Autopilot

Tesla Patent: Autonomous Driving System Component Fault Prediction

Featured image: Automotive News

Autopilot is still new to most people, so it can raise fear and doubt for some. People sometimes worry that Autopilot might fail and the vehicle would get into an accident. Nevertheless, all Tesla cars have technology that helps recognize potential malfunctions and/or predict them, ultimately preventing potentially dangerous situations.

At the end of December 2018, Tesla filed a PCT application for the "Autonomous driving system component fault prediction" patent. This invention relates to the autonomous driving of a vehicle; more specifically, it has to do with prediction of faults in autonomous driving components within the vehicle. In June 2020, the application reappeared at the Patent Office at the National stage, which means that some changes were made to the patent, which prompted a closer look.

Unmanned vehicles include many sensors and data-collection components. The data captured by these sensors are communicated to a central processor, which then processes the data to assist in making autonomous driving decisions--e.g., braking, accelerating, steering changes, etc. that are implemented by corresponding drive units of the vehicle. Such autonomous driving decisions may be performed while the vehicle is self-driving or when the driver of the vehicle is being assisted.


FIG. 2 is a block diagram illustrating a vehicular autonomous driving system constructed and operating according to various embodiments.

Tesla's autonomous driving system includes a fault prediction unit (225 on FIG. 2) configured to predict a potential future fault condition. An indication of the potential future fault condition can then be used to generate a fault message on the infotainment system, dashboard display, or other vehicle system.

In addition, alternatively, an indication of the potential future fault condition can be transmitted to a diagnostic unit associated with the operator of the vehicle or a service technician via a wireless communication link, a charging port of the vehicle, or other wired or wireless interface 218A-B. In this fashion, the indication of the potential future fault condition can be used to repair or replace an autonomous driving component before a failure occurs that might adversely impact the operation of the vehicle.

 



The fault prediction unit can compare the nominal component characteristic data to the current component characteristic data to predict a potential future fault and react in advance. Thus, Tesla cars and Autopilot systems are reliably protected and are able not only to notice a malfunction, but also to predict one before it actually occurs.

© 2020, Eva Fox. All rights reserved.

_____________________________

We appreciate your readership! Please share your thoughts in the comment section below.

Article edited by @SmokeyShorts, you can follow him on Twitter

About the Author

Eva Fox

Eva Fox

Eva Fox joined Tesmanian in 2019 to cover breaking news as an automotive journalist. The main topics that she covers are clean energy and electric vehicles. As a journalist, Eva is specialized in Tesla and topics related to the work and development of the company.

Follow me on X

Reading next

Tesla Accessories