Stable Performance Under Sensor Failure of Local Positioning Systems

Abstract

Local Positioning Systems are an active topic of research in the field of autonomous navigation. Its application in difficult complex scenarios has meant a solution to provide stability and accuracy for high-demanded applications. In this paper, we propose a methodology to enhance Local Positioning Systems performance in sensor failure contexts. This fact guarantees system availability in adverse conditions. For this purpose, we apply a Genetic Algorithm Optimization in a five-sensor 3D TDOA architecture in order to optimize the sensor deployment in nominal and adverse operating conditions. We look for a trade-off between accuracy and algorithm convergence in the position determination in four (failure conditions) and five sensor distributions. Results show that the optimization with failure consideration outperforms the non-failure optimization in a 47% in accuracy and triples the convergence radius size in failure conditions, with a penalty of only 6% in accuracy during normal performance.