The US Joint Direct Attack Munition (JDAM) family of inertial/GPS guided bombs became a household word with the extensive use of these weapons during the Enduring Freedom air campaign in Afghanistan. This was not the first use of the JDAM, delivered by the B-2A during the Allied Force campaign in 1999, the JDAM is credited with providing a critical all weather strike capability during periods of dense cloud cover, when the primary laser guided weapons used by the NATO force proved ineffective.

The JDAM has proven to be a highly effective weapon, offering new capabilities and very significant long term growth potential, but it is not without its critics. This two part feature will explore the current status of the JDAM and a number of related growth programs currently under way.

Inertial/GPS Guided Bombs

The origins of modern GPS guided bombs such as the JDAMs lie not in the domain of GPS satellite navigation, but in inertially guided bomb experiments performed during the 1980s.

Until that period, the dominant guided bomb technology was the laser guided weapon, first introduced like television guided weapons during the Vietnam war period. That conflict saw a long running and sustained war of attrition conducted by the US Air Force and US Navy against North Vietnam. While average loss rates of US aircraft to Russian supplied AAA and SAMs were fairly low, the cumulative effect over a decade long war was telling. This produced significant pressure for precision weapons, and the early GBU-2 laser guided bombs and GBU-8 HOBOS television guided bombs evolved primarily to reduce the number of aircraft exposed to defensive fire. The GBU-8 and the GBU-2 had significant limitations but were nevertheless highly successful compared to dumb bombs.

The guidance packages in these weapons were trivially simple by contemporary standards, reflecting the low density of period electronics. The cheaper and simpler laser guided weapons rapidly displaced the more complex television guided bombs, despite the higher accuracy of the latter.

The standard low cost GBU-10/12/16 series Paveway II laser guided bomb kit is a case study in simplicity. The quadrant seeker is fitted under a thick lens, and embedded in an aerodynamically aligned seeker head. Electronics in the guidance package sense the angular error between the bomb's velocity vector and the laser spot, illuminated by an aircraft of ground based laser designator. The angular error is then used to control solenoid valves which vent gas from piston / cylinder actuator assemblies, pressurised by a burning gas cartridge. The canard controls are either fully deflected or neutral in position, providing the simplest possible bang bang or non-proportional guidance.

The relatively dumb guidance technique in such weapons results in aggregate guidance errors of the order of several metres, generally irrelevant for a 2,000 lb bomb lethal radius.

Laser guided weapons have some very important limitations. Perhaps the most important of these is their dependency upon continuous laser illumination of the target aimpoint. If the laser is shut down, or the target is obscured by rain, water vapour (cloud/fog), dust or smoke, the bomb seeker is blind and the weapon is apt to follow a ballistic trajectory like a very ordinary dumb bomb.

This limitation was less important in the latter portion of the Cold War since low altitude delivery was considered an acceptable risk in a central European battle with the Soviets. Therefore fighters and bombers delivering these weapons would typically attack from short distances, well below cloud cover in most situations.

With the end of the Cold War tactics shifted. Loss of aircraft and aircrew became politically unacceptable, and bombing campaigns were mostly prosecuted from medium altitudes, well above the reach of AAA and shoulder fired SAMs. The latter accounted for the largest number of coalition aircraft losses in the 1991 Desert Storm campaign.

Medium altitude delivery presented serious issues for laser guided bombs. Loss of the sightline to the target would cause the weapon to go ballistic and frequently impact hundreds of metres from the intended aimpoint. In urban areas this would result in serious collateral damage, and politically damaging loss of civilian lives.

Another issue was the robustness of a simple non-redundant laser guidance system. Whether the guidance signal was lost through hardware failure, or loss of illumination, the weapon was almost guaranteed to go astray.

Adverse weather conditions and embarrassing collateral damage incidents in Desert Storm created the impetus for a production all weather inertial/GPS guided bomb kit.

Inertially guided bomb technology was the subject of intense US Air Force interest during the 1980s. Such a weapon would be initialised over a digital umbilical with target and aircraft coordinates before release, and then it would autonomously fly to impact using flightpath position and velocity information produced by its onboard Inertial Measurement Unit. Microprocessor and Kalman filter technology permitted these weapons to use very refined guidance and autopilot algorithms. The weapon's trajectory could be optimised for range, impact velocity or impact angle. Since the inertial system was self contained, the weapon could not be jammed.

An inertially guided bomb presented the prospect of a robust, digitally programmable, highly reproducable weapon which was jam proof and wholly oblivious to ambient weather conditions. The perfect precision guided bomb?

No inertially guided bomb ever entered production, since the cost of inertial units with the required accuracy proved to be prohibitive. The perfect yet unaffordable guided bomb.

