

Figure 11. High bandwidth comm relay configuration

The autonomous and programmable capability of the ULB makes it a good test platform for aircraft that may not even be designed to fly on Earth. The test aircraft has supported Lunar and Mars lander navigation and landing zone identification technology[1]. Figures 12 and 13 depict 2 of the 6 experiments flown during this test program. The ULB demonstrated that it could emulate the final approach and landing phase of extra-terrestrial flight trajectories, providing a safe and inexpensive method for developing and validating a variety of technologies.

Figure 12. Crater navigation sensor suite for development of lunar lander capabilities.

Take it to the Customer

Decision makers in the customer base are located all over the world and have very limited time to travel to manufacturer’s locations. The safety pilot on board concept guarantees that the ULB can be operated at the customer location. The availability of an observers seat in the cockpit insures that the customer can experience first hand the unique flight capabilities of the aircraft and system, as well as to witness that the safety pilot is not intervening in the operation of the aircraft. The ability to develop and upload missions using a simple laptop computer or via TCDL simplifies the cost and time of demonstrations.

Figure 13. Prototype flash LADAR for rough terrain landing zone assessment

On several occasions the ULB has been flown through civil airspace to customer locations in order to provide on-board demonstration flights to high level decision makers. Figure 14 is from an autonomous flight demonstration provided to various individuals at Redstone Arsenal in Hunstville, Alabama.

The safety pilot approach also insures that UAV related testing can be supported anywhere in the United States. Precision delivery using VTOL aircraft is a growing requirement and its military application may require operation to and from higher ground elevations. The ULB can safely and immediately conduct testing in any

MPG06-93788-006.ppt

9

environment to validate operational requirements in any terrain (Figure 15).

Figure 14. ULB performs autonomous demonstration flights at Redstone Arsenal in Huntsville, Alabama

Figure 15. Aerial resupply CONOPS

Logistics Footprint

The ULB test team is very compact and efficient. In a typical test scenario, the aircraft flies through national airspace as a manned aircraft to the test location with the test truck and crew in chase. The truck holds a ground power cart for extended ground test operations, the TCDL antenna tower, and a connex of spare parts and tools to support the aircraft during the testing (Figure 16).

Figure 16. ULB support hardware

Platform Selection

The MD530FF is a platform with a rich heritage. Initially conceived as the OH-6 for service in Vietnam, the airframe has enjoyed a long evolution and still shows much growth capability. The Mission Enhanced Little Bird operated by the 160th Special Operations Aviation Regiment based at Fort Campbell, Kentucky is the latest generation military variant of the MD530FF. The MELB helicopter has a military pedigree complete with a -10 operators manual, maintenance manuals and procedures, parts supply chain, and a wealth of experienced pilots, maintainers, and mission planners. A large variety of sensor, weapons, communications, and extended range fuel systems are qualified and available for the airframe. There are hard points all over the aircraft for mounting equipment and the pre- and post- flight inspections are quick, easy, and well documented. The electrical system is capable of 400 amps of 28 vdc electrical power output. Transportability by truck or military airlifter is customary and well understood.

10

Figure 17. MELB at a NASCAR event

The A/MH-6M growth variant is one of the few aircraft that can carry its own empty weight in fuel and payload, in fact 30% more than its empty weight (Figure 17). The still maturing aircraft has enjoyed a few decades of refinements and improvements driven by military customers over the world to help develop one of the lowest operational cost aircraft in its payload category. It also has many existing military qualified capabilities that can easily be made automated for the unmanned application.

Airframe Performance Growth Potential

In its current form the aircraft has documented performance for 4700 pounds, demonstrating a mission equipment and fuel capacity of 2400 lbs. Future changes to the aircraft will be driven by both the manned and unmanned customers.

The Rolls Royce 250-C30R/3M engine on the aircraft provides the existing design with high altitude HOGE capability. This is unlike many designs that are optimized for sea level operations and suffer tremendous weight and performance penalties when operating from higher ground elevations. Even with its current performance, additional engine power and drive train growth programs are taking shape.

