6

Vehicle Technology Directorate and Autonomous Systems Enterprise

INTRODUCTION

The Panel on Air and Ground Vehicle Technology (AGVT), along with selected members of the Panels on Survivability and Lethality Analysis, Sensors and Electron Devices, Armor and Armaments, Digitization and Communication Sciences, and Soldier Systems, reviewed the Army Research Laboratory’s (ARL’s) Autonomous Systems Enterprise at the ARL facilities at Aberdeen, Maryland, on July 11-13, 2011. The Robotics Collaborative Technology Alliance (CTA) and the Micro Autonomous Systems and Technology (MAST) CTA are integral parts of ARL’s Autonomous Systems Enterprise. The Vehicle Technology Directorate (VTD) was reviewed by the Panel on Air and Ground Vehicle Technology at its meeting on June 25-27, 2012. The Directorate has four divisions. Three divisions are aligned with the key scientific disciplines of mobility (propulsion, autonomous systems, and mechanics). The purpose of the fourth division—the Vehicle Applied Research Division (VARD)—is to provide notional concepts for the entire spectrum of Army vehicles. These notional concepts guide the Directorate’s investment decisions by ensuring that all technologies required for a class of vehicles are covered and by defining key technologies that enable several types of vehicles. The June 25-27, 2012, meeting was the first review of VTD’s new building and research laboratories called for by the 2005 Base Realignment and Closure Commission (BRAC). The review consisted of overviews given by management, presentations on a subset of current projects, and poster sessions at which project leaders were available for discussion.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 82
6 Vehicle Technology Directorate and Autonomous Systems Enterprise INTRODUCTION The Panel on Air and Ground Vehicle Technology (AGVT), along with selected members of the P ­ anels on Survivability and Lethality Analysis, Sensors and Electron Devices, Armor and Arma- ments, Digitization and Communication Sciences, and Soldier Systems, reviewed the Army Research L ­ aboratory’s (ARL’s) Autonomous Systems Enterprise at the ARL facilities at Aberdeen, Maryland, on July 11-13, 2011. The Robotics Collaborative Technology Alliance (CTA) and the Micro Autonomous Systems and Technology (MAST) CTA are integral parts of ARL’s Autonomous Systems Enterprise. The Vehicle Technology Directorate (VTD) was reviewed by the Panel on Air and Ground Vehicle Technol- ogy at its meeting on June 25-27, 2012. The Directorate has four divisions. Three divisions are aligned with the key scientific disciplines of mobility (propulsion, autonomous systems, and mechanics). The purpose of the fourth division—the Vehicle Applied Research Division (VARD)—is to provide notional concepts for the entire spectrum of Army vehicles. These notional concepts guide the Directorate’s investment decisions by ensuring that all technologies required for a class of vehicles are covered and by defining key technologies that enable several types of vehicles. The June 25-27, 2012, meeting was the first review of VTD’s new building and research laboratories called for by the 2005 Base Realignment and Closure Commission (BRAC). The review consisted of overviews given by management, presentations on a subset of current projects, and poster sessions at which project leaders were available for discussion. 82

OCR for page 82
VEHICLE TECHNOLOGY DIRECTORATE AND AUTONOMOUS SYSTEMS ENTERPRISE 83 VEHICLE TECHNOLOGY DIRECTORATE Changes Since the Previous Review Massive changes in VTD’s manpower, research portfolio, and location have continued to occur since the 2009-2010 ARLTAB report was published. Perhaps most significant are a complete change in VTD leadership and completion of the new VTD building at Aberdeen. The leadership change, coupled with the fact that a high percentage of VTD researchers have been hired only within the past 4 years, has caused a disruption in many research activities. In response to an ARLTAB recommendation to maintain a systems focus as it instituted changes in location, person- nel, and research portfolio, VTD established the VARD. The VARD defined eight capability concepts vehicles that address Army objectives and critical capability needs (see Table 6.1). VTD began to align its research portfolio to address the technological requirements of the eight capability concept vehicles. However, the VARD appears to have lost focus on the eight capability concepts, and the capability concepts are not being utilized to guide the research portfolio. Good research is being conducted, but how this research fits into the VTD mission is not well understood. Therefore, VTD management should refocus attention on capability concepts as a systems methodology to align its research portfolio to meet critical Army needs and requirements. Construction at Aberdeen Proving Ground of a 35,500 square foot building to house VTD personnel and corresponding research laboratories as required by the 2005 BRAC has been largely completed. Relocation of VTD personnel to the new VTD building is complete, and the development-checkout of TABLE 6.1 VTD’s Mobility Capabilities Concepts Approach Capability Concept Army Objective Critical Capability Needs Persistent Staring Intelligence, Improved and persistent situational High-endurance VTOL Surveillance, and Reconnaissance awareness for military operations Autonomous operation Cargo Unmanned Aerial Systems Overcome sustainment shortfalls High-speed VTOL associated with current supply methods Autonomous operation Automated cargo handling Multirole/ISR Attack Vertical 6K/95 armed aerial escort with higher Variable-speed power/drive Takeoff and Landing (VTOL) speed/longer range than current fleet Adaptable rotor performance Long-Range Heavy Lift Mounted vertical maneuver into austere Large stable rotor environments Large efficient propulsion Lightweight durable structure Advanced Ground Combat Vehicle Improved survivability and mobility for Reliable efficient propulsion with Unmanned Ground Vehicle armored vehicles Armored robotic vehicle Wingman Terrain Adaptable Tactical Wheeled Tactical transport with robust mobility in Reconfigurable suspension Vehicle austere terrain Advanced high power diesel Small Dexterous Robots Soldier tasks performed at a safe stand- Higher levels of autonomy off distance Dexterous manipulation Micro Autonomous Systems Tactical situational awareness Low-power mobility Distributed autonomous operations SOURCE: National Research Council. 2011. 2009-2010 Assessment of the Army Research Laboratory. Washington, D.C.: The National Academies Press. Page 71.

OCR for page 82
84 2011–2012 ASSESSMENT OF THE ARMY RESEARCH LABORATORY the new research laboratories is ongoing. A memorandum of understanding between VTD and NASA remains in place for the 11 VTD staff that will remain at the NASA Glenn or NASA Langley facilities. VTD leadership has continued to transition the research portfolio from a focus on large helicopters to a focus on the entire spectrum of Army ground and air vehicles. This transition has required the hir- ing of many new researchers and the establishment of many new research areas. There is evidence that a lack of good senior mentoring of these new researchers has limited the usefulness of their research. The two CTAs that have had continuity of leadership are in good shape relative to the rest of VTD research portfolio. Accomplishments and Advancements VTD’s most significant accomplishments over the past 2 years can be found in the areas of the Robotics and MAST CTAs and the start-up and checkout of the new VTD laboratory areas. Small Engine Altitude Performance and Heat Engine Systems Altitude Test Facility This state-of-the-art facility is unique relative to those of other government laboratories. Overall, the facility is well designed and provides diagnostic capability for testing a variety of engine types and environmental conditions. The facility has the capability to simulate sea level to 25,000 feet, provide air from –40 to +130 degrees Fahrenheit, and has two dynamometers, which together cover a span from 1 to 250 horsepower (hp). The current work is focused on small engines of 40 hp or less under a subat- mospheric variable-pressure environment, where engine performance and components (e.g., bearings and seals) will be evaluated. The testing of a Wankel engine is planned for the future. Combined with computational analysis of engine performance with a gas turbine suite of commercial modeling software, this testing will provide VTD with the ability to test all types of Army engines. Planned testing of a Shadow unmanned aerial vehicle (UAV) engine in a French facility will provide stand-to-stand valida- tion of this facility. ARL should support the purchase of the needed ancillary equipment to support the stand. The goals of the experimental effort should be clearly defined to fit the VTD goals of developing advanced modeling and a complete fleet of engines that can operate on JP-8. Gear Box Transmission Test Facility Although the test equipment for this facility is not yet in place, its description suggests that this is a unique, much needed Army facility. A VTD presentation discussing the Army’s experience with poor efficiency and reliability of hydroid drives provides an example of why this facility is needed. High-Temperature Materials Development Facility This facility contains the standard burner rig type of equipment that can be found in many labora- tories across the United States. VTD has need for high-temperature materials in many if not all of its capability concepts. Although it should be involved in high-temperature materials research given the serious flaws in such work conducted by the NASA Glenn Center for VTD, VTD has determined that other work is of higher priority and therefore has appropriately postponed developing a high-temperature materials program.

