Middleware-Based Kalman Filter Design For a Driver’ s Aid System

In this paper we present an implementation of the Kalman filter based on a middleware. This middleware LabMap is applied as a data acquisition and distribution layer in industrial automation area, serving as a software bus. We use its virtual handles to implement the algorithm of the Kalman filter directly at the level of the software bus. The advantage of this approach is that neither of the system components needs to be changes as the implementation is enclosed in the middleware. We show that the Kalman filter is useful to remove the speed noise in a driver's aid system. Note to Practitioners- This paper was motivated by the problem of the speed noise in the driver's aid system ErgoDrive Professional for the visual instruction of test drivers on chassis dynamometers. The speed data from the hardware has noise. This leads to trembling of the speed cursor in the driver's aid, which makes it is difficult for the driver to drive the test profile. In this work we applied the Kalman filter to remove the speed noise. Our experiments showed that we always were able to find suitable parameters for the Kalman filter for sampling rates tip to 100ms. For larger sampling time intervals performance degrades. In future research we will try to solve this problem with a predictive filter.