New main and tail rotors will also be incorporated in the coming years. The aircraft already has crashworthy and

ballistic tolerant fuel tanks for the UAV application as an added safety feature for the ground crew. Even though the design has demonstrated sloped landings greater than 7 degrees in all axes, the design team is looking to expand the landing envelope further, both in slope and contact velocity for maritime applications.

A UAV specific design enhancement would be the integration of an engine recouperator to increase the efficiency of the overall thermal cycle of the installation. Consequently the aircraft performance envelope would improve in terms of increased flight duration or increased mission equipment package payload capability due to the reduction in fuel burn.

Forward Thinking – What’s Next?

Time and funding appears wasted on true UAV programs during the development stage in safety reviews and extensive regression testing due to the risk associated with the loss of an aircraft. The ULB program has eliminated that loss possibility and can develop and understand critical flight and design capabilities quickly and safely.

Selection of an airframe already in the inventory, with all the certifications and a myriad of qualified weapons, sensors, and communications systems greatly reduces the cost of UAV development. Perhaps more importantly, this choice allows precious funding to be applied to developing UAV capabilities, rather than being spent on the basic development and certification of the airframe and fundamental systems. With only the UAV capable autopilot and command/control data link to develop and integrate, a VTOL UAV with powerful capabilities can be fielded much more quickly and for a fraction of the cost of traditional UAV programs.

The ability to operate as an optionally manned platform is a huge advantage. With shrinking budgets and growing requirements for both VTOL UAV and

11

manned VTOL platforms, military and civil customers can enjoy all the advantages of both types of platforms. As the ULB UAV concept can be deployed in a kitted form, airframe operators have the option of a growth path from manned to optionally manned airframes as CONOPS, requirements, and the regulatory environment mature.

The maritime environment presents special challenges for terminal operations and dynamic mission replanning. To that end, the ULB program has collaborated with NovAtel [2] to develop and test a moving baseline Differential GPS based navigation system. Initial testing of this low cost JPALS approach has provided very promising results. Formation landing approaches have been completed to a moving automobile (Figure 18) and a helipad has been installed on a tractor-trailer rig (Figure 4) for autonomous takeoff and landing development testing at speeds up to 50 knots. Work is ongoing to examine at-sea terminal operations demonstration for 2009.

Figure 18. Moving baseline navigation system development

Conclusion

The technical success of the ULB program can be attributed to several factors. Most importantly are the people; team members

that are willing to think outside the box and develop an entirely new approach to an existing problem. The ability of the aircraft to be pushed to the envelope limits, to test new payload capabilities, and to investigate new autonomous capabilities safely and rapidly is the key. High operational tempo operations promote excitement and high team morale.

The ULB concept appears to be the standard that all new designs must improve upon. Many design concepts falter upon reaching flight test and only after significant investments are made.

The Little Bird is not a stagnant design; rather it continues to evolve. In the unmanned mode, the excess payload capability and additional fuel capacity, extend the range of the UAV design to values that far exceed the current manned aircraft. Recently, a new test cockpit was installed in the ULB to allow data fusion testing, where the data can be validated at the aircraft level prior to being sent down to a ground station (Figure 19).

Figure 19. Second generation cockpit design

12

In summary, ULB is the most cost and calendar time efficient VTOL UAV development platform available for rapid prototyping. The design continues to evolve and mature along with the complimentary technology. ULB allows safe and effective exploration of CONOPS so that regulatory and operational requirements documents can be developed. The ultimate objective is to provide the customer the most capable, effective, and flexible platform possible, while allowing for continued growth as ideas continue to mature.

[1] M. Bayer, S. Berg, M. Hardesty, “Helicopter Flight Demonstration of Lunar and Planetary Lander Technologies”, Proceedings of the American Institute of Aeronautics and Astronautics Space 2008, San Diego, CA, September 9-11, 2008, AIAA-2008-7803

[2] T. Ford, M. Hardesty, M. Bobye: “Helicopter Ship Board Landing System” , Proceedings of the Institute of Navigation Global Navigation Satellite Systems, Long Beach, California, September 15th, 2006.