Speaker
Description
Low-cost ultrasonic distance sensors are widely used in embedded measurement and automation systems; however, their measurement accuracy can vary with target conditions and may contain systematic errors that may not be adequately represented by a single calibration relationship. This study experimentally investigates the distance-measurement behaviour of a low-cost Arduino-based ultrasonic sensor for four target surfaces-brick, glass, plastic and paper-for reference distances of 10, 20, 30, 40 and 50 cm. Multiple repeated measurements were acquired for each surface-distance condition and the sensor response was evaluated using mean distance, standard deviation, bias, mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE). The experimental results demonstrate a highly linear relationship between reference and measured distance for all four surfaces and resulting surface-specific regression models achieved R² values greater than 0.999. Despite the strong linearity observed in the measurements, systematic variations in measurement bias were identified across the tested surfaces. To characterise these differences, surface-specific empirical calibration equations were developed using a linear relationship between the measured and reference distances. The performance of the uncalibrated sensor, a universal linear calibration model and surface-specific calibration models was compared. At the condition-mean level, surface-specific calibration reduced MAE from 0.289 to 0.221 cm and RMSE from 0.375 to 0.268 cm corresponding to reductions of approximately 23.6% and 28.5%, respectively. The results indicate that high linearity of ultrasonic distance response does not necessarily demonstrate surface-independent accuracy and simple, interpretable empirical calibration can provide improved compensation of systematic measurement error. The study provides an experimentally grounded calibration approach for improving the reliability of low-cost ultrasonic ranging systems in surface-dependent measurement applications.
Keywords: HC-SR04; Ultrasonic distance measurement; surface-dependent error; empirical calibration.