Inertial MEMS Systems and Applications

This paper provides a survey of some of the available MEMS Inertial Measurement Units (IMUs) and some of the applications that MEMS are fulfilling. Inertial MEMS have been under development for over 20 years and numerous suppliers world-wide now exist. The performance of these IMUs spans the range from consumer/automotive grade (low performance) to tactical grade (~1 deg/hour, ~1 milli g). Tactical-grade IMUs were primarily developed for defense applications to reduce reliance on GPS. The application of automotive quality IMUs to three gun-launched munition tests is described. This is followed by a description of a path that led to a tactical grade IMU and subsequently to gun-hard common guidance IMUs. Other nongun-hard tactical grade IMUs are mentioned. A brief discussion on packaging improvements that permitted size reduction is presented. Some of the current MEMS IMUs at various levels of performance are described and typical defense and commercial applications are described. Finally, a brief insight into ongoing performance improvement initiatives and what the future may bring are presented. 1.0 INTRODUCTION MEMS inertial sensors and systems have become indispensable to the future of navigation [Ref. 1]. Also, navigation itself has become much broader than just providing a solution to the question ‘Where am I?’ It has moved into new areas such as games, geolocation, virtual reality, timing, tracking; all of which have been enabled by MEMS technology. Their small size, low power and weight, and low cost have led to increased use, new applications, increased mobility, increased integration (hence better performance) and extended operation. They are also very rugged and can survive tens of thousands of g’s. It been estimated that 752 million units of MEMS accelerometers and gyros were produced worldwide in 2008 with the market dominated by automotive and consumer electronics applications [Refs 2 – 4]. Reference 2 also predicts that the MEMS inertial market will be $3billion by 2013 of which $500million will be for defense and aerospace. Numerous consumer/commercial grade inertial MEMS systems have been available for several years. These have been integrated with augmentation sensors, such as magnetometers and velocity meters, and GPS aiding to provide an adequate navigation solution. In fact the near universal availability of GPS and other satellite navigation systems means that navigation-on-demand is now expected. The major problem is that GPS is not always available or cannot be acquired quickly enough (for short term applications of a few seconds) so that reliance on the inertial system is necessary. Nor can GPS provide the same flight control/autopilot capability that the inertial system can. This resulted in the development of higher performance MEMS inertial sensors and systems. Several manufacturers now provide all-MEMS inertial measurement units (IMUs) with tacticalgrade quality performance (approximately 1 deg/h for the gyro and 1 milli g for the accelerometer), which has enabled MEMS IMUs to be used in numerous military applications such as missiles, munitions, soldier Inertial MEMS Systems and Applications 3 2 RTO-EN-SET-116(2011) navigation, etc. The performance limitation to tactical grade is due to the gyro not the accelerometer. Current MEMS gyro technology is based on sensing the Coriolis induced force on a vibrating structure, which limits the gyro to behaving as a rate gyro [Ref. 5]. Reference 5 provides a broad description of inertial navigation sensor technology status and developments, including initiatives to develop alternate gyro technologies. This paper describes some of the applications of inertial MEMS systems. It begins in Section 2with the evolution of MEMS inertial systems for gun-launched applications. It continues with the development of tactical grade IMUs in Section 3. Section 4 provides a brief description of packaging for size reduction. Current MEMS IMUs and their applications are discussed in Section 5. Section 6 