Journal
INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH
Volume -, Issue -, Pages -Publisher
SAGE PUBLICATIONS LTD
DOI: 10.1177/02783649231221580
Keywords
Safety assurance; conformal prediction; statistical inference
Categories
Ask authors/readers for more resources
This study presents a framework that combines statistical inference technique with a simulator in order to tune warning systems to achieve a provable false negative rate. The framework is applied to driver warning system and robotic grasping application, showing low false detection rate while maintaining the guaranteed false negative rate.
When deploying machine learning models in high-stakes robotics applications, the ability to detect unsafe situations is crucial. Early warning systems can provide alerts when an unsafe situation is imminent (in the absence of corrective action). To reliably improve safety, these warning systems should have a provable false negative rate; that is, of the situations that are unsafe, fewer than epsilon will occur without an alert. In this work, we present a framework that combines a statistical inference technique known as conformal prediction with a simulator of robot/environment dynamics, in order to tune warning systems to provably achieve an epsilon false negative rate using as few as 1/epsilon data points. We apply our framework to a driver warning system and a robotic grasping application, and empirically demonstrate the guaranteed false negative rate while also observing a low false detection (positive) rate.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available