OCR for page 82
VEHICLE TECHNOLOGY DIRECTORATE AND AUTONOMOUS SYSTEMS ENTERPRISE 85 Combustion Facility The Army’s objective of utilizing JP-8 fuel in all of its vehicles makes this facility a necessity for VTD. The utilization of JP-8 over the range of engines of interest to the Army is a formidable chal- lenge, the depth of which VTD may not fully understand. JP-8 combines the worst characteristics of diesel and gasoline. The high combustion chamber surface area relative to combustion volume makes it especially difficult to run Wankel engines and small diesel engines on JP-8. The utilization of small injectors and heated fuel has been unsuccessfully tried by others in the past. Perhaps VTD should also add spark assist and catalytic breakdown of the JP-8 to its portfolio to investigate this difficult problem. The current combustion facility allows for some interesting experiments; however, it should be modified to cover the entire range of sizes and combustion configurations of interest to the Army (e.g., Wankel combustion). In addition, the facility should be modified to measure temperature and velocity in the combustion process. Also required is analysis of what sprays are needed in what engines to obtain the required mixture-ratio distribution by the penetration of the air by the droplets. A clearer explanation is also needed about the relation of the current experimental facilities to these various configurations. In a similar manner, analysis should be performed to indicate where vaporization is sufficiently slow to control the overall combustion rate. VTD should consider how all of the concepts discussed, such as flash heating of the fuel, fit into an overall program aimed at delivering JP-8 combustion over a range of engine varieties and sizes. Tribology Facility The tribology facility contains many of the test rigs that can be found in laboratories across the United States; however, this facility clearly fits a VTD need. VTD should develop a fully defined program to address the Army’s needs in this facility. Low Speed Wind Tunnel Because many of the Army’s needs involve air and ground vehicles at relatively low speeds, this facility fills a VTD requirement. However, plans to test the hover flight conditions and slow forward flight of micro air vehicles in this wind tunnel are not likely to yield realistic results. The size of the test section will cause the downwash from the lift jets to be reticulated by the proximity of the tunnel walls. The utilization of a free jet may be required to accurately test the micro air vehicles. ARL should sup- port the purchase of velocimeter equipment necessary to fully check out and test vehicles in the wind tunnel. The turbulence level in the test section of the tunnel has not been measured. There is risk that the turbulence levels stated for the tunnel will not be obtained in the tunnel. The addition of upstream screens and honeycomb has been utilized by many people to lower wind tunnel test section turbulence. VTD management should obtain help from outside experts before undertaking such an effort. Continuous Trailing Edge Flap for Helicopter Rotor Blades Research This is a well-conceived and well-designed program using a novel approach—embedding a piezo- electric actuated biomorph in a rotor blade for the purpose of trailing edge deflection. This concept offers an alternative approach to discrete flaps, which are commonly researched for application to helicopter rotor blades. An advantage claimed is that the proposed flap structure is simpler than a discrete flap because there are no mechanical components and moving parts. Additionally, a continuous blade flap

OCR for page 82
86 2011–2012 ASSESSMENT OF THE ARMY RESEARCH LABORATORY structure is considered to be more efficient because there is no gap between the flap and the primary structure as found with discrete flaps. The system offers the opportunity to eliminate the swashplate as the means of primary flight control, thereby reducing helicopter weight, drag, and maintenance require- ments. The concept is being developed using structural and aeroelastic analyses for design and bench and wind tunnel tests for model validation. Researchers are challenged with designing a trailing edge with sufficient flexibility to obtain large trailing edge deformation, but with sufficient stiffness to retain structural integrity, that is, to prevent any undesirable aeroelastic instabilities. Ducted Rotor System Research This work consisted of using computational fluid dynamics to predict the effect of diffusing the slipstream of a ducted fan. The researcher observed that the thrust decreased as the flow was diffused, rather than increased as expected, and the experiment was considered a failure. In fact, this result should have been expected, because the static thrust of ducted fans is increased by converging rather than diffus- ing the flow. This is another case when senior mentors would have been of great help to the researcher. Opportunities and Challenges The most critical challenge facing VTD is a complete lack of a plan for its research portfolio to meet critical Army needs. Without this overall plan the new research facilities do not have focus, and the research by the talented new researchers will not produce the critical technologies needed by the Army. Elements of this lack of a VTD plan have been evident for several years; however, the move to the new building with the new research facilities, the complete change in management, and the large number of new researchers make the lack of a plan critical at this time. Capability concepts are a tool to help with this effort, and it is extremely disappointing that VTD seems not to be using these concepts. The elements of the plan should start with critical Army needs and timelines. These needs should then be mapped into programs for each of the new research facilities, and a complete review of the VTD portfolio of research should be performed to determine what areas need added emphasis, what research areas are missing, and what research areas need to be stopped. The quality of VTD research is in decline and will continue to decline until an overall plan with timelines to meet Army critical needs is put in place. It is imperative that VTD’s management put in place within the next year an overall plan with timelines. New researchers are working with limited mentorship from new management and senior research- ers. An example of this lack of senior mentorship can be found in the work on physics of failure from multi-dynamic axial excitation. This research dealt with the physics of multi-axis loading and explored the advantages to doing vibration/dynamic load testing in a uni-axial superposition method versus multi- axial loading. The research was both good and valuable, although the results were not well understood by the researcher. The research showed vastly decreased time to failure in multi-axial vibration testing. However, it also confirmed that uni-axial results could be accurately superimposed on each other to get a similar result. The researchers did not seem to grasp that the results showed that if one is not going to go through the modeling and stress superposition exercise for uni-axial testing, then the only way to get an accurate time to failure is via a multi-axial testing technique. Essentially what was demonstrated was that if one would take the time to model the superposition of stress generated by uni-axial excitations, there would be little advantage to doing the testing in a multi-axial manner. All of these are good and useful results; however, the researcher’s conclusion that the uni-axial superposition method would not work is incorrect. This example, along with several others, spurs the suggestion that VTD management

