Adaptive target tracking algorithms for airborne ultrasonic rangefinders

The author proposes a method for adapting the sampling rate of digital filtering algorithms used for tracking targets moving relative to airborne ultrasonic pulse echo rangefinders. A method for the velocity dependent correction of estimated range is further introduced to compensate for the tracking errors arising as a consequence of the target sonar relative motion and the low speed of sound in air. Also, a simple gating technique is implemented for facilitating the detection and rejection of erroneous range measurements, caused either by noise spikes or multiple-time-around echoes, otherwise affecting the correct behaviour of the tracking algorithms. The author considers a case study concerning modified simple Kalman two-state filters for tracking a point target along a single physical dimension, namely its radial distance from the rangefinder. Monte Carlo sampling techniques are used for evaluating the performance of the tracking algorithms for different trajectories and model parameters. The proposed adaptive filtering algorithms perform satisfactorily even during manoeuvres; only a modest computational overload is required, compared to the standard implementation.