outlines what the future may hold. 2.0 EARLY GUN HARD MEMS IMU APPLICATIONS Current 5-inch gun-launched munitions are limited in range to about 12 nmi and as a consequence of dispersion effects (unsteady winds, propelling charge performance, aiming errors, etc) are not precision weapons and thus are not adequate for Naval Surface Fire Support. Rocket motors can be used for extra boost post gun-launch to increase the projectile range, however without an on-board guidance and control system, the uncertainty of the projectile impact point increases dramatically. The evolution to tactical grade performance is a direct result of advances made in key technology areas originally driven by gun-launched projectile requirements [Refs. 6 – 12]. These applications have a unique combination of requirements including, performance over temperature, high-g launch survivability, fast initialization and startup, small size, and relatively low overall system cost. Table 1 shows typical environments for tank, artillery, missile, and mortar munitions. Table 1: Munition Environments. Ref 12 Brown et al, "Strap-Down Micromechanical (MEMS) Sensors for High-G Munition Applications", IEEE Transactions on Magnetics, Jan 2001 The first MEMS gyroscope was developed by Draper Laboratory in the 1980s. At the beginning of the 1990s the possibility of MEMS IMUs with reasonable performance, when aided by GPS, had become a reality. The capability of MEMS technology in gun-hard applications was demonstrated through a series of technology demonstrations: the Extended Range Guided Munition (ERGM) Demonstration in a munition with rocket motors used for boost post gun-launch; the Competent Munitions Advanced Technology Demonstration Inertial MEMS Systems and Applications RTO-EN-SET-116(2011) 3 3 (CMATD) in a 5-inch artillery shell, and the Low Cost Guidance Electronics Unit (LCGEU) in an extended range munition. ERGM Demonstration and CMATD involved guided projectile tests of MEMS INS/GPS systems utilizing the best MEMS sensors available at Draper Laboratory at that time. LCGEU utilized COTS MEMS in a modular INS/GPS designed to be robust to GPS jamming [Ref. 13]. Another program with the goal of reaching tactical grade IMU performance was the DARPA MEMS IMU (MIMU) program in the early 2000s. The culmination of much of this technology evolution was incorporated into the Common Guidance IMU (CGIMU) program, which had the goal of incorporating tactical grade MEMS across multiple tactical weapon platforms. MMIMU and CGIMU are discussed in Section 3. Figure 1 provides a top-level description of the technology roadmap for the ERGM, CMATD, MMIMU, LCGEU, and CGIMU systems. Figure 1: System Technology Roadmap. 2.1 ERGM Demonstration INS/GPS In March 1995, the Naval Surface Fire Support branch initiated a proof-of-concept demonstration for the Extended-Range Guided Munition (ERGM). The ERGM Demonstration program was the first successful demonstration of a gun-launched MEMS-based INS/GPS system. The system consisted of a 126 in (2065 cc) avionics package containing a 6-degree-of-freedom MEMS inertial system, a Rockwell-Collins C/A-to-P(Y) code reacquisition GPS receiver with L1 tracking only, a TMS320C30 flight processor and power conversion and regulation electronics, which were sectionally mounted into a Deadeye Projectile. The relatively generous volume of 126 in3 (2065 cc) allowed conservative, rugged packaging technologies to be used to meet the required survivability goals. Five PWBs were hard-mounted to rigid aluminum frames C/A to P(Y) Code GPS Tightly Coupled Hybrid Technology ERGM Demo INS/GPS Vol = 126 in (2065 cc) Six 1-axis MEMS sensors 500/hr, 20 mg IMU 24 Watts 6,500-g gun launch 1995-1997 Miniature P Code Engine Direct P(Y) Reacquisition Tightly Coupled MCM /ASIC Technology CMATD INS/GPS Vol = 13 in (215 cc) Six 1-axis inertial modules 50/hr, 1 mg IMU 10 Watts 12,500-g gun launch 1996-2000 Deep Integration GPS SAASM GPS Direct P(Y) Reacquisition ASIC Chip Set Common Guidance IMU Vol = 2 in (33 cc) (3 in w/GPS) Single 6-axis inertial