WEARABLE COMPUTING

The Evolution of Army Wearable Computers

In 1989, the US Army envisioned a small wearable computer to assist soldiers with battlefield tasks. The concept has since grown from preliminary prototypes and a demonstration Soldier’s Computer into the current Land Warrior program and proposals for future systems.

earable computers will soon data transmission, image capture, integrated Global become a reality on the battle- Positioning System (GPS) receivers, and menu- field for frontline troops, under driven software. the US Army’s Land Warrior In 1990, Schoening and Zieniewicz teamed up program. Here, we trace the with John Flatt, Sal Barone, and Almon Gillette to Wevolution of Army wearable computers, from the ini- demonstrate an early surrogate system, the Soldier’s tial concept and first prototype, through downsiz- Computer, at the Army Material Command’s first ing and improvements, to future product directions. trade show in Aberdeen, Maryland (see Figure 1). We focus on two major programs central to the The Soldier’s Computer employed an Agilis brick- Army’s development of wearable type 386-based computer with an integrated packet Matthew J. Zieniewicz, Douglas C. computers: the Soldier Integrated radio system, which soldiers could load into their Johnson, Douglas C. Wong, and Protective Ensemble (SIPE) and backpacks. The system was relatively lightweight John D. Flatt the Land Warrior system. As the for the time, at approximately 10 pounds. It also Research, Development, and Engineering Land Warrior program nears included software for creating reports and display- Center, US Army Communications fruition, the Army continues to ing battlefield situation maps. Electronic Command advance the state of the art for In addition, a serial interface to an external GPS wearable battlefield computers. receiver let soldiers see their position on a map. The map was displayed on a ruggedized (metal case) hel- Early beginnings: The Soldier’s met-mounted quasi-VGA (720 × 280) display Computer (Reflection Technologies’ Private Eye display). It The history of Army wearable computers has its used a vibrating mirror and red LEDs to compose a roots in 1989 with James Schoening, a research ana- virtual 14-inch monochromatic (red-on-black) dis- lyst working at the US Army Communications Elec- play. Soldiers used a trackball for input and could tronics Command (CECOM), Research Develop- enter and transmit simple reports to other units. ment and Engineering Center (RDEC). (See the The system was a resounding success in demon- “Glossary” sidebar for terms used in this article.) strations to senior Army leaders and congressional Schoening envisioned a small wearable computer, staff members. integrated with a wireless link and helmet-mounted The next iteration of the Soldier’s Computer shifted display (HMD), that could help individual soldiers from a proprietary brick design to an open sys- on the frontline. Working with Matt Zieniewicz, tem–bus wearable design. The Natick Soldier Center Schoening transformed his idea into a system archi- in Massachusetts incorporated this concept as a tecture with targeted technologies, such as wireless key component of its SIPE Advanced-Technology

30 PERVASIVEcomputing US Government Work Not Protected by US Copyright Glossary

C4ISR Communications, command and control, computing, intelligence, Demonstration. The SIPE project, led by sensors, and reconnaissance Carol Fitzgerald, was the first time the Army CECOM Communications Electronics Command treated the various combat equipment com- HMD Helmet-mounted display ponents for the individual soldier as one IPT Integrated process teams integrated system rather than as a con- JCF AWE Joint Contingency Force Army Warfighting Experiment glomeration of individual components MDSE Mission Data Support Equipment (SIPE also included other advanced com- ORD Operational Requirements Document ponents in the areas of the fighting uniform, RDEC Research Development and Engineering Center load-bearing equipment, weaponry, and SIPE Soldier Integrated Protective Ensemble thermal imaging).1 TWS Thermal Weapon Sight The prototype design for the SIPE pro- WSS Weapon subsystem ject began in earnest in the spring of 1990. At that time, wearable computers were in their infancy. Steve Mann at MIT had pro- Figure 1. The Soldier’s Computer at the duced some early wearable computers,2 Army Material Command’s first trade show and during the summer of 1991, Carnegie in 1990. Note the small helmet-mounted Mellon University developed its VuMan VGA display. The visible cord is the VGA project,3 but the SIPE computer approach feed from the computer to the display. differed from the typical research project. The military still uses this monocular As part of a new digitized battlefield con- concept in an improved form. (The small cept, it aimed to implement desired battle- stub antenna for the integrated spread- field functions through technical means spectrum packet radio is not visible.) rather than explore an advanced technol- ogy and then develop an application for it. This key difference influenced the entire design process. Functionality and requirements The design team (see the “Soldier’s Com- Because this was the Army’s first attempt puter Design Team” sidebar) had to to bring computing devices to the individ- develop features (such as video capture) ual soldier, there were no preset system that could operate in a rugged environment. requirements, and users did not have spe- In simulated war-game exercises, actual sol- cific functions in mind. Initial brainstorming diers planned to test the system (10 proto- with the Infantry School—led by the sys- types) over several weeks in various out- portable display unit, preferably helmet tem’s software engineer, William Sanchez— door environments and during live-fire mounted. The time frame for developing developed key desired functions (listed in exercises. With this in mind from the out- the system was 24 months, with the last the next paragraph). At the time, none of set, the design team aimed to develop a three months reserved for field testing and the functions were commercially available portable, wearable battery-powered com- demonstrations. The budget for the com- in portable computers, but most were avail- puter with suitable battlefield applications puter-radio-GPS portion (exclusive of the able through various stand-alone electronic software. The computer needed to include helmet display unit) was US$500,000, or computer components. The challenge image capture, an integrated radio for including all labor, materials, software was to integrate these piecemeal compo- transmitting data between soldiers, and a development, and prototype construction. nents into a lightweight package that could

