Explainable Artificial Intelligence (XAI) for Point Cloud Perception in Autonomous Driving: A Survey

Abstract

This survey provides the first comprehensive review of Explainable Artificial Intelligence (XAI) methods applied to point cloud-based perception in autonomous driving. Although Deep Learning has significantly advanced 3D perception tasks such as object detection and semantic segmentation, the opacity of these models is a challenges for safety-critical systems. To address this, the paper examines existing XAI for point cloud approaches and categorizes them based on a previously defined XAI-specific taxonomy. The review highlights the challenges and limitations of the current state of the art, which includes the lack of standardized evaluation metrics and the limited number of XAI methods tailored specifically to point cloud data in the field of autonomous driving. In this work, we also outline future research directions, emphasizing the need for point cloud–native XAI techniques, stronger validation frameworks, and the integration of emerging vision-language models to enhance explainability and reliability in autonomous driving.

Aitor Iglesias

PhD Student

My research focuses on enhancing the reliability of Autonomous Driving functions using Machine Learning techniques.

Nerea Aranjuelo Ansa

Former PhD student

My research focuses on machine learning and computer vision for multimodal perception systems.

Ignacio Arganda-Carreras

Ikerbasque Research Associate Professor

My research interests include image processing, computer vision, and deep learning for biomedical applications.