OCR for page 82
VEHICLE TECHNOLOGY DIRECTORATE AND AUTONOMOUS SYSTEMS ENTERPRISE 87 should utilize workshops and consultants to obtain the help of outside specialists who can mentor its array of new researchers. The establishment of the VARD was the first important step to ensure that VTD focus on critical Army needs. In order to meet the needs of the Army, VTD needs to ensure that all the technologies required to support each of the Army’s spectrum of vehicles are contained in VTD’s portfolio of research or that VTD knows where it being conducted elsewhere in the United States. A review of the needed technologies would help VTD management to identify highly important technologies that support more than one capability concept and technologies in their portfolio that do not support any capability concept. Then the challenge would be to ensure that each researcher can relate how his/her research supports one or more of the capability concepts. VTD management should have VARD develop or adopt a complete set of capability concepts to meet the Army’s spectrum of critical needs and should conduct a review of its research portfolio vis-a-vis the needs of these capability concepts. Laboratory support equipment is required to support a state-of-the-art vehicle laboratory. The new VTD research facilities have the potential to compose a state-of-the-art vehicle technology laboratory. However, a large group of support equipment is needed to meet the requirements of each of the major facilities. Examples of this type of laboratory support equipment are a laser velocimeter, hardness testers, high- and low-cycle fatigue testing equipment, microscopes, and boroscope equipment. ARL manage- ment should ensure that the required support equipment is purchased. Several examples demonstrate that VTD lacks ownership of relevant Army problems. For example, a researcher working on the wear of hydroid drives focused only on the piece of the problem for which he has expertise. However, the real problem may be the surface finish of the gears or the windage of grease in the gear sets. Before launching a research effort, VTD management needs to ensure that the most important aspects of the problem are being addressed. Once again, senior mentoring and leadership would help to address this problem, but making sure that researchers own an entire higher-level problem, not just a piece of the problem, would also help to eliminate this lack of ownership. Overall Technical Quality of the Work Basic VTD Research Quality The BRAC decision to consolidate VTD at Aberdeen, coupled with VTD management’s focus on Army needs, has increased the quality of the VTD research portfolio. In 2010 the establishment of eight capability concepts that embody the technical breakthroughs needed to meet critical future Army needs was a major step toward focusing and upgrading VTD research. However, failure during this review cycle to continue to align the VTD research portfolio with these capability concepts represents a major lost opportunity. Focus can be lost when an entirely new management team has the pressure of estab- lishing and moving into a new facility; however, it is critical that VTD management quickly refocus on the technologies needed by the capability concepts. For example, VTD recognized in 2010 that a crawling bug type vehicle should be added to the micro autonomous vehicle portfolio of research, but 2 years later it has not been added. Similarly, combustion of JP-8 in the small volume of small engines is an example of a crosscutting technology area that would impact several capability concepts and is therefore a high-priority research area. While a new combustion facility is part of the new VTD, build- ing modifications to the facility will be required to support critical Army needs in this very important area. This also represents another area of lost opportunity. Without an overall VTD plan to meet critical Army needs, the overall quality of VTD’s research will continue to degrade.

OCR for page 82
88 2011–2012 ASSESSMENT OF THE ARMY RESEARCH LABORATORY Quality of Individual Research Projects The project on a continuous trailing edge flap for helicopter rotors evinced the highest quality among those presented and is of the caliber of the best in its field. Aspects of other projects, such as physics of failure from multi-dynamic excitation, separated flowing using nano-second pulsed plasma, and ducted rotor research, are important, but overall those projects are not of the highest caliber. Of inadequate qual- ity are research projects into material damage precursors in composite structure, variable speed power turbine research, compressive sensing robust recovery of sparse mechanical signals from incomplete measurements, and slowed rotor unpowered take-off. Autonomous Systems Enterprise A detailed discussion of the Autonomous Systems Enterprise is presented later in this chapter. Many, but not all, of the research projects in autonomous systems are of the highest caliber; the combined quality of the research contained in the CTAs is cutting edge. AUTONOMOUS SYSTEMS ENTERPRISE This section first provides an overall assessment of the Autonomous Systems Enterprise and then highlights the four subject areas reviewed—microelectronics and perception, intelligence, human-robot interaction, and manipulation and mobility. Overall Assessment Accomplishments and Advancements ARL should be complimented for its utilization of CTAs that have allowed ARL to connect with and leverage world-class research in the area of robotics. These CTA activities are unique in the history of government laboratories, and without them ARL would not have been able to pull outside expertise into ARL. Perhaps ARL’s most important contribution has been the use of the Micro Autonomous Systems and Technology (MAST) and Robotics CTAs to maximally leverage high-caliber research and talent across the United States, which is a significant accomplishment. Another benefit is the sense of com- munity and excitement about robotics of many bright early-career researchers involved in the programs. Opportunities and Challenges In terms of balance, the research portfolio covers the over-the-horizon area very well. More attention should be given to near-term accomplishments. In a similar manner, the portfolio is too heavily weighted toward analysis; more experimentation would provide a better balance. ARL should utilize two systems-oriented strategies to connect near-term needs to far-term research: (1) a road map of specific objectives, tasks, and challenges to reach the far-term goals and (2) off-ramps from this road map to achieve mid-term results. One example of these off-ramps is the TARDEC Appli- qué project, in which the Army’s ground fleet was converted from being manned to being optionally manned. Enterprise management did not present an integrated long-term plan (road map) showing how the individual research projects that make up the ARL Autonomous Systems Enterprise fit together, or how

OCR for page 82
VEHICLE TECHNOLOGY DIRECTORATE AND AUTONOMOUS SYSTEMS ENTERPRISE 89 and when ARL plans to transition this research into demonstration and/or development projects. Con- sequently, some of the research tasks have little or no connection to other research tasks taking place under the rubric of the enterprise. ARL should develop an integrated long-term plan (road map) that places additional emphasis on technology demonstrations. ARL’s role in developing the necessary 6.1- and 6.2-type technology for the future is extremely important to the success of Army systems. Alignment of the ARL robotics enterprise and overall Army strategic plans is not clear. Also not clear are whether the various ARL technology developments can be categorized as unique, essential, or relevant to the Army and how those developments position ARL and the Army for the near, mid, and long term. The following questions need to be answered: What gaps exist between the civilian and military markets? What can the Army leverage? What is missing? What will ARL provide? What will collaborators provide? Improvised explosive devices (IEDs) were stated as one of the main foci of the enterprise, but no significant discussion of IEDs was presented. The ARL process faces a difficult challenge in grouping technologies according to near-, mid-, and long-term needs. A second and perhaps more difficult challenge exists in ensuring that adequate fund- ing and a management structure are present to achieve cross-technology, integrated solutions. A third difficult problem relates to the application of revolutionary breakthrough technology solutions, arising from inventiveness of scientists and engineers, which cannot be planned. Reducing the tendency toward stovepipe solutions is a frequent challenge for scientific and technical organizations. ARL should balance its efforts across disciplines, ensure that resources are moved when technology transitions, and maintain strong participation in the overall effort. Long-term requirements should be addressed primarily by 6.1 efforts, which should reflect the need for both capability-linked programs and novel concepts or breathtaking technologies. These programs do not lend themselves to rigid milestones, and allocation of resources has to be based on opportunity, scientific judgment, and a vision of potential applicability to Army systems. ARL should determine what sharing is occurring, among whom, and what joint investments are being made. Commercial off-the-shelf robotic systems represent quantitative benchmarks against which the ARL should always measure its programs. Because some of the currently available systems are competitors with respect to what ARL is attempting to develop, at a minimum the option to develop or buy should be evaluated. Needed are metrics of success or failure for projects transitioned, stopped/divested, and ongoing. ARL did not present research efforts related to the deployment of remote weapon systems on unmanned air or ground platforms; yet the Army has a need for armed robots that can effectively engage threat combatants and other targets to help improve soldier survivability. The expected future Army employment of armed robots suggests a number of new research questions that ARL should consider addressing within this enterprise. Examples of relevant research topics include the following: • What types of robotic weapons are most suitable for deployment from each class of unmanned air and ground platform; • What issues are associated with firing a robotic weapon system on the move (and also from a stationary posture) and with being able to effectively engage a stationary or moving/fleeting target; • Whether robotic weapon systems should have an autotracker, and if so, what capabilities the autotrackers should have; • What is the maximum soldier-to-weapon and weapon-to-soldier communication latency that can be tolerated under various engagement conditions, and how small latencies can be achieved with wireless networked communications; and