The great enabler for inertially guided bombs was the US Air Force Navstar GPS satellite navigation system. By using a GPS receiver to bound the cumulative error produced by the inertial unit, an inertial bomb with GPS could achieve equal or better accuracy at very low cost, compared to a purely inertially guided bomb.

The first GPS aided inertially guided bomb to be built and deployed was the US Air Force's Northrop GBU-36/B GAM84 (GPS Aided Munition) 2,000 lb weapon, deployed on the B-2A as a gapfiller prior to production of the then embryonic JDAM. While the GAM was a relatively expensive weapon at cca USD 40k / round, engineered for early deployment rather than minimal mass production cost, it did prove the concept convincingly. More over, it also proved an important refinement for improving the accuracy of such weapons. This refinement was the use of platform referenced differential GPS, or GATS (GPS Aided Targeting System). When the B-2 programmed its GAMs before release, it included a list of which GPS satellites it was tracking. The bomb would track only these satellites, ignoring all others, and thus would see identical GPS position errors to the bomber. A GPS aided bomb without differential techniques would have a circular error probable of the order of 7 to 12 metres, using differential techniques the B-2A/GAM combo repeatedly demonstrated 6 metres or less, making it directly competitive against the established precision GBU-10 Paveway II.

The Joint Direct Attack Munition

The JDAM was the result of a hotly contested flyoff between McDonnell-Douglas (Boeing) and Martin-Marietta (Lockheed-Martin), bidding the GBU-31/32 and GBU-29/30 respectively. Boeing won what is likely to prove in time to be one of the most lucrative contracts for decades.

The baseline JDAM was to be an accurate rather than precision weapon, with a planned CEP without enhancements of 12 to 13 metres, corresponding to the systemic GPS P-code error and some guidance loop error. The initial plan was to enhance this basic weapon with future seeker technology to provide genuine precision capability.

The heart of the JDAM is a Honeywell HG1700 Ring Laser Gyro (RLG) inertial unit, which measures position, velocities and accelerations in all three axes. The brain of the JDAM is in its Guidance and Control Unit (GCU), which contains an embedded microprocessor running a Kalman filter, which accepts position measurements from the GCU's HG1700 and a Rockwell GEM-III low cost military GPS receiver. The Kalman filter continuously computes a best estimate of the bomb's position in space. This information, and the preprogrammed target GPS coordinates, are then used to feed a flight control algorithm. HR Textron actuators are used to drive three of the four tail surfaces. Power is provided by a thermal battery in the JDAM tailkit. Most JDAM variants employ a set of strap on aerodynamic strakes, intended to increase body lift and also reduce the weapon's stability to improve its pitch and yaw rates, and thus manoeuvrability.

The flight control algorithm can be configured before launch for vertical or horizontal (ie shallow dive) terminal trajectories, selected by the user for a specific type of target. A weapon intended for the basement of a tall building could be programmed to enter at ground floor level, wheres a weapon intended to enter a bunker shaft could be programmed for a vertical trajectory.

The use of Kalman filter technology allows for refined midcourse flight algorithms, which can manage the weapon's kinetic energy and maximise glide range. Compared to the primitive analogue guidance in a baseline Paveway II, the JDAM achieves close to twice the glide range under similar launch conditions.

The JDAM employs the US standard Mil-Std-1760 umbilical interface, incorporating the Mil-Std-1553B digital multiplex bus. Before launch the JDAM's embedded software communicates with the launch aircraft's stores management processor, no differently than a computer peripheral. Prior to release the JDAM is powered up using an umbilical feed from the launch aircraft. The JDAM executes an internal self test, warms up and aligns the HG1700 inertial unit. Once the JDAM is ready, it communicates status information to the launch aircraft, which then downloads GPS timing, GPS Almanac (ie nav message), GPS Ephemeris (constellation) and the GPS PPS crypto key. This information is used to initialise the GEM-III receiver.

Once the inertial unit is aligned and the GPS receiver initialised, the launch aircraft can download into the bomb the target GPS coordinates, fuse settings and impact parameters, all of which can be reloaded at any time before release. Prior to release the aircraft's position and velocities are downloaded.

After the weapon is released, the thermal battery is initiated, the GPS receiver acquires a satellite constellation, and the weapon autonomously flies itself to impact, using pre-programmed parameters, penetrating cloud with no loss in accuracy. Should the GPS signal be impaired, lost or jammed, the weapon can rely on its inertial unit and will suffer some modest loss in accuracy, dependent upon how late in the flight the signal was lost, and also depending on the tolerance errors in the HG1700 (some units may be slightly more accurate than others).