module 0.3/hr, 100 μg IMU < 5 Watts 20,000-g gun launch 2002-2009 GPS not included P(Y) Code GPS Add-on Deep Integration Compatible BGA/ASIC/MCM Technology MMIMU Vol = 8 in (131 cc) Two 3-axis inertial modules 1/hr, 100 μg IMU < 3 Watts 2000-2002 Deep Integration GPS Direct P(Y) Reacquisition BGA/ASIC/MCM Technology LCGEU INS/GPS Vol = 20 in (328 cc) Two 3-axis inertial modules 300-500/hr, 10 mg IMU < 12 Watts > 10,000-g gun launch 2002-2004 Inertial MEMS Systems and Applications 3 4 RTO-EN-SET-116(2011) and bolted into a cylindrical housing with end plates that served as the primary load bearing structure. The sensors and their discrete electronics were packaged using thin-film hybrid MCM-C technology and mounted in bulky hermetic metal housings structurally bonded to standard PCBs. The ERGM Demonstration sensor electronics packaging appears in Figure 2. Figure 2: ERGM Demonstration Sensor Electronics Packaging. A photo of the first of three (November 1996 and two in April 1997) successful ERGM Demonstration test flights is shown in Figure 3. This effort first demonstrated successful reacquisition of GPS after gun launch and proved the survivability and ability of MEMS inertial components to operate after launch and accurately measure body rates and accelerations. The MEMS-based system was composed of repackaged automotivegrade inertial components (Draper Laboratory TFG gyros and pendulous accelerometers with uncompensated performance of 1,000 deg/hr and 50 mg) and was able to perform down-determination successfully and provide inputs for a full navigation solution. Inertial MEMS Systems and Applications RTO-EN-SET-116(2011) 3 5 Figure 3: ERGM Demonstration Flight Test. 2.2 CMATD INS/GPS From March 1996 through February 2000, under funding from the Office of Naval Research, a series of three flight tests for the Competent Munitions Advanced Technology Demonstration (CMATD) Program was completed. The objective was to demonstrate MEMS-based guidance, navigation, and control (GN&C) within the fuze section of unguided artillery rounds. Figure 4 is a photo of the CMATD 5-inch (127 mm) projectile in-flight, 20 ms after the 6,500-g gun launch. The system GN&C is 13 in (215 cc) total, with 8 in (131 cc) for the G&N electronics. Although the MEMS gyros and accelerometers were similar to ERGM Demonstration, the CMATD sensor electronics were the first to be ASIC-based. This contributed to an order of magnitude improvement over ERGM Demonstration performance. Figure 4: CMATD Projectile in Flight. The volume constraint of 8 in3 (131 cc) for the electronics assembly was very aggressive, and is shown in Figure 5. This system consists of a flight computer module, three orthogonal accelerometer modules, three orthogonal gyroscope modules, a two-card GPS receiver, a TCXO clock board, and a voltage regulator card. A backplane and flex cables provide electrical interconnection between modules and external interfaces for Inertial MEMS Systems and Applications 3 6 RTO-EN-SET-116(2011) system initialization. The modules are over-molded with epoxy to provide mechanical and environmental protection and assist with thermal management. Modules fabricated in this manner have survived centrifuge tests in excess of 30,000 g and more than 400 thermal shocks from -55°C to +125°C. (a) CMATD Electronics Assembly (b) Integrated into Mk64 Guidance Figure 5: CMATD Electronics Assembly. The CMATD Program culminated in flight tests of three projectiles at Yuma Proving Ground on August 3, 1999, August 5, 1999, and February 2, 2000. All projectiles were launched at a setback acceleration of approximately 6500 g with initial velocities of approximately 2200 ft/s (670 m/s). The initial flight tests demonstrated survivability of the overall projectile design and subsystem functionality and performance. From the telemetry data acquired from the first two flights, all six MEMS instruments survived the gun launch and operated as expected and provided accurate in-flight measurements. The lessons learned in these test flights regarding the