PerceMon: Online Monitoring for Perception Systems

TL;DR

PerceMon is a runtime monitoring tool designed for perception systems in autonomous vehicles, capable of verifying complex specifications in real-time using TQTL logic within simulation environments.

Contribution

The paper introduces PerceMon, a novel runtime monitoring framework that supports TQTL logic with spatial operators for perception systems in autonomous vehicles.

Findings

Successfully integrated with CARLA and ROS platforms.

Monitored object detection and tracking algorithms in real-time.

Enhanced safety verification for perception modules.

Abstract

Perception algorithms in autonomous vehicles are vital for the vehicle to understand the semantics of its surroundings, including detection and tracking of objects in the environment. The outputs of these algorithms are in turn used for decision-making in safety-critical scenarios like collision avoidance, and automated emergency braking. Thus, it is crucial to monitor such perception systems at runtime. However, due to the high-level, complex representations of the outputs of perception systems, it is a challenge to test and verify these systems, especially at runtime. In this paper, we present a runtime monitoring tool, PerceMon that can monitor arbitrary specifications in Timed Quality Temporal Logic (TQTL) and its extensions with spatial operators. We integrate the tool with the CARLA autonomous vehicle simulation environment and the ROS middleware platform while monitoring properties on state-of-the-art object detection and tracking algorithms.