Soldier’s Computer Design Team

umerous engineers lent their support throughout the Sol- James Wright, project leader; Almon Gillette, packaging, electrical, N dier’s Computer effort, but certain key personnel ensured and mechanical interfaces; and Eric Hall, networking. In addition, the success of developing the computer-radio-GPS system. The Carl Klatsky provided valuable assistance during the final prototype core technical team members were Matt Zieniewicz, project construction and system checkout phase, and James Schoening leader, system architect, and video capture and compression spe- continued to work with the Infantry School on requirements; his cialist; William Sanchez, chief applications development software guidance and insight were essential throughout the project to engineer; John Flatt, networking and communications engineer; develop system concepts.

OCTOBER–DECEMBER 2002 PERVASIVEcomputing 31 WEARABLE COMPUTING

achieve the desired result without being too Engineering Laboratory, and Natick engi- minute intervals, along with digital reports bulky and cumbersome or requiring too neers and project leaders. Fortunately, the and captured still images, to a central gate- much power. The team decided early on to initial software’s functionality proved very way unit over an FM packet radio with a evaluate the best commercial components useful, and in fact, the Army later used it as range of up to one mile. At this fixed-gate- in each area (video capture, GPS, data com- the basis for the Land Warrior production way base station, messages were relayed munications, networking software, storage systems. (between two fixed, not mobile, stations) media, operating systems, programming to the Novell server over a wireless link languages, bus interfaces, and processor System architecture using a wireless LAN card. The soldiers boards) and then make trade-offs to arrive To satisfy the functionality required for used the FM radio because it offered an at the best possible system architecture. the Soldier’s Computer and its electronics increased range over a wireless LAN sys- They incorporated the functionally derived subsystem, the system team included the tem, and the packet mode better compen- hardware requirements in a custom hous- following key hardware components: a sated for intermittent connectivity. (LANs ing, developed within RDEC’s drafting, computer processor with memory, a GPS did not operate well under intermittent design, and fabrication division. receiver, a data radio, a video capture sys- conditions at that time, owing to the net- The new system aimed to digitize basic tem, a digital compass, a miniature color working technology’s limitations.) During battlefield operations to help soldiers camera, a video controller subsystem, an a data transmission, messages were relayed HMD, a power supply subsystem, wiring from the individual Soldier’s Computer to • Read maps, navigate, and maintain sit- harnesses, and packaging (for more infor- the gateway unit, to the server, and then uation awareness (so they could ask, for mation, see the “Hardware for the Soldier’s back to the gateway for transmission to the example, “Where am I, where are my Computer” sidebar). From a software per- appropriate Soldier’s Computer. Despite squad members, and in which direction spective, it was decided that it was best to the apparent multihop lag, soldiers did not am I heading?”) create one main application program that notice any degradation in service or time • Receive, prepare, and send written field could launch all the required functions delay. reports (so they could, for example, send through subprograms (see the “Software a call for fire or an operational order, or for the Soldier’s Computer” sidebar). On Feedback from soldiers prepare spot reports or Frago orders— the basis of this design approach, the pro- In the fall of 1992, the Soldier’s Com- puter was a key device demonstrated at By feeding the imagery from the bore-sighted Fort Benning, Georgia, as part of SIPE (see Figure 2). This was the Army’s first Thermal Weapon Sight (TWS) to the helmet attempt at “digitizing” the individual sol- dier, and the soldiers who used the system display, the soldiers could fire around corners or were impressed. The software functionality that the sys- out of foxholes. tem provided proved to be an asset under simulated battlefield scenarios. By feeding written military reports used by front- ject leader divided the software develop- the imagery from the bore-sighted Thermal line troops) ment into task areas and assigned them to Weapon Sight (TWS) to the helmet display, • Capture and transmit color still images appropriate specific project personnel. the soldiers could fire around corners or out for reconnaissance purposes The team then embarked on designing of foxholes, exposing only their hands and • Access battlefield operations reference the first custom Army wearable computer forearms to enemy fire. This feature received material (such as silhouettes of enemy to be demonstrated under field conditions. rave reviews from the user community. fighting vehicles, first-aid procedures, In effect, the team became a custom PC However, although the system enhanced common battlefield tasks, and standard clone manufacturer with a limited pro- the soldier’s fighting capability, it needed procedures) duction run. They carefully designed the to be more compact and operate longer on system by leveraging and integrating the a set of batteries before it would be battle- These functions were the basis for the latest hardware components and technol- field ready. More importantly, it needed to software application, developed in C. The ogy available and incorporating the best be lighter. The backpack-sized computer- team developed other functions as separate software practices, programming lan- radio-GPS unit weighed 18 pounds, and modules and included them in the main guages, and networking techniques. the HMD integrated into the fighting hel- program to provide modularity and ease of met tipped the scale at nearly eight pounds, testing. They also designed screens and Networking configuration with an additional 15 pounds for the high- screen layouts from scratch, using input The individual Soldier Computers sent voltage supply unit to drive the cathode ray from the Infantry School, the Army Human the soldiers’ current positions in one- tube-based display. Another drawback was