OCR for page 82
90 2011–2012 ASSESSMENT OF THE ARMY RESEARCH LABORATORY • What capabilities and features will future autonomous robotic weapons need and what new research will help to realize these capabilities. Three materials-related research tasks were described in association with the description of the p ­ roject on next-generation actuators and materials. These three tasks were apparently the only ARL materials research related to unmanned systems. The Army has a systemic problem with materiel weight, and unmanned systems are expected to be both weight- and size-challenged, because they will need to be transported to and from each theater of operation and some of these unmanned systems will also need to be carried into battle by soldiers. ARL has an opportunity to plan and undertake new research tasks focused on weight- and size-reduction of unmanned systems. Such research could potentially investigate new types of lightweight materials, structures, and components for future unmanned systems. The following activities were not identified in the presentation of the set of ARL’s autonomous systems enterprise activities and therefore appear to be missing from those activities; these should be considered: • Counter-robot activity. If the enemy is facing robots, then what actions will it take to counter the robotic threat?—for example, bat nets placed across windows and doors to trap flying robots. ARL should consider research approaches relevant to counter-robot activity. • IED sensors and activity. Although almost all researchers spoke of IEDs, the research portfolio contains no research on IEDs. A large task force, the Joint IED Defeat Organization, is address- ing the problem associated with IEDs, but ARL also should have some of its portfolio aimed at these problems. • Energy. Having soldiers carry 10 lb batteries for 15 minutes of robotic operation is not good; combustion of JP-8 in very small engines allows a factor of 100 more power density than sys- tems powered by electric batteries. Ways to achieve combustion of JP-8 in very small engines should be part of ARL’s portfolio. • Transfer of power from source to ground. The efficiency of robots in transferring energy of the power source to the ground is still two orders of magnitude less than that of biological systems. Work in this area should be included in the portfolio. In terms of impact on Army programs, ARL should consider the need to capture lessons learned with respect to IEDs, understand the ideal distance of soldier-robot separation, and identify what can be learned about UAVs that is applicable to unmanned ground vehicles (UGVs). Overall Technical Quality of the Work Many, but not all, of the research projects in autonomous systems are of the highest caliber; the combined quality of the research contained in the CTAs is cutting edge. ARL has organized a signifi- cant effort to leverage ongoing research across the United States. The overall technical quality of the work is very high for each of the key areas addressed: microelectronics, sensing, signal processing, and perception; intelligence; human-robot interaction (HRI); and manipulation and mobility. Specifically, the scientific quality of the research is comparable to that executed at federal, university, and/or national laboratories both nationally and internationally. The overall research reflects a broad understanding of the underlying science and research conducted abroad. The research team’s qualifications are very good. The facilities and laboratory equipment are state-of-the-art, and appropriate laboratory equipment and models are used.

OCR for page 82
VEHICLE TECHNOLOGY DIRECTORATE AND AUTONOMOUS SYSTEMS ENTERPRISE 91 The CTA exhibited a good mix of Army, university, and industry involvement. However, coordinat- ing the research at different locations and enabling collaboration among members of the alliance can be challenging. The following can improve the quality of the research: • A systems approach is needed. This is ARL’s responsibility, not that of the academic researchers or industry partners. • An element of the systems approach is termed “capability concepts.” For example, the descrip- tion of the MAST goals almost completely reflects a capability concept. That is, the MAST robot will weigh 200 grams, be utilized in the last 100 meters, and work for three different scenarios. To complete the MAST capability concept, time to complete the mission, reliability (in terms of getting there and back without breakdown and in terms of finding men and bombs when there), and availability of the robot to go on the mission should be added to the concept definition. • Reliability and availability goals and demonstration for robots are keys to the issue of soldier trust in the robot; that is, to design robots that are trusted by humans, reliability and availability goals should be added to the research. • Another example of the systems approach is the development of technology road maps tied to military need, identifying what is available now and what is needed, by when, in order to fill operational gaps. Greater emphasis should be placed on developing an overall robotics road map that defines the long-term Army needs and objectives and, therefore, individual program requirements. The road map should specify who will conduct the research and where it will be conducted.  • Another systems tool is the use of benchmarks. Benchmarking was illustrated with the demon- stration of the PackBot robot, which weighs 47 lbs and uses 10 lbs of batteries in 15 minutes of operation. In this example 60 people ran the robot through one of the MAST scenarios. The MAST goals are factors of 10 or more greater than the benchmark, so how the PackBot capabil- ity will improve with time is an important question to consider. • The MAST joint experiments are examples of system testing. However, they should be concen- trated on demonstration tied to the MAST capability concept goals. The results of the experi- mentation should be crafted into a clear statement of where the system is relative to its capability concept goals. The Robotics CTA should adapt the joint experiment system for its program. • The plan for how robots will go to war should guide research. For example, as robots get smaller they lose capability, and therefore small robots will most likely be used in swarms. At the start of WW II, the German strategy of how to use tanks was more important than the capability of the tanks themselves. • Many of the individual efforts are aimed at achieving the same objective but are poorly coordi- nated. For example, the robotic mapping or world view is a difficult problem that warrants more than one approach. However, each researcher should be aware of and attempt to leverage the efforts of other researchers in the same space. In addition, to conserve resources, ARL should identify and select the most successful approaches. Microelectronics and Perception Overall, the ARL research portfolio does a good job of covering the size range of robots. Fully autonomous robotic operation may take 30 years to fully develop. Clearly as robots get smaller their ability to carry sensors and support large computing activity decreases; therefore, demonstrating fully autonomous operation on large robots and then moving to smaller ones is recommended. The MAST

OCR for page 82
92 2011–2012 ASSESSMENT OF THE ARMY RESEARCH LABORATORY program is doing best-in-class work at reducing sensor size and weight. Off-ramps to divert developed technology into existing platforms should receive more emphasis. Also, scalability over robot platforms seems to be a major concern. There is a need to address scalability challenges with regard to capabilities (e.g., fidelity of the sensors, range and detection). The selection of sensors to accomplish the perception element of an autonomous system should be driven by the task and mission to be accomplished and the objects, activities, and events that one is trying to find and characterize. Bounding the problem in this way leads to identification of what can be observed and a selection of the sensor suite and measurements associated with the set of observables. There was little discussion of the observables needed and the justification for the selection of sensors and associated processing. Micro Autonomous Systems and Technology The goals of the MAST program include reducing the sensor size, weight, and power by a factor of 100. The hair inertial sensor is projected to be 1.5 mm × 1.5 mm, which is approximately 4 times smaller than the conventional microelectromechanical system (MEMS) accelerometer. The millimeter wave (MMW) radar work provides an approximate 100 times reduction in size, weight, and power with improved performance; the research resulting in this revolutionary technology development is of the highest caliber. An opportunity and challenge is to characterize and measure the effective range of this device for military utility. Once this is taken into account, the impact of the device can be accurately assessed. It is also important to illustrate more clearly how the work fits into an overall roadmap. The necessary next step in the maturation of the work is a technology demonstration as part of the future joint experiment to validate the utility of this technology. This work needs a comparison to the state of the art. Signal Processing The presentation on super-resolution processing described a technique for super-resolving three- dimensional (3D) range images acquired using a flash LADAR sensor for robotic applications in an urban environment. Work was done in collaboration with Carnegie Mellon University. Work was performed on the SwissRanger SR-3000 flash LADAR device, with a pixel count of 176 × 144 and 850 nm light modulated at 20 MHz. The system acquires range images at a maximum rate of 50 frames/second and has a field of view of 47.5 × 39.6 degrees. ARL’s super-resolution algorithm was used on the commercial off-the-shelf device to illustrate the benefit of super-resolution for flash LADAR imagery. For this work, the algorithm was operational in post-processing mode; however, it can potentially be implemented to run in near-real time on a robotics platform. The illustrative example of the operation of the algorithms is overly simple. The purpose of perception tasks is to reason about what is in the environment and what things in the environment are doing. Although the team has made progress in these areas, clarification of the extent of progress would be assisted by clearer articulation of the state of the art. Semantic understanding of static areas, particularly terrain and object classification, works very well over large data sets. Semantic understanding of dynamic areas (e.g., activity recognition) works reasonably well on small data sets and on a restricted set of activities. Work on distributed and collaborative perception (multiple robots, robots and people) is in progress.