control actuation system (CAS), roll control software, and launch signal subsystems were modified for the third projectile. Test Flight 3 was conducted February 2, 2000, and the GN&C system survived gun launch and all systems operated. GPS was reacquired successfully at 31 seconds after launch and the closed-loop guidance, navigation, and control executed as designed. A modified CMATD IMU was evaluated in a Precision Guided Mortar Munition (PGMM) application. The IMU was used for flight control during and after launch without GPS aiding. The PGMM was successfully tested in 2001 through a 7,900 g launch environment. 2.4 LCGEU INS/GPS The following information is taken from Reference 13. The US Navy initiated an investigation into providing alternate guidance electronic unit technology to address survivability across gun-shocks and enable a significantly lower unit production cost. The LCGEU program’s goal was to develop a gun-hard, low cost INS/GPS system using COTS inertial sensors and other COTS components. The LCGEU contains Analog Devices gyros and accelerometers and is designed to be modular so as to fit into various airframes. The IMU is a 20 cu in (328 cc) system, hardened to 18,000g, and has deep integration software for enhanced GPS antijam capability. It was the first ‘common’ guidance unit for US Navy projectiles and was successfully tested in long range guided flights of two Ballistic Trajectory Extended Range Munition (BTERM) in September 2003 Inertial MEMS Systems and Applications RTO-EN-SET-116(2011) 3 7 (range 53.6 nmi) and in one EX-171 Extended Range Guided Munition (ERGM) in October 2003 (range 44.8 nmi). The two BTERM projectiles impcted 20.5 and 21 meters from the target; the Ex-171 ERGM impacted 10 meters from the target. The BTERM projectile is a rolling airframe (15 – 25 Hz) employing a single actuator driving a pair of canards to compensate for trajectory dispersions along a nominal ballistic trajectory to the target. The Ex-171 ERGM is a non-rolling airframe utilizing four independently articulated canards to stabilize the projectiles orientation and steer it to the target along a boost-glide trajectory. Figure 6 shows an exploded view of the LCGEU. Figure 6: Low Cost Guidance Electronics Unit. Because the LCGEU uses commercial grade sensors, the navigation error increases dramatically with loss of GPS. To compensate for this additional anti-jam GPS antenna electronics were included and the navigation filter-formulated data used Draper’s Deep Integration algorithms. Previously, in December 2001, BAE Systems also participated in successful ERGM Control Test Vehicle flight tests in which their SiIMU was used [Ref. 14]. 3.0 PATH TO TACTICAL GRADE PERFORMANCE As long as GPS is available then the use of low performing inertial sensors is acceptable. However, to reduce reliance on GPS, the need for MEMS to reach tactical grade or better is essential. The size, weight, power and cost advantages that MEMS offers would also make MEMS IMUs very competitive with RLG and FOG systems. Therefore significant effort was (and still is) put into improving the performance of MEMS IMUs. 3.1 MMIMU The Special Projects Office of the Defense Advanced Research Projects Agency (DARPA) and the Air Force Research Laboratory (AFRL) sponsored Draper Laboratory from 2000 to 2002 to develop and demonstrate the world’s highest performance MEMS IMU, the DARPA MMIMU [Ref. 15]. The performance requirement was a 10 deg/h, 500 μg IMU with a performance goal of 1 deg/h, 100-μg, and with a unit production cost goal Inertial MEMS Systems and Applications 3 8 RTO-EN-SET-116(2011) of $1200. Designed as a low-cost, smaller, low power alternative to Honeywell’s ring laser gyro-based tactical-grade HG1700 IMU, the MMIMU is 2.7 inches (68.5 mm) in diameter and 1.4 inches (35.5 mm) high (8 in) with a weight of 260 grams. The IMU was designed to perform over a temperature range of -40 to +85°C and consume less than 3 W. Development experience and the need to design for ease of manufacture and low cost drove the MMIMU designs