32 PERVASIVEcomputing http://computer.org/pervasive Figure 2. Testing and aligning the SIPE helmet display with the Soldier’s Computer in July 1992. The visor reduced ambient light and was a flip-up, flip- down display. It also provided ballistic and laser protection. The right-mounted sensor on the helmet’s top was an image intensifier for night vision capabilities. The large brown case is the computer- radio-GPS unit.

The next phase The Natick Soldier Center completed its SIPE project in two and a half years, and the Army’s Chief of Staff was enthusiastic about furthering efforts to field an integrated fight- ing system with a wearable computer-radio- GPS unit. The Army also continued explor- ing digitized components for the individual the delay in capturing and sending a still (9,600 bps), capturing and transmitting soldier under various programs. For exam- video image. Owing to the limited pro- images could take 45 to 75 seconds, during ple, the Twenty-First Century Land Warrior cessing speeds on the video capture board which time soldiers couldn’t use the sys- project examined advanced computing and and communications channel capacity tem for other operations. electronic products and concepts. Also, the

IEEE Security & Privacy

Ensure that your networks operate safely and provide critical services even in the face of attacks. Develop lasting security solutions, with this new peer-reviewed publication. Top security professionals in the field share information you can rely on: • Wireless Security • Securing the Enterprise Don’t run • Designing for Security the risk! • Infrastructure Security • Privacy Issues Be secure. • Legal Issues • Cybercrime Order your • Digital Rights Management charter • Intellectual Property Protection and Piracy subscription • The Security Profession today. • Education

http://computer.org/security WEARABLE COMPUTING

Hardware for the Soldier’s Computer

o achieve the design requirements of the Soldier’s Com- server. The device had approximately a one-mile range and the T puter, the design team integrated several hardware compo- added bonus of having a toggle device that could be used to switch nents (see Figure A). They aimed to develop one main application to voice mode. Thus, one device could provide the soldier with both program that would control all hardware components and periph- voice and data communications. erals—they didn’t want to launch the components from separate To let the frontline soldier capture a color still image and trans- applications. This restriction influenced their choice for many of mit it back to a commander, the team had to select a video cap- the components, and they selected devices that provided well- ture system and integrate and program the appropriate compo- defined APIs and C libraries. nents. The team selected a 16-bit ISA still-imagery video capture The central component was a 16-bit ISA-based single-board card that allowed a National Television System Committee video computer consisting of a 20-MHz Intel 386SX microprocessor, image to be gen-locked and overlaid over a VGA image (gen- with a 387 coprocessor and 16 Mbytes of RAM. The system used a locked is when two video signals are synchronized so they can be passive backplane architecture with an ISA bus structure. The sin- overlaid one on top of the other). An important feature of this gle-board computer plugged into the passive-backplane card. This card was that it provided both VGA and NTSC outputs. This was approach made the system modular and easy to upgrade, and, necessary to drive the helmet display, which required an NTSC most importantly, gave it the desired physical footprint. A mother- input for viewing video imagery. This overlay feature let the video board approach would have resulted in a large rectangular board image appear in a window within the overall application program, with expansion boards inserted at 90-degree angles. Employing a without taking up the full screen. Most video capture equipment passive backplane allowed the cards to be stacked longitudinally did not allow for such tight integration with other programs. alongside each other, resulting in a denser package overall. For the video sensor, the team selected a state-of-the-art cigar- sized daylight color camera, a Sony XC-999, for its resolution, porta- Location functionality bility, and ability to power the camera from a 12-volt source. (This To achieve the required location functionality, the design team camera was very advanced for its day and remained the camera of added an ISA-type Global Positioning System receiver card (devel- choice for many developers for years to come.) They also designed oped in 1990). They selected a NavStar model because an API an external plastic case to house the camera and digital compass. library existed to provide low-level interfaces to the GPS data. This The digital compass selected was from KVH, a digital compass man- device also provided the best accuracy at that time by providing the ufacturer, and allowed for custom programming interfaces through most channels. The antenna was a puck-type antenna mounted a serial port. The team installed two small toggle switches on the externally at the top of the soldier’s backpack frame. case to let the soldier capture an image through a freeze-frame technique. The signals were fed through a wiring harness, which Data transmission consisted of plastic-coated copper-stranded wire encompassed in The team incorporated a 2-Watt FM packet radio transmitter, cloth mesh tubing. Velcro fastened the video capture enclosure to also in a 16-bit ISA form factor, to achieve the required data trans- the soldier’s suit. mission. For similar reasons, they selected a model that allowed for The soldier captured the actual images by viewing a green-type integrated software control of this device from the main application monochrome live video display in his or her binocular helmet program. They developed a simple program interface to transmit display and pressing a small capture button to freeze the image. The the files to a gateway unit, which then relayed the files to a Novell digital compass, which also had an API and serial connection,

