HWIL: Technological Advances Add Reality to Missile Tests
Testing a rocket-propelled grenade or new shoulder-launched missile is relatively easy. First you design your system. Then you build it. Finally you make a set of prototypes, hook them up to monitoring systems, place a target on the firing range, and BAM! If the system worked, the target is most likely gone.
Testing multimillion-dollar guidance systems that are attached to incredibly complex ballistic missiles interceptors is not as straight forward. The expense and complexity of testing the systems limits the number of real flight tests to just a few. Conversely, soldiers have to count on the performance of these ballistic interceptors because the complexity of the systems directly reflects the importance of their task: protection from weapons of mass destruction. In 1986, it was with these factors in mind that the Ballistic Missile Defense Organization (BMDO) created the Kinetic Kill Vehicle Hardware-in-the-Loop Simulator (KHILS) at Eglin AFB (Valparaiso, FL).
Hardware-in-the-loop (HWIL) means essentially what it says: put the guidance, navigational, or other subsystem in an automated, computer-controlled, closed loop environment with sensing input data modeled on real-life engagements and see how it performs. Of course, the better the simulation, the higher the confidence that a system will perform as well in the air as it does on the ground. Until 1986, each of the U.S. armed forces had its own testing facilities. As these weapons of war became more complex, they required newer, more intricate testing mechanisms. In order to maximize its investment, the BMDO, then known as the Strategic Defense Initiative Organization, created a "center of excellence" for HWIL testing and HWIL test technology development.
Although the U.S. is no longer improving its nuclear stockpile or the rockets that deliver them, it has continued to develop smarter interceptors: air-to-air missiles and surface-to-air missiles that intercept incoming aircraft or other missiles. These systems have to operate in a variety of environments -- from lower atmosphere to the frigid wastes of outer space. Considering this, KHILS has had to develop a wide variety of simulators. Some of those systems include the adaptation of commercial graphics computers to perform real-time 3D rendering, steerable lasers, and large-scale 2D resistive-element arrays to simulate high-intensity targets, decoys, and debris. Even newer simulators will allow the missile and projector to actually mimic the ultracold backdrop of space while undergoing structural vibrations due to onboard chemical thrusters; while others will use a target simulator that combines radio frequency and IR into a common aperture.
Missiles have long depended on IR guidance systems to track the heat from a jet engine or the ferocious intensity of a rocket plume. Their guiding sensors require that they detect the object at a great distance and correct their trajectory to meet their objective in a very short period of time.
Over the years, the KHILS facility has developed large-format, resistive-element projection arrays to provide the IR input into missile guidance systems. Phenomenology and modeling codes are constantly upgraded to optimize speed and include new types of environmental conditions and targets. In order to adequately test the guidance system, these codes must be able to accurately mimic target size growth and spatial representation as it relates to the guiding sensor.
A relatively new development at the KHILS facility is dual-band IR projection technology. The system uses two Wideband IR Scene Projectors (WISPs), each with its own 512 x 512 resistive-element array. Each array emits broadband IR imagery that can be calibrated for banded emission in any desired waveband. According to KHILS Senior Engineer, Tony Thompson, newer missile systems use focal plane arrays to perceive radiation in multiple IR bands. Such focal-plane arrays can be tailored to be sensitive to both midwave IR (3 to 5 µm) and long wave (8 to 12 µm) bands, or subsets of either band. Resistor arrays have been designed to produce measurable radiation from very short wavelengths to beyond 20 µm. However, Thompson said that, as yet, KHILS has not been given a requirement to test a system at the very long wavelength end of this spectrum.
To test dual-band IR systems, radiation from two arrays is combined and sent through collimating optics so that the target appears to the missile system to be at infinity. According to Thompson, missile systems are designed to focus at infinity because, for the majority of engagements, the targets are very far away. Special alignment and calibration techniques are incorporated to ensure that the imagery from both bands is properly registered and scaled to simulate the correct inband energy. A similar system, called the Multi-Spectral Scene Projector (MSSP), is under development by several branches of the U.S. armed services and will combine IR images with radio frequency data to simulate the combined signatures for IR and radar sensing systems.
The resistive-array is an ongoing area of development at KHILS. In addition to expanding the size of the array to more than 1024 x 1024, researchers at KHILS and Honeywell are reconfiguring the array so that it updates all pixels simultaneously. "Ideally, you want to update [the array] all at once. A missile is approaching very fast so, if the [missile] sensor were to collect a frame halfway through [a projector] update when half the target is small and half of it has been updated to a larger size, it could create a problem. Always providing a consistent image would also simplify interfacing the projector with the sensor," Thompson explained. Also, the array currently updates at 120 fps. KHILS researchers would like to get that up to 200 fps. Other areas of improvement include the reduction or elimination of bad pixels, higher temperature emitters, improved emitter nonuniformity, and larger format devices. WISP arrays have shown better than 99 percent operation, but high yield arrays can experience groupings of dead pixels that are unacceptable, as they cannot be overcome with optical oversampling. Nonuniformity in the resistive array means that its effective dynamic range is limited by the maximum value of the lowest operational pixel. Existing WISP arrays have contrast ratios in excess of 1000:1. A multiservice development project is expected to begin this year for emitter array devices having 1024 x 1024 emitter elements.