The autonomous capability in the JDAM is without precedent and a key advantage of this weapon against laser guided bombs. The latter are dependent upon laser illumination, as a result of which the aircraft can engage only one target at a time. While a good operator can pickle off bombs several seconds apart for a level medium altitude strike, and move the laser spot from aimpoint to aimpoint during an attack, in practical terms this permits strikes only on clusters of targets and depends critically on operator proficiency. The JDAM has no such limitation.

The JDAM can fly a boresight trajectory similar to a ballistic drop, but can also fly off axis trajectories, to engage targets to either side of the flight path, with some loss in range. Therefore, an aircraft can pickle off multiple JDAMs almost simultaneously, each independently targeted, with the sole limitation that the targets must be within the kinematic footprint of the weapon. The weapon can be released from altitudes as high as 50 kft, at speeds up to Mach 1.3, with medium altitude drops yielding standoff ranges of several nautical miles. A supersonic high altitude drop (F/A-22A) almost doubles range performance due to the much higher initial energy of the bomb.

A heavy bomber carrying dozens of JDAMs can obliterate dozens of targets within a given footprint, in a single large drop, as each bomb can be independently preprogrammed before release. The catchcry for the laser guided bomb was one aircraft, one bomb, one target - in the JDAM era this becomes one aircraft, many JDAMs, many targets.

Integration of the JDAM is relatively simple, the principal prerequisite being that the launch aircraft is equipped with a Mil-Std-1760 digital weapon station interface. With this capability, software changes are the only modification to the launch vehicle. Clearance testing is required since the JDAM is aerodynamically different to the Mk-84/83/82 series slick bombs.

The JDAM GCU module was sized from the outset to fit the internal volume of a Mk.84, Mk.83, Mk.82, BLU-109/B and BLU-110/B tailcone. At this time production of the JDAM encompasses the GBU-31 (Mk.84/BLU-109), GBU-32 (Mk.83), GBU-35 (BLU-110) models, with the GBU-38 (Mk.82) in development with a planned 2004 IOC.

The GBU-31 has been most widely used, primarily as a replacement for the GBU-10 in strategic strike (Serbia/Afghanistan), battlefield interdiction and close air support roles (Afghanistan). The US Navy has used the GBU-32 and GBU-35 widely during the Afghan campaign. It is expected that the GBU-38 will become a preferred weapon for battlefield interdiction, close air support and especially urban combat - in these roles low collateral damage is more important than lethal blast effect. Directly interchangable with the Mk.82 slick, the GBU-38 will provide aircraft like the B-52H, B-1B, B-2A, F-111C and F-15E with formidable firepower.

To date the JDAM has been used only in its basic configuration, without additional seekers installed. Even with this limitation, the weapon has proven to be a robust replacement for the Paveway II.

The capability of the JDAM to punch through a solid cloudbase has revolutionised close air support and battlefield work, since historically such combat required either very low level strikes using dumb bombs, or medium to low altitude strikes using laser guided bombs. Inclement weather offered cover to a clever opponent. The JDAM has closed this strategic loophole forever.

JDAM Accuracy and Jam Resistance

The accuracy of the JDAM is frequently criticised, the bomb being often described as much less accurate than the widely used GBU-10/12 Paveway II weapons. This argument is lame and not representative of more recent developments in technique and technology.

The baseline accuracy of the weapon cited in mid 1990s glossy brochures is a very pessimistic number, based on worst case GPS accuracy for the period. Since the 1999 Allied Force campaign, the US Air Force has generated predictions of GPS accuracy variations over a 24 hour cycle for targets of interest, or areas of interest. These computer models analyse an effect termed Geometrical Dilution Of Precision (GDOP), which arises as a result of the relative positions of satellites in the constellation a reciever can see at a given point in time and space. As the orbital positions of the satellites in time, the GDOP error increases or decreases. Where and when an unusually favourable constellation is seen, the GDOP error can be very low, and GPS errors resulting can be a fraction of the textbook figure. The practice followed by the US Air Force since 1999 is to plan non-time critical strikes to fall into time periods of minimal GDOP for the target of interest, to achieve defacto precision accuracy.

The US Air Force planned in the late 1990s a series of Product Improvement Program (PIP) incremental block upgrades to the JDAM guidance package, but no details have been disclosed more recently as to which have been implemented to date.

One candidate is the use of platform referenced differential GPS, which is relatively undemanding to implement since it involves only software changes to the aircraft and bomb embedded code (OFP), and GPS receiver operating code. These force the bomb to acquire only a programmed constellation of satellites. The principal errors in bomb delivery are then dominated by the accuracy of the synthetic aperture radar or thermal imager/laser rangefinder used to produce target coordinates, and the guidance loop error in the bomb. Experience with the B-2A suggest this technique results in 6 metre or better CEPs, with the GDOP error dominating the GPS error under most circumstances.

