Scholarly record

TUNING AND EVALUATION WITH EXPERIMENTAL DATA OF A SMART ALGORITHM FOR GPS DATA PREDICTION TO FUSE THEM WITH THE HIGH RATE INS DATA

Teodor Lucian Grigorie

Abstract

The paper exposes the tuning and the evaluation with experimental data of an algorithm used to predict the lost data in the GPS readings due to the acquisition rate significantly different from that of the Inertial Navigation System (INS). The developed algorithm provides a real time prediction of the GPS data at the same rate with the INS data in order to be ready for the processing step, which realizes the fusion of the two systems. To acquire the data, an INS/GPS hardware structure based MEMS sensors has been developed in the Aerospace Engineering laboratories at the University of Craiova, Romania, as a part of a research project. Its detection components are an inertial module and a satellite positioning module, the data provided by them being integrated with a microcontroller. The data from inertial sensors have been acquired with a rate of 100 samples/s, while the GPS receiver provided data once per second. The prediction algorithm implements an intelligent approach based on adaptive neuro-fuzzy inference systems (ANFIS). The neuro-fuzzy algorithm is an extrapolator for the data received from the GPS system, but whose training mechanism, uses in the first phase data from the INS system, which, after initialization, in the short term, does not significantly degrade the performance of the navigation solution under the influence of sensors errors in the inertial detection unit. The structural scheme of the Fuzzy Inference System training process includes six channels: three speed components and three global coordinates. The INS/GPS hardware structure has been boarded on a testing vehicle and experimental data acquired for many testing situations. The evaluation results with the experimental data proved a very good operation of the developed algorithm, it being ready to be incorporated in the INS/GPS integrated structure for the next stage of the research project.

Publication Impact Profile

• Captures
• Mendeley - Readers: 4

PLUMX see details

DOI resolver: doi.org/10.5593/sgem2018/2.2/s09.139

Publication details

Back to publication list