This paper presents a calibration method for consumer-grade accelerometers, gyroscopes, and magnetometers. Considering the calibration of consumer-grade sensors, it is essential that no specialized equipment is required to create reference signals. In addition, the less is required from the reference signals, the more suitable the method is on the field. In the proposed method, the novelty in the calibration of the gyroscopes lies in the exploitation of only the known net rotations between the positions in a multi-position calibration. For accelerometers and magnetometers, the innovation is that the direction of reference signals, the gravity and the magnetic field of the Earth, are estimated with calibration parameters. As a consequence, no precise absolute alignment of the sensors is needed in the calibration. The rotations need not be done about a constant axis. In the proposed method, the biases, scale factors, misalignments, and cross-coupling errors for all the sensors as well as hard iron and soft iron effect for magnetometers were modelled. In addition, the drift of the sensors during the calibration was estimated. As a result, all the sensors were calibrated at once to the same frame, exploiting only a cube and a jig and thus, the method is eligible in the field. To estimate the quality of the calibration results, 95 % confidence intervals were calculated for the calibration parameters. Simulations were done to indicate that the calibration method is unbiased.Index Terms-Multi-position calibration, inertial measurement unit (IMU), accelerometer, gyroscope, magnetometer, confidence interval. This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication.The final version of record is available at http://dx.

No abstract

No abstract

This study investigates the fusion of pole force measurement, inertial speed measurement, and video analysis to determine crosscountry skiing performance in field conditions. As a proof of concept, a preliminary study was performed with different grip designs and double poling technique. The test showed that with exploiting inertial measurements, the average speed could be determined for any number of full cycles or separately for each cycle, which may be difficult with other methods in field conditions. The exploited measurements were appropriate for determining the key characteristics of the double poling cycle, which along with the estimated speed data can be used for comparing skiing economy, determining maximum performance, and finding differences in ski equipment.

No abstract