OCR for page 82
VEHICLE TECHNOLOGY DIRECTORATE AND AUTONOMOUS SYSTEMS ENTERPRISE 93 The team should consider active use of multiple sensor input. This would involve joint simultaneous training and classification with multiple sensor streams, not just fusing best outputs from classification on individual sensor streams. The team should consider testing on data sets that the learning algorithm was not exposed to. Although it is common to train on one part of the data and test on other parts of the data, it is not so common to train on one data set and test on an entirely new data set, which would add value to the work. If possible, the team should consider acquiring data from the field—perhaps utiliz- ing video, LADAR, and radar footage from deployments in Iraq and Afghanistan. Future effort should address processing in real life for real-time decisions. There is a need to explore work being done by other organizations, such as the Air Force Research Laboratory, the Defense Advanced Research Projects Agency (DARPA), and industrial laboratories. Semantic Perception The purpose of this effort is to reason about what is in the environment—for example, whether a group of pixels is a car or a window—and what things in the environment are doing—for example, moving in a threatening way. If successful, then the system should be able to perform these tasks auto- matically, albeit with offline human supervision or input. The presentation clearly captured the technical barriers that make this work challenging. Complex interpretation leads to computationally intractable (optimization) problems because perception is noisy; fully supervised training is unrealistic; and integration of non-sensory sources of information (e.g., domain knowledge) is difficult (though clearly important). The presentation described, at a high level, the technical approach as a focus on objects and activities relevant to robots and soldiers—not general objects and activities—using contextual cues (e.g., external information and domain-specific information). The work aims to develop new learning and optimization techniques to make perception problems tractable. At present, the problems are intractable even when the focus is restricted to objects and activities relevant to robots and soldiers, and when context is built in. The presentation provided a clear evaluation strategy for reviewing the state of the art and setting and achieving specific quantitative goals for the program. The goals include standardizing metrics and data sets (the data sets will be made public—a laudable goal) and producing publications. The concept of semantic perception connects machine visual perception that is largely driven by physical objects in the visual scene with knowledge (i.e., domain knowledge, mission knowledge, and cultural information). This connection represents an advance because it allows context to drive expec- tations and reduce the space of possibilities produced by bottom-up visual recognition of edges and vertices. The team has made clear progress in all areas. Although sensing was not addressed in the presenta- tion, the following topics were addressed in some detail: • Semantic understanding of static areas: terrain and object classification, which works very well over large data sets; • Semantic understanding of dynamic areas: activity recognition, which appears to work reason- ably well on small data sets and a restricted set of activities; and • Distributed and collaborative perception: multiple robots, and robots and people, which is a work in progress.

OCR for page 82
94 2011–2012 ASSESSMENT OF THE ARMY RESEARCH LABORATORY Intelligence Overall Assessment Robotic intelligence is defined as whether the robot knows its location, what path to choose, and what the surrounding threats and opportunities are. This key grouping of capabilities needs to be addressed by technologies that have to be developed for autonomous robotic operation. The presentation of this work showed slow but incremental progress in this area. Because this area is very challenging, ARL has appropriately been applying many approaches to finding solutions. However, many of these approaches are not fully coordinated with other approaches in the same area, and therefore researchers have not leveraged the lessons learned and emerging results of the other efforts. For example, it was mentioned that robots and warfighters do not actually collaborate, which makes the role of the warfighter in this project unclear. Nevertheless, some of the approaches are sufficiently developed for standardization—for example, mapping inside a building should now be standardized, and additional research should be aimed at object identification inside the building. ARL should lead a mapping effort to ensure that all areas are covered by the research and that maximum leverage is being obtained from the different approaches. Advancements and Accomplishments  The robotics enterprise is addressing some critical sub- problems whose solutions are necessary for increasing robot intelligence (e.g., mapping, path planning, machine learning, robot control, and architectures for cognition). The team has made several incremental advances, such as modeling the multi-robot patrol problem in a new way and using machine learning in a variety of ways to improve robot intelligence. The quadrotor control is very impressive; however, it relies on some strong simplifying assumptions (e.g., having perfect localization from a fully instrumented laboratory) that would make it very difficult to apply to real Army missions. It is difficult to point to particular advances that could change the game in terms of robot intelligence for Army applications, but this research is still solid, incremental work. Opportunities and Challenges  ARL should better define robot intelligence as it relates to warfighter needs. For example, what type of decision-making capability is needed? Relatively little was commu- nicated during the presentation about the types of intelligence and decision making that are required to achieve the vignettes that have been outlined. A better vision of what is needed for Army applications would help to focus the research in intelligence. In particular, ARL should identify the forms of robot intelligence that are uniquely required for military applications but are not addressed by the significant amount of civilian work being done on robot intelligence. For most of the research in intelligence, it is difficult to measure the accomplishments and progress. It is often not clear how the research fits into the state of the art or how the problems being researched directly affect an Army mission objective. More work can be done to benchmark the research, including the definition of short-, mid-, and long-term benchmarks of Army relevance. More solid demonstrations that are compared to benchmarks would support the case that relevant progress is being made. Also, the autonomous robot challenge is an excellent opportunity to demonstrate how to empirically measure progress. For example, when successful autonomous operation range increases from a few hundred yards to a few miles, measurable progress has been made. The researchers should better elucidate the fundamental limitations of the models being developed, as well as better address issues of uncertainty and robustness throughout all the research tasks. This comment has broad applications. For example, in the MAST CTA, several of the intelligence projects and integrated demonstration did not acknowledge the size, weight, and power constraints or the sensing

