A pop-up SAM that was never on the laydown, one aircraft, one RWR, and bearings that arrive one at a time. The Kalman filter fuses them as they land — 1/σ² = 1/σ₁² + 1/σ₂², one measurement after another — but a bearing measures direction, never distance. A stationary emitter off the route is observable from a straight leg; a moving one is not, until the own-ship turns. Bearings alone can't tell you range — you have to move to see, and the turn that sees best spends the most exposure. The Final Project checkpoint is due today.
From L14's batch least squares to a recursive estimator, the Kalman filter's predict and update, the EKF's linearized bearing, the bearings-only observability conditions, and the price of the maneuver that buys range.
FlashcardsKey terms and relations from the lesson, in a self-test deck.
Pre-flightFive questions to work before class — anyone can be called to the board.
SlidesThe deck we work through in class — an HTML deck you can annotate in the browser, with a print view for saving a PDF.
DemoRun an EKF on one aircraft's RWR bearings and watch the error ellipse collapse — or refuse to — then fly a dogleg and count the seconds it costs inside the threat ring.