Another more potent candidate is the use of Wide Area Differential GPS (WADGPS) techniques, pioneered in the US Air Force EDGE and WAGE trials. This family of techniques involves the deployment of a network of precisely calibrated GPS receiver ground stations surrounding the theatre of operations, which continuously measure the error in the recieved GPS signal against the precisely surveyed location. Data from these ground stations is fed over low data rate landlines or satellite links to a central ground station, which runs a complex computer model incorporating parameters such as solid earth tide (bulge) and wet / dry tropospheric delay. The system continously computes a set of correction parameters for use in an enhanced Kalman filter, these are encrypted and broadcast via a radio link (EDGE) or unused encrypted GPS Almanac page (WAGE). The compensated GPS errors achieved using this technique are as low as several inches in all three axes.

An aircraft and JDAM configured to use WADGPS techniques can achieve true precision accuracy, 100% of the time, without the cost penalty of a seeker package.

Experience from Afghanistan suggests that the most frequent cause of JDAMs going astray were either bent fins resulting from mishandling, or more frequently the fat finger factor to use the colourful americanism. Human errors in entering aimpoint coordinates on keypads, entry of other than the intended coordinates, and in one instance possibly a ground forward air controller mistakenly transmitting over the radio his own coordinates rather than those of the enemy!

Like the alleged inaccuracy of the JDAM, its vulnerability to jamming is very frequently overstated by its critics. To date there is no published evidence of successful use of jamming to defeat a JDAM, or indeed any GPS aided weapon.

The baseline GEM-III receiver has built in provisions to resist GPS jamming. Regardless of these, successful jamming of a GPS guided bomb is not as simple as JDAM critics like to suggest. For a jamming effort to work properly, the jamming signal must be coupled into the mainlobe of the bomb's antenna, preferably from the very instant the bomb is released, or even earlier. This is easier said than done, since the GPS antenna on the JDAM is mounted on the tail, and therefore if the jammer is colocated with the target, the antenna mainlobe is always pointing away from the jammer. The only jamming signal which can couple in is what little attaches to the skin of the bomb and tailkit as a creeping wave. Creeping waves tend to be weak in magnitude, and are easily suppressed with coatings.

Even should GPS jamming increase in popularity (US reports suggest more recent AGM-88 HARM versions will have provisions for homing on GPS jammers), the installation of improved GPS antennas and receivers would defeat most techniques. Neither represent unusual integration challenges for a modular design such as the JDAM.

One issue JDAM critics seem to universally overlook is the reality that it takes very little effort in any inertial/GPS system to incorporate code which monitors the difference between the GPS and inertially predicted bomb positions. Should the GPS position read from the receiver suddenly change by a large amount, the software can simply reject the GPS measurement and continue to fly the bomb using inertial data until impact, or until the GPS signal behaves as it ought to. Unless the jammer is unusually effective, odds are that gaps in jamming will occur and the bomb guidance can use these to grab valid GPS measurements. With a flight time of mere minutes or tens of seconds, the cumulative inertial system error seen since the last valid GPS measurement could be very small indeed.

It is worth noting that a JDAM is potentially more robust than an analogue laser guided bomb in the event of a guidance component failure. For instance a hardware failure in a GPS receiver or inertial unit could be handled by rejecting its output and flying to impact on the remaining source of position and velocity data. Boeing have not disclosed whether this technique is used.

In summary, most of the criticisms directed at the JDAM (and very popular in some Canberra circles) are very lame and assume a very clever technological peer competitor opponent. Whatever limitations the JDAM might have, these are generally of less significance than the enormous gains in capability and firepower offered by this weapon. At unit costs under USD 20M, the JDAM is one of the best bang for buck choices in the market today.

Without doubt the most important near term application of the JDAM has been its use as a near precision conventional weapon for US Air Force heavy bombers, previously limited to dumb bombs. The 2,000 lb GBU-31 fitted to the Mk.84 or BLU-109/B warheads was the first to see widescale combat use. The JDAM was blooded in 1999 when the B-2A bombed Belgrade with the weapon. In 2001, the decisive blows to the combined Taliban/Al Qaeda ground forces in Afghanistan were inflicted by B-52H and B-1B bombers delivering GBU-31s against a wide range of battlefield targets (Boeing/USAF).

The US Navy's primary JDAM variants are the GBU-32 and GBU-35, designed for the 1,000 lb Mk.83 and BLU-110/B warheads standard for this service. The Boeing F/A-18C/D/E/F will be the primary near term delivery platform for naval JDAMs. Loadouts are likely to be identical to the existing Mk.83, but using smart Mil-Std-1760 racks with Mil-Std-1553B bussing to the bomb umbilical connectors. The baseline JDAM can be retargeted up to the point where it is released (Boeing).