Enhanced sensor coverage for autonomous vehicles with a preliminary integration of fault detection scheme

Abstract

It is becoming evident that, tomorrow's vehicles will be more electrified, automated, and connected respect to today's vehicles, which will make the future safer, cleaner, and more enjoyable. By now, several prototypes of self-driving cars have been built to demonstrate the concept of automated system and the usefulness of the system in situations like improving the traffic efficiency. However, tremendous issues still have to be addressed prior to achieve a mass production of high-level autonomous driving. One of those issues is the perception, which still dramatically expensive and hard to implement. More-over there are always the possibility that diverse components particularly sensing elements may fail. The paper presents the current state of the art and key progress made both in academia and industrial sector. A deep study carried out proposing and demonstrating the capability of low-cost sensor in covering needed area for collecting information for the safe control of the vehicle is elaborated. In addition, the paper proposed a fault detection approach based on the thresholding, whereas soft faults are projected as direct consequence. The model presented simplified the fault identification task by tracking whether the threshold is reached or not in critical area such as sensing components.

Keywords

• Autonomous vehicle
• Perception
• Sensor
• Control
• Fault detection
• Dynamics