OCR for page 82
VEHICLE TECHNOLOGY DIRECTORATE AND AUTONOMOUS SYSTEMS ENTERPRISE 95 and computational limitations inherent in the MAST robot vision. All research tasks in the MAST CTA should directly address these issues. Possibly missing from this research portfolio, especially for the MAST CTA, are power-aware com- puting, intelligent power management, and algorithms for managing bandwidth. In addition, some areas of important teaming research were not mentioned, such as the overall organization of the teaming (e.g., centralized versus decentralized intelligence across teams, hierarchical teaming, or other organizational strategies) and multi-robot decision making at the mission level. Many individual projects address the issue of robot mapping; however, these project teams are not talking with one another, which raises the following questions: What mission objectives are being addressed by these different mapping projects? Why not perform comparisons across mapping projects for a variety of Army applications? Why not study whether the key strengths of different mapping algo- rithms can be merged into a single approach? Why not study the tradeoffs for each type of mapping for various Army-relevant applications? For example, multiple projects are under way to autonomously map out the interior of buildings, and these projects appear to be making solid progress. Because most build- ings worldwide are of multiple stories, it is time to conduct a down-select of interior mapping systems and focus the resulting system on some stair-climbing approach such as was demonstrated so effectively in the autonomous stairwell assent project. It is very important to consider all stair-climbing concepts, such as tracked, legged, and wheeled platforms, in the overall program of autonomous mapping of the interior of multiple-storied constructions. The importance of instantaneous interior building mapping to the military mission remains somewhat dubious. How important would it be to map the inside of the building if it could be instantaneously and reliably determined that there were no combatants or other dangers inside it? Overall Technical Quality  The robotics enterprise is grappling with difficult problems that have been around for a long time. The enterprise’s approach is to pursue many different algorithms for these prob- lems, and then at some point in the future decide which is most beneficial for a given Army application. Research advances are being made to the state of the art, mostly of an incremental nature. Not all of the research approaches are well-justified (e.g., the hybrid mapping research, which pursues techniques based on the available sensor, rather than best approach to solving the problem). A better justification of the research approaches being pursued is needed in many areas of the intelligence research—particularly in terms of military requirements. Robot research in the civilian market continues to dwarf ARL’s efforts. ARL should strive to fill the military-relevant gaps left in the civilian research, not duplicate that research. Multi-Robot Persistent Surveillance Planning as a Vehicle Routing Problem This work addresses a long-existing problem using a new technique, that of multi-vehicle patrolling. The objective is to solve the problem exactly. Comparing the approach to a baseline of the traveling salesman solution is commendable. The ability to generate solutions that vary the visit times is a nice side-effect, because it reduces the external predictability of the system. This research should be pre- sented more carefully, recognizing that the approach is actually not generating a globally optimal result and that heuristic approaches can have value over exact solutions, specifically for solving large-sized problems in this probably intractable domain. Using this technique as a foundation for addressing the general task allocation problem with spatio-temporal constraints is an idea worth pursuing. Feedback should be added to the approach in order to close the loop on the control. There is a lack of knowledge regarding similar past and recent relevant activities conducted by NASA, DARPA, and other government organizations. Additionally, this work should better address the

OCR for page 82
96 2011–2012 ASSESSMENT OF THE ARMY RESEARCH LABORATORY issues of noise and robustness. References to the traveling salesman problem were refreshing, but if the power, weight, and energy challenges for a single robot are significant, then it would seem that the task of choreographing a team of robots for which we lack remotely relevant battery or energy technolo- gies is premature. These problems can become quite intractable with a moderate increase in problem dimensionality, and there is a need to explore super-computing platforms in this context, especially the parallel implementation of heuristic optimization algorithms to solve the larger scale problems. It is not clear whether a practical problem formulation is possible that would admit an exact solution as proposed by this preliminary work. Nevertheless, this has taken positive first steps to solving a difficult problem. Machine Learning for Robotics An overview presentation outlined the various ways in which machine learning is being used to improve robot performance. Using machine learning to tackle the challenging problems in robot intel- ligence is a good approach with many avenues for fruitful study. The advocated approach for control learning, which begins with a model for nominal control and then applies machine learning to optimize the control parameters, is an appropriate way of applying machine learning. Storing a database of expe- rience to assist machine learning is a useful way to gather data for machine learning, although much theory needs to be developed to determine how to generate and identify the right data for a given prob- lem. In the area of designing for learning ability, however, other than properly instrumenting the robot with sensors and monitors, the main ideas are not clear. The presentations lacked sufficiently concrete technical information on all of these learning topics. Because the civilian market is already doing a lot of work in this area, ARL should focus its efforts on the aspects of machine learning that are uniquely military in nature and not the focus of civilian efforts. Developing Hybrid Maps to Promote Common Ground This research attempts to develop topological maps in indoor environments that can be represented in human-understandable terms. To date, the main contribution has been the development of an online approach that works in cluttered environments. The approach involves analyzing point clouds from a Kinect sensor and decomposing the environment into regions and portals. Generating human-understandable maps has potential benefit to Army applications. However, using the Kinect sensor just because it is available is not an adequate justification for the selected approach. This presentation could have been strengthened if a better justification for the selected approach had been given. Old techniques that look at the ceiling to determine the likely 2D footprint of the walls are most likely a better starting point. As with the other mapping research, this work would also likely benefit from increased conversations among the various performers who are researching localization and mapping. The focus was on mapping Western structures and their contents—for example, opening doors, climbing stairs, identifying beds, tables, and chairs. In much of the third world, interior openings may be covered with a blanket or other hanging, ladders may be used to access other levels, and furniture may consist of cushions, mattresses, and the like. Even in modern buildings robots may be stymied by leaving the door open and hanging a blanket up in its place. Robots designed for Western architectures may also be stymied by hanging beads in windows and doors or by installing window screens. These problems should be addressed.

OCR for page 82
VEHICLE TECHNOLOGY DIRECTORATE AND AUTONOMOUS SYSTEMS ENTERPRISE 97 Joint Experimentation of Distributed Mapping This presentation reported on experiments in multi-robot simultaneous localization and mapping (SLAM) in indoor environments, as part of the MAST CTA emphasis on cooperation. In this experiment, a team of ground robots used laser, camera, and communications links to map out an indoor area. Robots recruited help at intersection points. The novelty of this mapping approach lies in the use of graphical models for data association and the use of structures (e.g., doors, signs, and posters) as mapping land- marks. The use of high-level features also reduces the amount of required inter-robot communication. The resulting accuracy of the system was reported to be approximately 1 inch. This research is solid and demonstrates some important SLAM and multi-robot capabilities. However, the research approach does not consider the constraints specific to the intended small-scale MAST robots—that is, size, weight, and power—or limited sensing and computational constraints. ARL reported that the graphical model approach reduces the typical computational requirements of mapping from O(n2) to O(n lg n), which is an important contribution. However, the research needs to do a better job in general of explicitly addressing the size, weight, power, sensing, and computational constraints. The software developed for this demonstration requires robots at the capability level of the Robotics CTA, rather than the MAST CTA. Human-Robot Interaction Leveraging Human Cognition Most of the robot mobility and manipulation research leverages the mechanics of biological systems (e.g., insects, birds with flapping wings), which is not necessarily the case for robot cognition. However, human cognition is a hugely parallel system, and almost no ARL research addresses massively parallel processing. For example, in the area of perceptual expertise, detection of IEDs and similar threats can be quite complex, yet humans do these tasks, and some humans do them better than others. Why not examine human skill as we do the flapping wings of birds to reverse-engineer the human mechanisms? Another area of ARL robotics work, radar signatures, could also benefit from an understanding of the types of cues that human observers use to identify threat. On the positive side, a few presentations harnessed human mechanisms. In the human-robotic interaction group of the Robotics CTA the areas of robotic trust and teamwork are benefiting from an understanding of human trust and teamwork. The work on neurocognition envisions the use of EEG signals to identify anomalies in the environment to which an expert detector would attend. The work on neurocognitive application to robotics highlights a broad range of robotics applica- tions based on the use of physiological measurements. These applications range from improving robot performance based on recorded human responses to varying a soldier’s tasking based on assessment of fatigue stress or other mental state indicators. The robotics enterprise works closely with the neuro­ science group at ARL’s Human Research and Engineering Directorate (HRED) to focus the research and technology maturation in a direction that has a potential for transition to useful Army applications. Human-Systems Integration and Function Allocation Human-systems integration (HSI) is much more that HRI or human factors for robotics. HSI is sys- tems engineering with humans at the center. Timing is very important; one cannot adequately engineer a system or integrate the human in the middle or at the end of system development. Human consider-