National Training Center-94 Soldier System a weapon system that, amongst many The Army leadership liked the SIPE sys- as well as Task Force XXI were large war things, could identify a soldier’s location, tem’s capabilities, so they incorporated game exercises conducted in the mid 1990s, his or her fellow troops, and the enemy. many of its functions into Land Warrior. in which frontline soldiers used a ruggedized First and foremost, the system aimed to However, they also added new functions portable computer in field exercises at Fort enhance a soldier’s ability to move, shoot, and tried to achieve a lighter, smaller, lower- Irwin, California, to effect command and communicate, and survive in modern war- powered, and more rugged system. Like control operations. fare. To achieve this, the Land Warrior any successful wearable computer or com- The Army’s main focus, however, was System relied on communications, com- puting system, Land Warrior had to be easy on producing an integrated fighting sys- mand and control, computing, intelli- to use, weigh almost nothing, work all day, tem. In 1993, it held a kick-off meeting to gence, sensor, and reconnaissance (C4ISR) and be comfortably placed and conve- initiate the development of Land Warrior, technologies. niently located.

34 PERVASIVEcomputing http://computer.org/pervasive High-voltage converter Helmet-display batteries tracked the direction the camera was Helmet Computer-radio-GPS aimed at the time of capture, as well as the soldier’s direction of travel for navi- Input system and control device gation purposes. The signals from both the camera and compass were fed through a cloth mesh tubing wiring harness. LEMO (a manufacturer of vari- ous specialty connectors) and military- type circular connectors were used on all external connections for strength and reliability under harsh conditions.

Interface To interface the VGA output from the computer and capture system to Figure A. The Soldier’s Computer. the monochrome binocular helmet- worn display that S-Tron designed, and to interface to a custom joystick controller, Dick Tuttle’s dis- was rectified with a ground strap, none of the 10 units had a play team from the Electronics Technology and Devices Labora- hard disk failure. tory of the Army Research Laboratory designed a custom input/output card. The synchronization signal had to be slightly Power sources modified to display properly in the helmet. This card also multi- Two nonrechargeable lithium batteries (BA-5590s) provided plexed the video signals from the weapon-mounted thermal three to five hours of operating time. The system’s average power weapon sight and the daylight color camera, directing the sig- consumption was 19.5 Watts for the computer, radio, GPS receiver, nals either to the video capture card or directly to the helmet. and daylight color camera. The helmet-mounted display was pow- The card also provided standard keystroke inputs from the cus- ered by its own separate battery and voltage inverter. The system tom joystick to the keyboard and mouse input ports on the sin- bus was powered by feeding the battery voltage to two separate gle-board computer. DC-DC converters with appropriate inline fuses. The entire system was enclosed in a 16″×9″ (with a 6-inch depth) aluminum chassis Storage with shielding material added to reduce electromagnetic interfer- To store the operating system, application program, captured ence with another voice radio that the soldier used. The system was still images, and maps, the team installed a 3.5-inch form factor mounted on a backpack frame and weighed approximately 18 40-Mbyte ruggedized hard disk along the case’s inside perime- pounds including the two batteries and external camera and com- ter. It could withstand 10G operating and 100G nonoperating pass case. (This did not include the helmet assembly and its power shock values. It had an initial grounding problem early in the source and inverter. The high-voltage inverter used with the display testing that resulted in some hard disk crashes, but once that added significant weight to the system.)

