1. [Publications](/publications)
2. ML-based Fault Injection for Autonomous Vehicles: A Case for Bayesian Fault Injection
 
 # ML-based Fault Injection for Autonomous Vehicles: A Case for Bayesian Fault Injection

  ![Publication image](/sites/default/files/styles/wide/public/default_images/default.jpeg?itok=qUFsuJCP "Publication image")

 The safety and resilience of fully autonomous vehicles (AVs) are of significant concern, as exemplified by several headline-making accidents. While AV development today involves verification, validation, and testing, end-to-end assessment of AV systems under accidental faults in realistic driving scenarios has been largely unexplored. This paper presents DriveFI, a machine learning-based fault injection engine, which can mine situations and faults that maximally impact AV safety, as demonstrated on two industry-grade AV technology stacks (from NVIDIA and Baidu). For example, DriveFI found 561 safety-critical faults in less than 4 hours. In comparison, random injection experiments executed over several weeks could not find any safety-critical faults



 ## Authors



Saurabh Jha (University of Illinois at Urbana-Champaign)

Subho S. Banerjee (University of Illinois at Urbana-Champaign)

Timothy Tsai (NVIDIA)

[Siva Hari](/person/siva-hari)

[Michael B. Sullivan](/person/mike-sullivan)

Zbigniew T. Kalbarczyk (University of Illinois at Urbana-Champaign)

[Steve Keckler](/person/stephen-keckler)

Ravishankar K. Iyer (University of Illinois at Urbana-Champaign)

 

 

 ## Publication Date



Monday, July 1, 2019

 

 ## Published in



[arXiv](https://arxiv.org/abs/1907.01051)

 

 ## Research Area



[Artificial Intelligence and Machine Learning ](/research-area/machine-learning-artificial-intelligence)

[Autonomous Vehicles](/research-area/autonomous-vehicles)

[Resilience and Safety](/research-area/resilience)