OCR for page 82
98 2011–2012 ASSESSMENT OF THE ARMY RESEARCH LABORATORY ations need to occur up front. In the case of soldier systems, researchers need to ask very early in the conceptualization and design of equipment or systems whether the soldier will carry it, use it, like it, and need it, and whether the soldier’s mission will improve as a result of its use. This observation should be repeated in all projects. In the beginning, the question of function or task allocation also becomes important: What tasks are best allocated to the robot and what tasks are best left to the human? These decisions require input from the warfighter. In contrast, the enterprise seems to be pushing technology that the soldier is not asking for and does not describe, with supporting rationale, the functions and tasks that will be allocated to the robotic system and to the human. Although ARL investigators talk with warfighters, this interaction provides only rudimentary infor- mation, because the warfighter does not often have the breadth of technological and scientific knowledge to know what is possible or where weaknesses reside. The discipline of cognitive engineering possesses a body of tools and methodologies for cognitive task analysis that provide systematic approaches to eliciting this type of information (e.g., see Pew and Mavor1). It also is possible that near- and mid-term solutions may allocate more tasks to the human, strategically postponing allocation of the functions that are more challenging to automate. Some researchers reported that they talked with operators and warfighters as they developed the concept of the robot as co-combatant, but these discussions did not happen at the right level and were ad hoc and undocumented. If operators and warfighters are driving requirements, then interactions with them should be structured to the best extent possible. For example, showing a warfighter what ARL is doing with robotic learning differs from asking the warfighter what kinds of things a robot should assist with. The MAST CTA and the Robotics CTA should collaborate more, especially in the area of HRI. When this issue was raised during the presentations, the response was, in some cases, that HRI is being handled by the Robotics CTA, which implies that the HRI considerations happen in parallel to MAST research. However, these considerations should be better integrated from the beginning, or the end result may not be used by or be useful for the human. As is often the case in large organizations, there are stovepipes that should be better connected. Coordination and knowledge sharing should be improved within the CTAs and between the MAST CTA and the Robotics CTA. It may be beneficial to leverage the work or lessons learned from the previous Robotics CTA. Additionally, connections should be forged across CTAs and the ARL autonomous sys- tems enterprise, between CTAs and external industry and academic units, as well as across DoD. Common ground can be better achieved by providing a common data set to share within and outside of DoD. This common data set would provide ground truth (e.g., location of threats) and therefore could be used to compare alternative technologies. Expanding HRI Research ARL’s employment of simulation in its research is good and should be extended and leveraged. HRI’s role should be expanded beyond testing of swarming robots to cover all aspects and stages of robotic research. Systematic tools for doing HSI needs analysis should be used to drive definition of mission and scenarios. Just as biological systems are leveraged, ARL should take advantage of knowledge about human cognition in perception and intelligence applications. In the longer term, ARL should use more real robots and should consider testing at Fort Benning with intended user groups. 1Pew, R. W. and Mavor, A. S. (Eds.). 2007. Human-System Integration in the System Development Process: A New Look. Washington, D.C.: The National Academies Press.

OCR for page 82
VEHICLE TECHNOLOGY DIRECTORATE AND AUTONOMOUS SYSTEMS ENTERPRISE 99 HRI needs to be part of MAST research and design. There is good synergy between ARL and university researchers, but there should be more interaction, at ARL and within the CTAs, with HRI researchers in other military laboratories (e.g., the Naval Research Laboratory and the Office of Naval Research, Army Research Office) and academia. To better inform the researchers of ongoing work and needs, ARL should conduct more outreach to invite instruction from academic and military experts and to leverage their work. The long-term vision is for soldiers to have robotic teammates. In the mid-term, robots could be used as tools, with functions and tasks allocated according to supporting analyses of human and robot capabilities. Overall Technical Quality Work on robot reliability and on dynamics of human-robot teaming and RIVET2 simulation software is notable. The ARL programs have embraced the noble goal of utilizing robots as trusted team members rather than as tools, which is even more challenging than the goal of fully autonomous robotic operation. ARL’s Human Research and Engineering Directorate should be involved with and given early priority in all robotics programs. HRI individuals should collaborate with non-HRI groups as horizontal integrators. Manipulation and Mobility Advancements and Accomplishments The research portfolio includes devices at the macro-scale (Big Dog), mesoscale (Rex), and microscales (micro flyer). With respect to legged locomotion, Big Dog and Rex represent the state of the art; they are excellent examples of legged robotic systems. Legs represent capabilities that wheeled and track vehicles cannot deliver; adaptability and robustness to different terrains and scenarios are possible. This is a most appropriate undertaking for ARL. In the case of Rex, theoretical considerations were effectively combined with pragmatic issues in mobility—for example, the ability of a machine to climb a hill or a flight of stairs and the inherent dynamic stability of the platforms. Big Dog is robust with respect to disturbances, and its legs allow for coverage of complex terrain. ARL is performing high-caliber work on the inherently stealthy micro-flyers. The component-level research in terms of MEMS fabrication, combining actuators with structural functionality, is pushing the state of the art. This system is potentially an organic asset for soldiers for the last 100 meters. Model- ing of the flyers was performed to develop ideas about power required, but the details of those models were not provided during the presentation. In addition to component-level research, system issues were addressed to determine the potential range and payload of such a system. Work on flapping vehicles performed on micro-flyers at the University of Maryland using computational fluid dynamics should be applied to the ARL micro-flyer. The work on controllers for micro-flyers is laudable as well. The use of commercial off-the-shelf robots is appropriate, because it allows researchers to focus on the other research issues at hand. Use of state-of-the-art PackBots for some applications such as map- ping for more realistic environments might be helpful. Analysis, modeling, and experimentation are evident within the program, which is appropriate. For example, modeling of robotic motion, actuation, and platform dynamics needs to be performed to develop an understanding of how power is utilized and delivered in legged robotic applications. 2RIVET stands for robotic interactive visualization and exploitation technology.

OCR for page 82
100 2011–2012 ASSESSMENT OF THE ARMY RESEARCH LABORATORY Another good piece of work is the characterization of novel, highly compliant, dielectric and con- ductive materials, which have application in isolation or as dielectric elastic actuators and electronic components. Characterization and functionality extends through 100 percent strain. Opportunities and Challenges There is an opportunity to redefine locomotion for legged systems to increase their efficiency and robustness. Not enough resources are available to create new platforms, but the platforms under consid- eration can be improved, bringing these robots closer to utilization by the warfighter. The portfolio should be examined to identify how projects can contribute to concepts of optimal vehicles or vehicle features needed to meet application scenarios. Understanding the efficiencies of vehicles with respect to mission needs and considering vehicles that have not been traditionally studied (such as snakes or blimps) may result in a new paradigm in vehicle attributes that can be leveraged to surpass the defined technical gaps and hurdles. Specific gaps in power generation, power delivery, and countermeasure development should be included in this assessment. The researchers should more clearly articulate and justify their work—particularly with regard to scenarios for the Robotics CTA. The locomotion researchers should develop joint experiments with other researchers in the ARL, such as the HRI researchers. HRI’s vision could be better integrated into the manipulation and mobility projects. Some problems such as sorting of parts, path planning, and robotic arms are not well integrated into any scenarios or even into the Army mission. The issue of reliability should be quantified and captured as part of the research for these platforms. Systems analysis should be performed for more platforms for various scenarios and should consider energy source, power utilization, robotic counter measures, and payloads. This does not mean that ARL should assume responsibility for research to counter measures, but it raises the issue of the robustness of platforms. Fluid flow analysis should be applied to the micro flyers in order to inform understanding of how to control them. There is a lot to learn about these machines and their performance, and so the interplay of experiment and modeling should continue. Power for Robotic Vehicles  More work should focus on power for the robotic motion, specifically power source (engines and energy storage) and power delivery (actuation, rotation, and thrust). Because of its great advantages in energy density, power density, and the availability of JP-8 fuel, the use of combustion should be studied. New concepts could lead to smaller combustion engines useful for small robots and small unmanned aerial vehicles. In addition, smaller combustors on larger robots could lead to designs with distributed power throughout the robot, which should offer some efficiency advantage. The area of energy and power utilization presents opportunities for huge advancements. Humans and large dogs use about 100 watts of power on average. A horse is at roughly one order of magnitude greater in its average energy usage at 1 kilowatt, and an elephant uses about 10 kilowatts on average. Why is it necessary, then, for the DARPA/Boston Dynamics Big Dog Robot to require a 15-hp internal combustion engine, which had a little more power at its maximum than an elephant? The issue is that the animal, bird, or human can store energy in a chemical form that is distributed throughout the body near muscles and readily convertible to mechanical energy when needed. Compared to the Big Dog Robot, which uses a hydraulic actuation system, the animal has the equivalent of stored energy in accumulators distributed throughout the body. So, the animal can produce a burst of power without need of a huge engine, while today’s robot does not have the accumulator system and has to speed up its huge engine to get the needed power. A robot with a smaller engine running at nearly constant speed and with a well-