Developing system requirements it must demonstrate threshold values of next phase—material development—the In 1994, the Army began a formal key performance parameters listed in an Training and Doctrine Command System requirements process, quantifying battle- ORD in contractor and development test- Manager for Soldier Systems at Fort Ben- field functions and required operations in ing and operational testing. ning (a government program management a performance-based document known as For Land Warrior, the Infantry School at office) presented the user requirements to an Operational Requirements Document. Fort Benning, Georgia, provided the initial the program manager’s office, Program An ORD defines a desired system’s func- fighting doctrine as described in the ORD. Manager (PM) Soldier at Fort Belvoir, Vir- tions, operational capabilities, and perfor- The year-long process involved numerous ginia. (The Infantry School still reviews mance, quantifying many performance meetings with both users and technical changes made to these requirements.) parameters with both threshold and objec- experts, who reviewed, in detail, the require- After the Army documented the formal tive values. Before a system can be fielded, ments’ feasibility and applicability. For the system requirements in the Land Warrior

OCTOBER–DECEMBER 2002 PERVASIVEcomputing 35 WEARABLE COMPUTING

Software for the Soldier’s Computer

he software system employed the legendary Disk Operating could display a virtual keyboard to construct fragmentary or oper- T System, with a custom package developed in C with a win- ational orders when necessary. The video mode let a soldier see a dowing toolkit. This let the system emulate a Windows environment video feed from either the Thermal Weapon Sight or the daylight and let the user select software buttons using a joystick interface color camera in his or her monochrome (green on black) helmet that emulated a mouse. The main menu navigation bar at the top of display. The soldier could then choose to capture one of these the screen let the soldier select the different functions: mapping and images to send back to the base station. The system automatically navigating, sending and receiving reports, using video mode (for time- and date-stamped all images with the sending unit’s identity both capture and weapon firing), communicating, and accessing and logged the images into a video database. However, it first reference material. compressed the raw images into JPEG files, which could be trans- For mapping and navigation, soldiers could see both their own mitted in approximately 30 seconds. location, provided by their GPS receiver, and that of their fellow The reference material section consisted of several scanned images soldiers indicated as small icons on a scanned and registered map. of enemy fighting vehicles for field identification purposes along with The reports section let platoon leaders send and receive several information about the weaponry characteristics. Also included were basic battlefield reports. They could construct the reports through common field manuals, evacuation procedures, first-aid information, a series of pull-down menus, requiring very little typing. They range card procedures, and prisoner of war procedures.

ORD, the program manager developed a an obsolete system. The Army aimed to Furthermore, Land Warrior has been performance-based system specification, leverage mature and emerging technolo- incrementally built and tested using the rapid stating what the system should do but not gies, packaged for the warfighter’s envi- prototyping approach. Both the require- how it should do it (for example, the spec- ronment, to field a supportable weapon ments and specification evolved as the IPT ification might say “transmit reports” but system. However, while incorporating the learned lessons throughout the development not “transmit reports using an FM-based latest trends, the system still had to satisfy process. Early testing identified the prob- digital radio”). For interoperability rea- its ORD requirements and the constraints lems of obtaining adequate bandwidth and sons, interface standards were specified of the Army’s Joint Technical Architecture. range from the communication system, between components and for external con- In addition, the Land Warrior Inte- because Land Warrior requires transmitting nections to other systems. grated Process Teams (IPT) of government voice, data, and imagery within a squad. The PM Soldier Systems and Project Man- and contractor design engineers had to In Fall 1999, the Land Warrior team of ager, Soldier Electronics offices, under the make key design decisions on technical government and contractor engineers Program Executive Office, Soldier, were pri- standards, approaches, and tools used to started working on the first rugged design marily responsible for developing the Land build the devices. Such decisions had to of Land Warrior, Version 0.6. They aimed Warrior system. They had to write the sys- facilitate an open, modular, and flexible to present it at the Joint Contingency Force tem performance specification and contract technical architecture that suited the sol- Army Warfighting Experiment (JCF AWE) for developing the system. An SPS translates diers and could operate in their environ- in September 2000 (see Figure 3). This pre- operational requirements and other system ment, including under water, at extreme liminary effort used commercial off-the- constraints into system requirements and sys- temperatures, and under constant abuse. shelf and government-furnished compo- tem architecture. The Army awarded the At the same time, the system had to min- nents packaged to survive the soldier’s Land Warrior contract to a consortium of imize audible, radio frequency, infrared, environment. contractors, who worked with the govern- and visible emissions. So, the IPT had to ment to allocate requirements to the sub- ask design trade-off questions such as, The JCF AWE system level. The contractors performed rotating disk or semiconductor (flash) Soldiers equipped with the Land Warrior, detailed design, build, integration, and test memory? Infrared communications or Version 0.6 participated in three missions tasks to produce the system. Bluetooth? AMLCD (Active Matrix Liq- during the JCF AWE. The system provided uid Crystal Display), LOCS (Liquid Crys- a tremendous advantage to a platoon of Key design factors tal On Silicon), or OLED (Organic Light infantrymen from the 82nd Airborne Divi- A significant challenge facing Land War- Emitting Diode) display? Wireless, USB, sion (Fort Bragg, N.C.) in this field test rior was keeping pace with current tech- or FireWire? PCMCIA or RS-232 inter- against the conventionally equipped oppos- nology and implementing a modular faces? Centralized power or numerous ing force at the Joint Readiness Training replacement strategy to avoid maintaining batteries? Modular or integrated? Center in Fort Polk, Louisiana.

