Scale-free PSO for in-run and infield inertial sensor calibration. (December 2019)

• Record Type: Journal Article
• Authors: Poddar, Shashi; Kumar, Amod
• Is Part Of: Measurement. Volume 147(2019)
• Publication Date: 2019-12
• Subjects: Inertial sensor -- In-run calibration -- Evolutionary optimization -- Particle swarm optimization -- Scale free network; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2019.07.077
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: In run self-calibrating algorithm for inertial sensors. Scale free PSO investigated as evolutionary optimization technique. Simulation data generated for two different test setups is experimented. Performance improvement seen in uncalibrated accelerometer readings. Abstract: Inertial sensor calibration is one of the most important aspects of estimating motion using an inertial navigation system. The traditional calibration technique requires mounting of sensor in different orientations using costly equipment which are not available many a times for calibrating low cost MEMS sensors. The infield calibration scheme helps in obtaining the calibration parameters directly by minimizing a multi-dimensional cost function built around principles that these sensors follow. In this paper, a novel in-run calibration algorithm is proposed that does not require the sensor to be mounted in specific orientations. This scheme updates calibration parameter even without dismounting it from its location. In this work particle swarm optimization technique with scale free network is compared with other PSO variants and recommended to be used for inertial sensor calibration. The proposed in-run calibration scheme is run on simulated and real world dataset to investigate its efficacy on uncalibrated sensor readings.