This paper is proposing a sensor fusion architecture that minimizes the influence of a sensor fault. Experimental results are presented simulating faults by introducing displacements in the sensor information from the KITTI dataset. | Journal of Automation and Control Engineering Vol. 4, No. 6, December 2016 Multi-sensor Fusion Module in a Fault Tolerant Perception System for Autonomous Vehicles Miguel Realpe1,2, Boris X. Vintimilla2 , and Ljubo Vlacic 1 1 Intelligent Control Systems Laboratory, Griffith University. Brisbane, Australia Escuela Superior Politécnica del Litoral, ESPOL, CIDIS - FIEC, Campus Gustavo Galindo Km Ví a Perimetral, Guayaquil, Ecuador Email: {mrealpe, bvintim}@, 2 software imperfection. When activated during system operation, a fault leads to an error. A failure occurs if an error is not detected, resulting in the vehicle behaviour that is inconsistent with its specification. Sensor faults can be divided into two general types, hard and soft faults. Hard faults are presented in a stepwise form when sensor data changes abruptly from its normal state to a faulty one, while soft faults are slow degradations in the sensor data through time. Soft faults are more difficult to detect and eliminate because it takes time before the sensor data leaves a limit of confidence. Meanwhile, the faulty information (unless it is detected) may be still used by the system and, thus, may be added to the set of correct data. Abstract—Driverless vehicles are currently being tested on public roads in order to examine their ability to perform in a safe and reliable way in real world situations. However, the long-term reliable operation of a vehicle’s diverse sensors and the effects of potential sensor faults in the vehicle system have not been tested yet. This paper is proposing a sensor fusion architecture that minimizes the influence of a sensor fault. Experimental results are presented simulating faults by introducing displacements in the sensor information from the KITTI dataset. Index Terms—fault tolerance, data fusion, multi-sensor fusion, autonomous vehicles, perception system I. INTRODUCTION The deployment of driverless vehicles has