to much simpler implementations. The MMIMU assembly shown in Figure 7 features a set of four plug-in modules with screw attachments. From top to bottom are the IMU processor, power conversion electronics (PCE), accelerometer, and gyro modules. Each module consists of a PCB mounted to a Kovar frame, and alignment pins are used to guide module-to-module assembly and connector mating. Top and bottom seam-welded covers provide a hermetically-sealed system. Figure 7: MMIMU Assembly. To meet the volume constraints, a novel planar stacked disk approach was developed for system packaging. Inertial sensors are hermetically sealed in an leadless ceramic chip carrier (LCCC). A separate 3-axis accelerometer and 3-axis gyroscope instrument module are configured with ASIC electronics, further improved to reduce the number of off-chip components. The two 3-axis inertial instrument modules contained the latest MEMS TFG and pendulous accelerometer sensor designs. Ceramic mounting blocks are employed for the orthogonal set, and a separate mixed-signal CMOS ASIC in a chip-scale package operates each sensor. ASIC electronics implemented in CMOS processes provide excellent performance at low power and cost, while enabling the small size objectives to be met. Much of the gains in performance experienced in TFG 3-Axis Accelerometer Module • X01 • RM -1 • Ceramic circuit 3-Axis Gyro Module • TFG USP • HP -1 ASI • Ceramic circuit PC • 5-Vd input • Conversion and regulation • Laminate circuit Processor • TMS320VC33 • Laminate circuit TXC Inertial MEMS Systems and Applications RTO-EN-SET-116(2011) 3 9 development are directly attributable to the ASIC electronics. The MMIMU gyro ASIC, the HighPerformance Gyro digital ASIC (HPG-1), was developed under the DARPA MMIMU Program. This thirdgeneration ASIC uses a high-speed Σ–∆ converter to acquire the gyro rate information directly from the preamplifier. All sense axis processing is performed digitally, eliminating errors associated with drift in the baseband electronics while achieving dynamic ranges in excess of 140 dB. The accelerometer electronics form a differential capacitive measurement system consisting of a carrier signal generator and demodulator circuitry providing a DC output proportional to acceleration. During closed-loop operation, control compensation and rebalance drive circuitry are added. An existing second-generation analog ASIC (RMA-1) was used for the accelerometer. The ASICs were packaged in custom ball grid array packages. Build of two MMIMUs was started, with one completed and tested by August 2002. Table 2 presents a summary of the MMIMU goals and the actual IMU test data. As of today (2010) these results still represent the highest published performance data attained to date on an all-Silicon MEMS IMU. The DARPA MMIMU Program was the culmination of three years of focused development on advancing the performance of MEMS gyroscopes. This effort leveraged 10 years of development and over $100M of Draper IR&D and government investments in the development of MEMS inertial technology. Table 2: MMIMU Sensor Performance. Parameter Gyros Accelerometers Goal Actual Goal Actual Bias Turn-on Repeatability deg/h or mg 1 (1σ) 3 (1σ) (1) 0.5 2 (1σ) (1) Bias In-run Stability deg/h or mg 1 (1σ) 5 (1σ) (2) 0.1 (1σ) 1 (1σ) (2) Bias Drift deg/h or mg 5 (1σ) (3) 2 (1σ) (3) Scale Factor Turn-on Repeatability ppm 140 (1σ) 70 (1σ) (1) 210 125 (1σ) (1) Scale Factor In-run Stability ppm 140 (1σ) 100 (1σ) (2) 210 600 (1σ) (2) Axis Misalignment mrad 0.5 1 0.5 1 Input Axis Repeatability mrad 0.1 0.2 0.1 0.2 Maximum Input deg/s or g 1,000 1,000 (4) 50 45 (4) Bandwidth Hz 150 150 (4) 100 100 (4) Angle Random Walk deg/√h or m/s/√h 0.030 0.050 0.035 0.02 Notes: (1) Turn-on to turn-on over 0°C to +70°C, power-down, and mount/dismount, 14 days (2) Post-compensation over 0°C to +70°C (3) RSS of turn-on and in-run performance (4) Typical; other ranges available 3.2 CGIMU The goal of the US Army’s Common Guidance IMU program was to develop a new system that extends the capabilities of