Validating Terminal Guidance Without Expending the Aircraft

The Phase You Cannot Afford to Test

On a loitering munition or one-way attack platform, terminal guidance is where the entire system either delivers or does not. It is also the shortest phase, the most dynamic, and the only one where the conventional test approach consumes the aircraft. The predictable result is that terminal behaviour gets validated across a handful of expensive live trials, producing a sample size too small to characterise anything.

Almost all of it can be tested without expending anything, provided you separate the guidance problem into its components and test each against ground truth.

Decompose Before You Test

Terminal guidance is at least four distinct functions and they fail differently. Target acquisition: does the seeker detect a valid target at the specified range and aspect. Lock quality: once acquired, does it hold through the closing geometry. Track continuity: does it maintain lock through aspect change, scale change, and partial occlusion. Guidance law execution: does the airframe fly the commanded solution within its manoeuvre limits.

A platform that acquires reliably at 800 m but breaks lock at 200 m as the target fills the frame has a scale-handling problem in track continuity, not an acquisition problem. Tested as one monolithic capability — did it hit or not — that distinction is invisible, and the engineering team receives a pass/fail with no diagnostic content.

Testing Each Piece Against Truth

Acquisition and track are testable in a reusable configuration: fly the seeker on a recoverable airframe, run the terminal profile against representative targets, break off at a safe altitude, and record the seeker's track output alongside RTK ground truth. You get a full track history for every run and you get the aircraft back, which means dozens of runs instead of three.

Guidance law execution is testable in the same sortie by comparing commanded acceleration against achieved acceleration. What you are looking for is command saturation — the point at which the guidance solution demands more than the airframe can deliver. Saturation late in the terminal phase is the classic cause of large miss distances, and it appears clearly in logged data long before it appears as a miss.

Field Example: The Lock That Broke at Two Seconds

Evaluating a platform with an electro-optical terminal seeker, live trials had produced inconsistent results with no pattern the team could identify. We flew 34 recoverable terminal runs against a vehicle target over two days.

Track output showed the failure clearly and repeatably. Lock was solid from acquisition at 900 m down to approximately 180 m — then, at roughly two seconds to impact, the target's angular size exceeded the tracker's template scale limit and it began tracking a high-contrast sub-feature, in this case a wheel arch, rather than target centroid. The aimpoint walked steadily off centre through the final seconds.

In a live trial this reads as a near miss with no explanation. In 34 instrumented runs it is an obvious, fixable scale-handling limit in the tracker. The team corrected it in firmware and we re-flew the same profile to confirm.

What Reusable Testing Buys You

Sample size, primarily. Terminal performance is statistical — a platform that hits in three of three live trials has told you very little about its actual distribution. Thirty instrumented runs across aspect angles, target types, lighting conditions, and closing speeds gives you a characterised envelope and, more usefully, tells you where the edges are.

Reserve live trials for end-to-end confirmation of a system you already understand. They are the wrong instrument for discovery.