36 PERVASIVEcomputing http://computer.org/pervasive Figure 3. Land Warrior Version 0.6, September 2000.

The first mission was to parachute onto and secure an airfield at night. After reat- taching their HMDs and headsets and turning on the system, the soldiers could see their own location, where they were headed, and the location of their fellow troops overlaid on the assembly area map. Wireless voice and message communica- tion, previously not available to all soldiers, proved beneficial, and everyone reached the assembly area in record time. The second mission, which began at 2:30 am, was an assault on a village with several buildings (to simulate urban ter- rain) and enemy soldiers. The Land War- rior system automatically transmitted posi- tion reports for eight leaders in the platoon Helmet subsystem • Helmet-mounted display, speaker, and microphone to higher-echelon software systems. • Provides soldier audio and video interfaces The third mission was a night ambush. Soldier control unit and communication subsystem Land Warrior let the soldier’s view their • Provides system controls and soldier radio night vision image intensifier with one eye • Power on, smart card login, joystick, volume control, and their HMD with the other. (A rubber brightness control, and push-to-call • Soldier radio boot attached to the HMD eliminated • Communications processor detection by reducing light emissions.) Weapon subsystem PM Soldier learned numerous lessons • Weapon user input device, day video sight, thermal from these exercises. For example, the sys- sight, multifunctional laser, and compass • Provides the soldier with sensors and controls for aiming, tem’s disposable LiMnO2 batteries and target location, and target identification lower-energy Li-ion rechargeable batteries were too expensive and impractical, so the System power • One battery on each side of the soldier system needed different power sources. • Rechargeable or disposable smart batteries Other lessons learned included the need for (a) fewer cables with less exposure, improved reliability and ruggedness, and a reduction in electromagnetic interference. Overall, Computer subsystem however, the system performed well, • Manages system configuration, messages, and alerts • Stores standard map product, mission data, and manuals improved fighting capabilities, and im- • Generates map with graphical overlay of position and situation pressed the soldiers. Navigation subsystem Land Warrior, Version 1.0 • Provides GPS and magnetic heading After completing the JCF AWE, PM Sol- • Utilizes dead reckoning device when GPS signal is not present • Provides soldier location and heading to computer for map dier and the consortium of Land Warrior display, automatic position reporting, and target location contractors began to design the first field calculation version (Version 1.0), now called the Land Soldier equipment Warrior Initial Capability (see Figure 4). • Clothing, boots, gloves • Assault helmet • Modular lightweight load-bearing equipment, and ruck sack • Hydration system • Body armor Figure 4. Land Warrior Initial Capability, (b) Version 1.0: (a) front and (b) back.

OCTOBER–DECEMBER 2002 PERVASIVEcomputing 37 WEARABLE COMPUTING

Incorporating lessons learned from Land Warrior Version 0.6, which used a cen- tralized server, Land Warrior Version 1.0 uses a more distributed software archi- tecture. Also, the Land Warrior system does not contain enough storage for worldwide coverage of maps, and wire- less downloading of maps is problematic owing to bandwidth issues. PM Soldier thus developed a separate system, called the Mission Data Support Equipment, to Figure 5. Land Warrior’s computer help load mission data before a mission. Figure 7. Soldier control and subsystem. The MDSE consists of a laptop computer communications subsystem. and USB-to-Ethernet adapters. Its soft- ware includes the Mission Data package, Design rationale which lets soldiers organize unit tasks and pounds and consists of the computer assem- The IPTs incorporated the lessons learned create situation maps, help files, operation bly, flash memory, and video board, pack- and addressed other fighting and opera- orders, and mission overlays. aged in the computer subsystem box. The tional issues, leveraging commercial com- box has a single external connector for ponents, packaged and configured in a cus- Major subsystems and components power, USB, and IEEE 1394 FireWire con- tom fashion to meet battlefield conditions The Land Warrior system is characterized nections. The flash drive stores Land War- and requirements. by multiple integrated subsystems to achieve rior application software, National Imagery To address the battery problems, the IPT a more effective infantry unit. The computer and Mapping Agency (NIMA)-approved decided to use smart batteries that included subsystem (a Pentium) runs Windows and map products, field manuals, and system an SMBus 1.1 (System Management Bus, is the core of the Land Warrior computer information. Version 1.1, a commercial standard) read- subsystem (see Figure 5). It weighs 1.79 The helmet subsystem (see Figure 6) con- out of the battery’s charge status and other sists of the HMD, hearing devices, and data. However, SMBus had to be converted microphone devices. The HMD is an 800- to USB, so a smart-battery adapter was Figure 6. Helmet display. ×-600-pixel full-color display using an developed. In addition, the IPT used novel organic light-emitting diode display viewed power management techniques to extend through a high-efficiency plastic prism battery life. Now, when the soldier flips up encased in a protective housing. It allows his or her HMD, a switch turns off all video the soldier to interface with all Land War- components and places the computer in rior functions. During tactical movement standby mode (voice communications and and contact, the soldier will primarily use other functions still operate in this mode). it to view his or her location, other friendly The integrated handheld display and key- locations, and his or her direction of travel board let platoon leaders view maps with a (heading) superimposed on the map. larger display (in addition to the HMD) and The soldier control and communications rapidly enter graphics and text for mission subsystem (see Figure 7) is the system’s pri- planning. However, manufacturers are still mary soldier input and interface device. The trying to develop a color SVGA display (800 soldier control unit lets soldiers manipulate × 600), six to nine inches diagonal, which system configurations and generate and is the ideal size from a human-factors and send tactical messages. The communica- form-factor standpoint. This display would tions subsystem transmits voice and data allow for easy map reading while still fitting so that soldiers can communicate in their in a soldier’s Battlefield-Dress-Uniform squad. A mesh concept that forwards pack- cargo pocket. Furthermore, display manu- ets to soldiers in multiple hops enhances the facturers are working to make a touch- system’s range, and the Army will issue an screen display that is visible in all lighting AN/PRC-148 multiband inter/intra team conditions and meets all other environ- radio to squad leaders (and above) for mental requirements, such as a wide range longer-range communication and interop- of operating temperatures. erability with higher-echelon radios.