the previous systems further yet in terms of highest performance with high launch loads, antiInertial MEMS Systems and Applications 3 10 RTO-EN-SET-116(2011) jam GPS, decreased volume, and low production cost ($1200 per IMU and $1500 per INS/GPS in high volume) [Ref. 16]. CGIMU was a multi-year program in three phases as shown in Table 3 [Ref. 16]. Significant development effort was planned within the scope of this program to shrink packaging volume, power and overall system cost while improving sensor performance. The ultimate goal was to develop a common system for use in the majority of the Department of Defense’s tactical applications. Table 3: CGIMU [Ref 16]. Parameter IMU Volume Gyro Performance Accelerometer Performance Expected Completion Phase 1 8 cu in 200 °/h (3σ) 9 mg (3σ) April 2003 Phase 2 4 cu in 20 °/h (3σ) 4 mg (3σ) October 2004 Phase 3 2 cu in 1 °/h (3σ) 0.3 mg (3σ) October 2006 Initially several organizations participated in the CGIMU program. Honeywell, in partnership with Draper Laboratory, was funded in 2001 with the MMIMU technology as the Phase 1 baseline. During this development period effort was also expended on improving anti-jam capability through improved algorithms. Figure 8 (courtesy of Honeywell) outlines Honeywell’s Common Guidance technology progression through the subsequent phases; the HG1930 being 4 cu in and the HG1940 2 cu in. Figure 8: Honeywell’s Common Guidance Technology Progression (Published with permission of Honeywell). 3.3 Tactical Grade MEMS IMUs There now exist several suppliers of tactical grade IMUs and INS/GPS systems. Honeywell has the HG1930 and HG1940 IMUs. In 2006 Integrated Guidance Systems (IGS), a Honeywell and Rockwell Collins joint venture, introduced family of Deeply Integrated Guidance and Navigation Units (DIGNUs) [Ref. 17]. IGS’s first product was the IGS-200, a 13.7 cu in (225 cc) gun-hardened system. Subsequent products are the IGS202 (16.5 cu in) and IGS-250 (7.8 cu in) INS/GPS systems [Ref 17]. BAE Systems (now Atlantic Inertial Systems (AIS) a B. F. Goodrich Company)), Northrop Grumman LITEF, and Systron Donner Inertial (SDI) Inertial MEMS Systems and Applications RTO-EN-SET-116(2011) 3 11 have all developed tactical grade MEMS IMUs. AIS have the SiIMU02 (~5 cu in) and SiIMU04 (~8 cu in) IMUs as well as the SiNAV (~20 cu in) Inertial Navigation System [Refs. 18 20]. LITEF have developed a 20 cu in IMU [Ref. 21] for AHRS applications. SDI have developed the SDI500 IMU (~19 cu in) shown in Figure 9 (courtesy of Systron Donner Inertial). The SDI500 is based on quartz technology. The high signal to noise ratio of the piezoelectric readout from quartz gives the SDI500 very low Angle Random Walk. Systron Donner also has the SDN500 (~25 cu in) miniature integrated GPS/INS system [Ref. 22]. Figure 9: SDI500 with Schematic of Internal Mounting of the Sensors (Published with permission of SDI). It should be noted that, while the existence of tactical grade IMUs has reduced reliance on GPS, they have not eliminated the need for GPS (whenever available) or other augmentation sensors to ensure mission success. For example, in purely inertial mode, a tactical grade IMU’s position error would be around 150 meters in 3 minutes. 4.0 PACKAGING Packaging considerations are very important for performance as well as survivability and significant packaging challenges arise when trying to minimize the size and cost of a product expected to last 20 years in an uncontrolled environment. In particular, for very high g applications, typical environments may include a 20,000 g setback acceleration, angular rates up to 250 revolutions per second, and a 5000 g set-forward acceleration upon exit from the gun barrel. Also, MEMS sensors tend to have a fairly high sensitivity to external vibration. Therefore, the inertial sensors and electronic assemblies must be properly mounted, especially if they are to survive a gun launch environment. Shock mounts are typically employed for inertial components to attenuate and dampen the high-g launch loads and in-flight