Another system used in conjunction with WISP, the Steerable Laser Projector (SLP), adds the ability to simulate extremely hot objects such as flare countermeasures within the overall IR simulation. The SLP uses six independently steered midwave lead-salt laser diodes to project subpixel IR sources simulating debris or countermeasures. Once the target is resolved, the SLP can be used to illuminate portions of the extended body to simulate an ablating nose on a re-entry vehicle or a rocket nozzle on a boosting missile. The SLP can achieve a contrast ratio of 50000:1
Currently, tests are conducted with the missile resting either on an optical bench or in the center of a three-axis gimbal system, commonly referred to as a flight motion simulator (FMS). In the first scenario, the line-of-sight motion is digitally computed with the graphics computer; in the second, the missile is moved to simulate relative motion between the interceptor and the target. According to Thompson, a better scenario would have both the missile and projector capable of moving relative to each other.
To address this, KHILS has contracted with Carco Electronics (Menlo Park, CA), to produce a five-axis flight motion simulator where the missile is mounted in the center of a three-axis gimbal and the projector is free to move in 2D on the outer two axes. This new hydraulic-powered system is similar to the five-axis flight table installed at the Guided Weapons Evaluation Facility (GWEF) located at Eglin AFB, but will feature extreme gimbal travel to accommodate and accentuate BMDO scenarios.
"This allows you to test the Inertial Measurement Unit directly. It's not necessarily a new technology, but what is new is that people are starting to put arrays out on the target gimbals," Thompson said. "The resistor arrays have been the first projection technology that was small enough to do that unless you used a simple black-body cut-out." Also under development is a high frequency gimbal system that will have 1000 Hz dynamic response in order to simulate missile vibration associated with flight and its effect on guidance systems. Such a system can either stand alone or be mounted to an existing flight table.
To further improve the realism and flexibility of the simulation, KHILS has moved from 2D scene rendering to 3D rendering. A significant contributing factor behind this move is the advent of cheaper, graphics-intensive commercial computing power from Silicon Graphics, Inc. (Mountain View, CA). According to KHILS officials, early attempts to create 3D scene renderers resulted in expensive, one-of-a-kind systems that did not have a clear upgrade path. The advent of the SGI Onyx/Reality Engine, and now the Onyx-II/Infinite Reality suite, has added basic 3D rendering available at a more reasonable price.
Although leveraging on commercial equipment developed for the human eye has given KHILS a valuable tool, the sensing systems used in missile guidance are far more discriminating. Specifically, intensity differences and subpixel sampling rates are inadequate for missile guidance tests. Commercial SGI systems sample each pixel at four different locations to generate an intensity level; however, this can result in intensity variations several times greater than the actual intensity. To resolve this issue, KHILS scientists altered the SGI's programming to oversample 64 times within each line-of-sight pixel, reducing errors to less than one percent of the actual intensity.
For some tests, the background complexity and requisite number of polygons, or basic units in a 3D scene, also necessitated adding additional graphics processing units to the SGI system. According to Thompson, the additional computing power is more an issue when the SGI is used as a LADAR (laser radar) simulator, which produces more complex backgrounds for ground-based targets.
Although WISP provides a versatile building block for HWIL testing, some scenarios require additional environmental measures. The KHILS Vacuum Cold Chamber (KVACC) creates an atmosphere similar to that of space. Programs under development for Theater and Strategic Missile Defense such as THAAD and the National Missile Defense Program would intercept incoming weapons of mass destruction in outer space before they enter the atmosphere.
The 15 x 38-in. optical bench in the KVACC sensor chamber can be cooled to 75° K using a closed cycle helium refrigerator. All reflective optics provide for 95 percent transmission through a 30-cm aperture between the projector and sensor chambers. The WISP array has been operated as low as 145° K, which should satisfy most BMDO sensor test requirements. If colder projector temperatures are required, the WISP array can be replaced with special cryogenic arrays developed by the Defense Special Weapons Agency (DSWA; since changed to the DTRA -- Defense Threat Reduction Agency).
This year, the KVACC is expected to begin its first dual-band IR tests with the addition of a second WISP unit. These and the many other developments at Eglin AFB's KHILS facility will help weapons designers continue to improve their systems while controlling the costs associated with live-fire testing.