38 PERVASIVEcomputing http://computer.org/pervasive Army milestones Industry and research milestones DOS 3.0 released 1985 386 Microprocessor introduced

Soldier's Computer idea conceived

Soldier's Computer design team formed Reflection Technologies develops head-worn display (Private Eye) 486 Microprocessor introduced 1990 First Soldier's Computer prototype demonstrated (Oct. 1990) VuMan wearable demonstrated by CMU SIPE version—Soldier's Computer (Fall 1992) Windows 3.1 released Pentium introduced DARPA Smart Modules program begun Land Warrior requirements formulated Pentium Pro 1995 Land Warrior development begins

1st International Symposium on Wearable Computers

Pentium III Windows 2000 and Pentium 4 released Land Warrior, V 0.6 tested 2000 Land Warrior, V 1.0 tested

Figure 8. A timeline of Army wearable computer systems versus industry and academic developments.

The weapons subsystem (WSS) has a of other Land Warrior systems, which are fight, to a ubiquitous system that embeds mounted Daylight Video Sight and TWS automatically broadcast periodically. those products into an all-for-one system for sighting. Depending on the duty posi- The full Land Warrior system includes that a soldier wears to fight. The Objective tion, the soldier can mount currently not only the electronics but also all the other Force Warrior system focuses on electron- issued aiming lights, an infrared pointer, items that constitute the soldier’s combat ics embedded in an integrated combat uni- or a multifunctional laser. The laser com- load, including clothing, armor, weapons, form, and researchers at Carnegie Mellon bines multiple functions of currently and ammunition. Many integrated elements University and Georgia Techare exploring fielded systems into one device and inte- comprise the Land Warrior fighting system, similar concepts.4,5 grates a laser range finder and digital com- not just a computer—though it is a key com- In addition, the Army continues to inves- pass. A peg grip on the weapon’s stock has ponent. For more information, see https:// tigate advances in wearable computing buttons that let soldiers make calls, tran- www.pmsoldiersystems.Army.mil/public/ devices and the use of handheld devices to sition between sighting systems, capture default.asp. augment or replace wearable systems in cer- images, and locate targets without remov- tain situations. Under various research ini- ing a hand from the weapon. The WSS tiatives within CECOM’s Command and routes all target information and image Control Directorate, the Army is exploring capture to the computer, which automat- advances in computer hardware and soft- ically determines target location and fills ware applications that can run on small message fields as applicable. and Warrior continues to evolve portable-computing platforms. Also, Land The navigation system integrates a GPS from a system built around the Warrior, Version 1 (now called Land War- receiver with an antenna on the left shoul- soldier’s equipment, to a system rior IC) has demonstrated an early version der, a magnetic compass heading sensor, integrated with the soldier’s of speech recognition, one of its objective and a dead reckoning module, which equipment,L toward a system built within requirements. There are also two small busi- extrapolates the last known position should the soldier’s equipment (see Figure 8). It ness innovation research (SBIR) contracts the GPS fail or receive insufficient signal. It will progress iteratively from an all-in-one for “heads-up situation awareness for the also graphically overlays the soldier’s posi- wearable system that replaces portable dismounted warrior.” These contracts tion on a digital map, along with positions C4ISR products and enables soldiers to address the need to superimpose situa-