OCR for page 82
VEHICLE TECHNOLOGY DIRECTORATE AND AUTONOMOUS SYSTEMS ENTERPRISE 101 engineered system of distribution and storage of energy would be much more efficient. Distribution of several smaller engines would also be superior. To address the need for small engines that run on JP-8 fuel, ARL should review the recent successes at the Rochester Institute of Technology in the development of spark-ignited two-stroke heavy fuel engines. Flapping Flight  The experimental work being done on the flapping flight of nano-scale air vehicles is exceptional. However, the use of rotors and fans should also be investigated. Although some early research indicated that flapping flight may be more efficient than steady rotary motion at nano-scales, recent research does not support this idea. The power for steady motion of a rotor requires overcoming the induced drag and profile drag of the rotor. The power for flapping flight requires power for overcom- ing induced drag, profile drag, and the inertial power required to stop and reverse the flapping motion. Flapping may not be the best use of the limited power available on nano-scale air vehicles. Compar- ing flapping to rotary wing flight might be an appropriate research task that probably should be done, but it might be easier to devote a similar effort to developing rotary winged nano-scale air vehicles. Current work on the quad fan and maple seed concepts may provide a starting point for rotary wing nano-scale air vehicles. Another opportunity is the engineering of flapping systems; this involves not only their construction and testing, but also how to engineer these platforms. Because they are not fixed-wing aircraft, there is no coherent way to design these systems with respect to power, stability, and controls. Palm-Sized Aerial and Ground Platforms  The objective of this project is to provide the fundamental aeromechanics and ambulatory tools to enhance MAST objectives. The research involves very significant multi-university partnerships, with funding from multiple governmental agencies. Physics related to the vehicles are very different because of low Re and the dominance of the viscous effects. Traditional computational fluid dynamics codes do not have good predictive capabilities, and experimental results are not easy to replicate for the length scales under consideration. Flight vehicles are also highly suscep- tible to atmospheric disturbances and, therefore, call for new approaches to flight control. The approach consists of experimentation with concepts that loosely model winged flight in nature (e.g., small birds and insects). The focus is clearly on flapping-wing flight, although the lead investigator’s background in rotary wing flight has introduced innovative micro-air vehicles such as the quad rotor and the ducted fan rotor. The mass of the vehicles considered in the work ranges from 12 g to 100 g. A parallel effort to understand the mechanics of small ambulatory ground vehicles is less developed. The work’s promise lies in developing a deep understanding of flapping-wing aerodynamics and a much better appreciation of the engineering scaling laws that apply in this environment. The progress in this regard can at best be characterized as limited. The computational fluid dynamics models have shown some promise in qualitatively characterizing the experimentally observed flow patterns for hovering flight. Simple flat plate models have been developed to study the highly coupled aero-structural behavior; a flapping-wing rig has also been built to do fundamental flow measurements to augment the simplistic computational models. Similarly, interaction of flows between two flapping wings has been visualized in the tunnel. If successful, there is considerable value in the work to push the envelope on developing micro-air vehicles. To date, however, the success has been limited, and such micro aerial platforms will develop largely through a build-and-test approach with limited understanding of the physics. The flight of the cyclocopter is an example of such an approach. The principal impediments to microscale ground vehicle ambulation are not well understood. The current approach emulates the mechanics of motion of small insects; designing mechanisms at this length scale may be the goal of this work. Design of both

OCR for page 82
102 2011–2012 ASSESSMENT OF THE ARMY RESEARCH LABORATORY air and ground vehicles at this scale should consider power needs, and very little about this important consideration was presented. ARL Millimeter-Scale Aerial Platforms  This project’s objective is to provide the Army with a low- cost, low-observable, mobile sensor platform using PZT and MEMS technology. The study is focused on a feasibility analysis of millimeter-scale robotics. Key challenges in this work include developing and integrating appropriate power sources at this length scale, providing adequate load-bearing capacity for flight vehicles designed using MEMS production techniques, and appropriate framing of the mobil- ity design problem. The focus resides in studying flapping-wing propulsive devices to look at aerial vehicles ranging in mass from tens to hundreds of mg. There is significant dependence on the promise of thin-film battery technology for delivering the power needs for this project. The approach completely embraces the idea of build-and-fly test concepts, with a focus on MEMS production techniques. Very simplistic aerodynamic theories are used to predict the flight characteristics of these devices, and further improvement is promised as other portions of MAST provide the necessary tools for more refined analysis. To date, the work has resulted in millimeter-scale flapping-wing designs that have been demonstrated on static test rigs. At this length scale, this work represents the state of the art and is one of the most promising developments at ARL in this area. A significant amount of formal design work (including coupled multidisciplinary analysis and optimization, albeit with simplified analysis modules) has been performed in support of this project. Understandably, the work has been well received in the literature, as evidenced by both conference and archive journal papers. The lack of more robust aerodynamic calculations should not hold back this development. The requirements of mechanical power, however, could be a major impediment to success. Parallel developments at appropri- ate power electronics to get the voltages required for appropriate deformations in pitch and flap should be a continuing focus in this regard. Overall Technical Quality ARL is to be complimented for the range of robotics sizes and the different types of mobility devices in its robot research portfolio. Also, the research portfolio is well balanced in terms of analysis and physics-based modeling and experiments. It addresses real-world effects and has great focus on meeting a specific need. Metrics are fairly well defined, and the inherent requirement of a test vehicle drives the system thinking and approach. The work on the “micro flyers” and legged robotic systems are best-in-class. There was a good portfolio of vehicles and platforms from small scale to meso- and microscale. Specific efforts were judged to be leading the state of the art in their areas. Briefers are aware of and addressed the system perspective. A candidate area for increased emphasis is discrete awareness of mission, sensors, and power requirements to meet the application vision and scenario. The efficiency of existing robotic systems in transferring energy from the engine to the environment is still several orders of magnitude worse than biological devices, and therefore continued work in this area is required. Because of the burden imposed on soldiers by battery packs and robots’ limited time on mis- sion, research that improves the overall energy density and the efficiency of converting energy to motion is required. In particular, research on the combustion of small JP-8 combustion engines and research on the efficient creation and transfer of force to the environment should be added to the research portfolio.