vibration inputs. The size of the inertial sensor modules originally dominated the size of MEMS IMUs. This was not due to the size of the sensing element itself, but from the electronics. Initially, commercial field programmable gate arrays (FPGAs) were often used for many digital functions. Subsequently the electronics were packaged in Application Specific Integrated Circuit (ASICs) with configurable gate arrays (CGAs) to replace FPGAs, and passive conditioning electronics into a custom ball grid array (BGA) package. As ASIC development became more advanced it virtually eliminated all off-ASIC components on the board. This resulted in a ~50% volume Inertial MEMS Systems and Applications 3 12 RTO-EN-SET-116(2011) reduction in board size. Current packaging is now able to stack (bump bond) 7 or 8 chips in one BGA about 2 mm high. Other advanced packaging techniques use Flat No Leads packages, using landing pads not BGAs. Another major advance in packaging size reduction resulted from the design and fabrication of planar sensors that could measure in-plane (x-axis and y-axis) or out-of-plane (z-axis) rate or acceleration. Originally, threeaxis measurements were achieved by rotating sensor packages on ceramic mounting blocks as shown in Figure 10(a) to get measurements along x, y, and z axes. The ability to use planar mounting of all the sensors further minimizes overall system packaging volume since the ceramic mounting blocks are eliminated. Utilizing a combination of these new sensor technologies and the aforementioned ASIC electronic advances with CGA devices custom packaged in a BGA format enables the realization of a complete 6-axis inertial instrument module as shown in Figure 10(b). This yields a total volume reduction of >75 percent over that in Figure 10(a). (a) Two 3-Axis Sensor Modules (MMIMU) vs. (b) One 6-Axis Sensor Module Figure 10: Improved Packaging. 5.0 CURRENT MEMS IMU APPLICATIONS As of today, MEMS IMUs have been successfully demonstrated in numerous applications, including the high g applications (e.g., artillery shells, extended range (rocket assisted) munitions, mortar shells, anti-tank weapons) mentioned previously. MEMS IMUs can now be purchased for commercial and military uses from numerous companies. In fact, MEMS is poised to become the worldwide industry standard for the design of nearly all modern smart weapons. Table 4 [Ref. 23] shows typical generic inertial bias from 1991 for missile guidance applications (assuming no GPS, but with other initial or terminal aiding). As shown in the table, MEMS is now able to meet most of these bias requirements. The other major requirement is low noise and achieving this is still an issue for MEMS IMUs. Table 4: Generic Bias for Missile (no GPS) [Ref. 23]. Mission Gyro (°/hr) Accelerometer (μg) Air-to-Air Surface-to-Air Air-to-Surface Mid-Course Guidance Surface-to-Surface Long-Range 10 100 10 – 100 0.1 – 10 0.01 – 1.0 0.001 – 0.1 0.001 – 0.01 500 1000 500 – 1000 50 – 500 20 – 200 5 – 100 5 – 50 Blue – MEMS can meet. Red – MEMS not yet demonstrated Inertial MEMS Systems and Applications RTO-EN-SET-116(2011) 3 13 Tactical grade MEMS IMUs have been fielded in a number of existing and developing weapon systems, such as Excalibur, Small Diameter Bomb, E-Paveway, Hellfire, Joint Common Missile, ERGM, BTERM, etc. For many applications tactical grade performance, while desirable, is not essential and commercial grade MEMS, with appropriate GPS aiding and other augmentation sensors, will suffice. Table 5 [Ref. 24] shows typical generic bias, scale factor, and random walk for a GPS/inertial guided artillery shell and guided bomb. It can be seen that, even with a certain amount of jamming, inertial MEMS with commercial grade performance is sufficient. However, for the majority of the military applications a tactical grade IMU (or better) is required. Table 5: Generic Bias, SF and RW for GPS/Inertial Guided Munitions [Ref. 24]. Gyroscope Accelerometer Range (km) GPS Off (km) Bias °/h SF ppm ARW °/√h Bias mg SF ppm Guided Artillery Shell 12 5.6 10