Ballistometer performance simulation

Ballistometer performance simulation

Prins, J.J.M.

National Aerospace Laboratory NLR

1993-09-06

With a set of linear accelerometers in a particular configuration, it is possible to measure angular velocity and linear and angular acceleration of the body the sensors are attached to. In this report the performance of a sensor system of this type, the so-called ballistometer, is studied in some detail. In particular, the sensitivity to sensor system imperfections is investigated. This investigation is performed by means of simulation. The report describes the basic theoretical background, including a particular post-processing solution for the ballistometer. The simulation and post processing programs are briefly presented. A short analysis is performed on the sensitivity of the sensor system for typical measurement errors like: quantisation, bias error, scale factor error, sensor location error, and centre of mass location error. Main conclusions: •The processing required to calculate the angular velocity is significant. At low or equal velocities, sign ambiguities occur. Due to these properties, the ballistometer seems less appropriate for real-time low-cost applications (this conclusion is drawn for the given post-processing scheme). For applications with non-real-time sensor data processing this is less of a problem. Once the angular velocity is calculated, the linear and angular accelerations are calculated with relative ease. •The sensor system performance depends strongly on the motion input, therefore, a simple performance indication is not possible. For relatively large angular velocities, the performance is good.

http://resolver.tudelft.nl/uuid:ff49a57f-ab5c-4277-8696-bbdbc1b57aff

Nationaal Lucht- en Ruimtevaartlaboratorium

Campus only

NLR CR 93425 L

Aerospace Engineering Reports

report

(c) 1993 National Aerospace Laboratory NLR