Verifiable Obstacle Detection

TL;DR

This paper presents a safety verification method for a classical LiDAR obstacle detection algorithm in autonomous vehicles, establishing bounds that ensure compliance with safety standards, unlike neural network-based systems.

Contribution

It introduces a formal verification approach for classical obstacle detection, providing bounds on system capabilities and sensor requirements for safety compliance.

Findings

Established strict bounds on obstacle detection capabilities

Determined LiDAR sensor properties for safety standard compliance

Provided empirical validation with real-world data

Abstract

Perception of obstacles remains a critical safety concern for autonomous vehicles. Real-world collisions have shown that the autonomy faults leading to fatal collisions originate from obstacle existence detection. Open source autonomous driving implementations show a perception pipeline with complex interdependent Deep Neural Networks. These networks are not fully verifiable, making them unsuitable for safety-critical tasks. In this work, we present a safety verification of an existing LiDAR based classical obstacle detection algorithm. We establish strict bounds on the capabilities of this obstacle detection algorithm. Given safety standards, such bounds allow for determining LiDAR sensor properties that would reliably satisfy the standards. Such analysis has as yet been unattainable for neural network based perception systems. We provide a rigorous analysis of the obstacle detection system with empirical results based on real-world sensor data.