OCTOBER–DECEMBER 2002 PERVASIVEcomputing 39 WEARABLE COMPUTING

tional information on the HMD to help the AUTHORS avoid fratricide. Suomela and Lehikoinen presented similar concepts for augmented Matthew J. Zieniewicz is a senior electronics engineer for the Research, Development, reality at ISWC 2000.6 Government and Engineering Center, US Army CECOM. He formed and led the initial Soldier Com- puter team for the Soldier Integrated Protective Ensemble and was the initial computer- research engineers are examining low- radio-GPS technical lead for the Land Warrior system. He is exploring mobile database power computing devices, tablet PCs, and issues, Web technologies, and handheld applications for the dismounted warfighter handheld computing devices. As handheld and advanced system architecture issues for the Objective Force Warrior. He received a BSEE and an MSEE from Fairleigh Dickinson University, and he did post-graduate work devices become more powerful, the need at Princeton University. He is a licensed professional engineer in the state of New Jersey. for a wearable computer for certain appli- He is a member of Eta Kappa Nu, the national electrical engineering honor society, and of the ACM, IEEE, cations diminishes. Along these lines, there and Internet Society. Contact him at [email protected]. is also an SBIR solicitation (request for business contract proposals) calling for a Douglas C. Johnson is a system engineer on the Land Warrior program at Fort Mon- location-aware handheld computing device mouth for PM Soldier Systems, specializing in computers, personal area networks, and interoperability with other Army Battle Command Systems. He received his BS with integrated long-range (greater than in electrical engineering from Rensselaer Polytechnic Institute, N.Y., a masters in 500 km) communications. However, wear- electrical engineering from NYU, and a masters in computer science from FDU. He is able computers will always have a place on a member of the IEEE and Army Aviation Association of America. Contact him at [email protected]. dismounted soldiers, who need both of their hands free to perform missions while the computer augments their capabilities. In the applications area for mobile mil- Douglas C. Wong is team leader for CECOM RDEC’s Command and Control Direc- itary computing platforms, four technolo- torate. He manages the electronics technologies, products and systems integral to gies show particular promise: the Land Warrior System. He also leads a team of technical subject matter experts, and technical, readiness, software, and acquisition managers, who come from the depth and breadth of CECOM organizations. He has a BS in aerospace engineering • Intelligent agents on wireless wearable from Polytechnic Institute of New York, an MBA in strategic management of tech- computers communicating with remote nology from Monmouth University, N.J., and he is a graduate of the Army Manage- ment Staff College, Fort Belvoir, Va. He is a member of the Army Acquisition Corps. servers Contact him at [email protected]. • Java-based collaboration tools with whiteboarded military maps to plan and John D. Flatt is a senior electronics engineer for the US Army’s Program Manager, IE. He worked on the rehearse missions initial Soldier’s Computer effort under SIPE and has over eight years experience with mobile computing. • Speech recognition in the battlefield’s He received a BSEE and MSEE from Fairleigh Dickinson University. Contact him at [email protected]. high-noise and high-stress environments • Mobile wireless database retrieval and synchronization with handheld devices

Java and the Jini architecture show promise REFERENCES Electronic Textiles Modeling and for many portable networked applications Optimization,” Proc. 39th Design that support the network-centric battlefield. 1. M. Nugent, “The Soldier Integrated Automation Conf. 2002 (DAC 2002), The Army is exploring Bluetooth and other Protective Ensemble,” Army Research, AMC Press, New York, 2002, pp. 175–180. communications technologies to reduce Development, & Acquisition Bulletin, Sept./Oct. 1990, pp. 1–5. cabling issues associated with wearable 5. S. Park, K. Mackenzie, and S. Jayaraman, computing devices. 2. S. Mann, “An Historical Account of the “The Wearable Motherboard: A Framework for Personalized Mobile All these technologies will play a key role ‘WearComp’ and ‘WearCam’ Inventions Developed for Applications in ‘Personal Information Processing,” Proc. 39th in the Army’s vision of the future, as its Imaging,’” IEEE Proc. 1st Int’l Symp. Conf. Design Automation (DAC 2002), Objective Force Warrior system emerges Wearable Computers, IEEE Press, ACM Press, New York, 2002, pp. 170–174. over the next 10 years. Soon, the Army will Piscataway, N.J., 1997, pp. 66–73. have soldiers with integrated battlefield 3. A. Samilagic et al., “Very Rapid 6. R. Suomela and J. Lehikoinen, “Context systems consisting of the Land Warrior sys- Prototyping of Wearable Computers: A Compass,” Proc. 4th Int’l Symp. Wearable Computers, IEEE CS Press, Los tem. For now, the granddaddy of it all, the Case Study of Custom versus Off-the- Shelf Design Methodologies,” Proc. 34th Alamitos, Calif., 2000, pp. 147–155. backpack-sized Soldier’s Computer, will Ann. Design Automation Conf. (DAC soon become part of the Smithsonian’s per- 2002), ACM Press, New York, 1997, pp. manent collection, so you’ll be able to see 315–320. For more information on this or any other comput- it on a future visit to the Washington D.C. 4. D. Marculesa, R. Marculescu, and P. ing topic, please visit our Digital Library at http:// area. Khosla, “Challenges and Opportunities in computer.org/publications/dlib.

40 PERVASIVEcomputing http://computer.org/pervasive