Dependability overview for autonomous vehicles and reliability analysis for basic safety messages

Autonomous Vehicles (AV) communications promise to increase roadway safety by providing each vehicle with 360-degree situational awareness of other vehicles in proximity, and by complementing On Board Unit (OBU) and sensors such as radar or camera in detecting imminent crash scenarios. These vehicles provide communication services among one another (V2V) or with Road Side Infrastructure (V2I). Vehicular Ad Hoc Networks (VANETs) employ multichannel to provide a variety of safety and non-safety applications. Safety applications necessitate appropriate and reliable transmissions, while non-safety applications require performance and high speed. Broadcasting of Basic Safety Messages (BSM) is one of the fundamental services in today's autonomous vehicles. With the excitement related to the revolution in technology comes the challenge to make these systems dependable to ensure the safety of the consumer and the public. We focused in this paper, on the reliability of AV as a whole. Our scientific approach was structured with methods that combine qualitative approaches and quantitative methods based on the data published by the California Department of Motor Vehicles concerning the accidents and the disengagement reports of AV. Evaluating these problems leads to a sensitivity analysis related to reliability parameters, which helps further innovation in vehicle automation and automobile engineering. Moreover, an analytical model for the reliability of BSM communication was proposed.