Integrating hybrid FMEA methodology with path planning decisions in autonomous vehicles

Abstract

This paper demonstrates how to choose the best performing path planning algorithm for an autonomous vehicle to address different traffic conditions with the help of FMEA model. An autonomous vehicle uses path planning algorithms to navigate but no one algorithm is sufficient/address all kinds of traffic issues efficiently. This paper proposes an evaluation method. The quality and performance of each of the path planning algorithms are analyzed by a Hybrid FMEA framework which predicts the best path for a given set of traffic conditions. This decision can then be used by the vehicles master program to select the best path for execution instead of executing multiple algorithms. The Hybrid FMEA framework selects the appropriate algorithm for different road conditions/road curves/directions resulting in optimizing execution time and aids in maintaining real-time software conditions. This paper includes a case study of FMEA framework applied to autonomous driving vehicles to support decision-making in different traffic condition.